UTILIZATION OF AI IN DIGITAL MARKETING

ABSTRACT

This study explores the role of artificial intelligence (AI) in digital marketing, specifically
focusing on its integration into strategies such as search engine optimization (SEO) and
customer relationship management (CRM). The research aims to understand how businesses,
particularly small to medium-sized enterprises (SMEs), utilize AI tools like text generators
(e.g., ChatGPT), and the impact this has on their digital communication efforts. In addition,
the study addresses ethical and legal concerns, including issues such as GDPR compliance,
which affect the adoption of AI. A qualitative method was employed to gather insights into
the experiences of businesses in using AI to optimize marketing strategies. Findings suggest
that while AI contributes to improved margins, cost reductions, and enhanced quality,
challenges like skill gaps persist, particularly for larger businesses with limited
implementation capabilities. Conversely, SMEs tend to adapt more quickly, utilizing AI tools
for growth and competitive advantage. The study concludes that AI is a valuable tool for
digital marketing, with its effectiveness and profitability depending on the size and resources
of the business. The research highlights the growing importance of AI in shaping modern
business practices, while also emphasizing the need for ethical and regulatory considerations
in its use.

Keywords: Artificial intelligence (AI), Digital marketing, Search engine optimization (SEO),
Customer relationship management (CRM), Gathered data, Ethics, Effectiveness, Regulation
Acronyms

Return of investment (ROI)

Artificial intelligence (AI)

Machine Learning (ML)

Search engine optimization (SEO)

Search Engine Results Pages (SERPs)

Digital Marketing (DM)

Entrepreneurial (EP)

Holding Company (HC)

Chief Executive Officer (CEO)

Chief Marketing Officer (CMO)

Product Innovation Manager (PIM)

Solutions Architect (SA)

 1. Introduction
The purpose of this chapter is to introduce the reader to the topic of this thesis. Additionally,
the background will include information about the main concepts of the research question.
Finally, the purpose and research question will be presented.

1.1 Background
Digitalization has increased in the last few years, making it a part of many people's daily life.
In a similar manner, the digital technology has succeeded in gaining the trust of organizations
and now occupies a significant portion of each person's day, both informally and formally
(Smith et al., 2021). A higher level of digital activity among businesses leads to a higher
level of digital interaction with consumers. Businesses must invest more time and effort to be
more considerate in their approach to market communication. Thus, making sure that the
brand image remains relevant and focused on addressing the unique needs and preferences of
their specific customer segments (Yankelovich & Meer, 2006). Combined with this digital
shift, Customer Relationship Management (CRM) systems have gained importance. The
latter mentioned system helps businesses in overseeing their interactions with current and
potential customers. When merging consumer data and streamlining processes, CRM systems
play a significant role in enhancing business relationships. Hence, leading to customer
loyalty and elevating profitability (Tien et al., 2021). Furthermore, the role of
Search Engine Optimization (SEO) can not be overstated in today's digital era. SEO
involves boosting a website's visibility, and enhances it’s ranking on search engine
results pages. This is accomplished when changes have been made to the website design
with suitable and relevant content, making it more appealing to the search engines result
pages. The final aim of the utilization with SEO, is to improve the quantity and quality
of organic traffic to the site. Furthermore, boosting the exposure of the website to potential
customers and clients (Yalçın & Köse, 2010).

Artificial Intelligence (AI), refers to the ability of a machine to mimic, communicate, reason
and operate independently. Ultimately, simulating human intelligence and similar manner as
humans (Du et al., 2020). AI tools in marketing have witnessed significant growth in recent
years, evolving areas such as customer segmentation, marketing automation, content creation,
and predictive analytics (Chaffey & White, 2022). These tools have made it possible for
businesses to make their marketing efforts more efficient. Furthermore, creating personalized
and targeted campaigns that effectively engage consumers, and elevates sales for the
company when utilizing AI tools (Vlačić et al., 2021).
Efficiency in AI, is generally referred to how well AI systems use provided resources when
assisting the businesses to achieve their goals. Along with the ever-evolving of the digital
and social media field, various AI-tools have also been launched and developed due to the
interest from businesses, the general population and the rapidly changing market (Loureiro et
al., 2021). A generous amount of businesses has expressed a high interest, and is currently in
a state of being able to implement AI into their marketing efforts and strategies. Therefore,
AI could be a suitable tool in market communication, with the addition of being suited for
any type of business that is involved with digital marketing. In that sense, AI could be a
useful tool to implement in order to support businesses marketing efforts. Additionally,
making the process of achieving financial goals more accessible, due to the capacity of work
that could be done in the same period of time. This way, data driven decisions could be the
solution to making repetitive tasks managed by AI from the data that it has gathered
(Makridakis, 2017). By providing faster and more accurate predictions while automating
repetitive tasks and identifying new investment opportunities, businesses could see AI as a
part of a Return of Investment (ROI) (Brynjolfsson & McAfee, 2017). Therefore, businesses
who are integrating AI into their digital marketing strategies can participate in the digital
evolution, while effectively addressing the needs and expectations of customers and
stakeholders within their sector.

Despite the advantages AI can supply, the eventual difficulties also have to be taken into
consideration. Automation-spurred, job loss is a major worry associated with development
and deployment of AI (Bessen, 2019). AI in forms of eg. robots/machines are becoming more
capable of performing tasks that traditionally have been operated by humans. Hence, as AI
becomes more advanced, it will have the capability to replace human labor in different
industries. Henceforth, leading to widespread unemployment. The threat of job loss due to
automation has always been a concern for societies (Huang and Rust, 2018). In the 19 th
century, the industrial revolution was a great example of automation. Machines created more
efficiency and productivity, leading to decreased human labor in many industries. Studies
show that as many as 47% of jobs in the US would be automated in the upcoming decades
(Frey & Osborne, 2013). The reason behind it is the speed at which AI is growing and
advancing. Industries that involve repetitive and routine tasks, such as manufacturing,
transportation or tasks that involve the digital world, like finance and marketing. Machines
can outperform humans in these industries by completing the task faster, and hopefully with
greater accuracy on account of the gathered data (Arntz et al., 2016).

Privacy violation is another risk that is connected to the expansion of AI (Bessen, 2019). As a
result of the quantity of data that can be processed, gathered and analyzed by AI from a
variety of sources, such as social media platforms and search engines (Steimers & Schneider,
2022). Concern arises when AI collects personal or confidential information without an
individual's knowledge or consent, and if it is used to monitor their behavior and day to day
life. Furthermore, this information can be used to target individuals for advertising and
marketing purposes, influencing the individual’s decision making.

Another danger of AI, is that it could exacerbate existing socioeconomic inequalities

(Makridakis, 2017). Hence, AI involves a substantial investment that starter-ups and small
businesses may not have the capacity to invest. For instance, a business that invests in AI to
automate repetitive operations, will result in the loss of low skilled professions. This will lead
to inequality by expanding the gap between highly skilled and low skilled employees.

Furthermore, another danger that AI could contribute is a quick and unexpected change in
market prices can occur when a large number of trades are performed in a short period of
time (Makridakis, 2017). Furthermore, AI has the possibility to optimize the financial
markets, by providing faster and more accurate predictions, automating trading processes and
identifying new investment opportunities.

Thus, due to the fast growth of this phenomenon, the ethical and legal in this segment have
not been properly established. In conjunction with the pressing issues surrounding ethics and
legal concerns, the General Data Protection Regulation (GDPR) has been in effect since 2016
with the aim of safeguarding individual rights for a more ethically sound community
(Hoofnagle et al., 2019). This is something that has to be taken into consideration when using
AI to generate materials for you, since the gathered data could be taken from a source that is
not relevant or true. Furthermore, the ownership of the data could belong to a private or
co-user source.
1.2 Research problem:
Digital marketing requires a deep comprehension of market demands and consumer
preferences, necessitating immediate and impactful responses based on this knowledge. In
the constantly changing landscape of today's market, this can not be enough for the
businesses. The implementation of AI in marketing strategies has become an essential part
for success (Camilleri, 2017). Businesses abstaining from integrating AI into their marketing
strategies risk being left behind. This is because they fail to utilize AI- powered tools to
support data-driven decision-making processes, a crucial requirement in the contemporary
market. The reliance on AI for smart, efficient, and strategic decision-making in marketing is
no longer a luxury, but an essential, in order to keep your competitiveness as a business and
be responsive to dynamic market trends (Camilleri, 2017). Therefore, businesses that ignore
applying AI in their marketing businesses risk being left behind, struggling to navigate
through the complications of the new-age market.

A significant portion of the research study is motivated by interest by the authors, in addition
to the necessity to closely examine to what extent AI-tools has been implemented into
businesses. Additionally, what kind of precautions regarding: Digital marketing, ethical and
financial were taken into consideration prior to the implementation. Studies from secondary
data gathering shows the importance of implementing AI tools into digital marketing efforts,
how it can be used in businesses, moreover the risks and the precautions that emerge when
utilizing AI in business strategies. Hence, enhancing the raised importance of showing the
financial, ethical, legal aspects and precautions following the utilization. One major ethical
and legal concern with AI is data privacy. With the vast amounts of data being collected and
processed by AI, ensuring the protection of sensitive information and the privacy of
individuals becomes paramount (Mittelstadt et al., 2016). The utilization of AI in the market
has the potential to bring about notable advantages such as heightened efficiency and
productivity. Nonetheless, it also poses significant risks, including the replacement of jobs
and the widening of inequalities. Furthermore, this research creates a proposal on the
involvement of AI in the digital market. Additionally, enhancing the benefit of businesses to
adapt this new technology in order to potentially raise their revenue. Furthermore, in the near
future, businesses will use it as their main tool to optimize not only their profit, but also the
customer traffic and their brand image across the market. The study will be conducted on a
Nordic level, more specifically geographically directed to the Swedish market. Thus, to
understand and opt the effect of AI in the region, and the outcome of AI in a digital
marketing context.

1.3 Purpose and Research question:

The purpose of this study is to examine how organizations use AI technology in their digital
marketing, and how this contributes towards the effectiveness of the digital marketing
process. This, in conjunction with any repercussions regarding ethical and legal concerns that
may occur through the utilization of AI. Thus, the following will be the research question for
this paper:

- How do businesses utilize AI and how does it impact digital marketing?

 CHAPTER II
Literature review

1.4 Artificial intelligence

AI is a computerized system that gathers data from multiple sources to perform tasks more
efficiently and effectively than humans are capable of doing, in order to reach the full
potential of success (Van Esch & Black, 2021). AI advances are needed to handle the
substantial volume, velocity and variety of data that is out there which is also duplicating
every 1 to 2 years (De Mauro et al., 2022). The ultimate objective is to improve business
within different fields: marketing, strategy and finance According to Allam Hamdan et al.,
(2021) it has become necessary to introduce new products and new production methods, in
order to meet the consumer needs, wishes, and the productivity of the business. Hence,
making customer satisfaction one of the main goals in a business for the purpose of gaining
loyalty. This can only be carried through, when focusing on technical development and AI, in
order to meet the market needs (Alareeni & Hamdan, 2022). There are several technological
incorporations of AI, but the following are the most important. According to (Du et al.,
2020) this intelligence is built using complex algorithms and mathematical functions:

Machine learning - Provides a machine with the capability to learn from data and
experience through algorithms.
Deep learning - A machine that learns techniques that teaches techniques which
come naturally to humans.
Neural network - This is used to learn things, this is a replica of structure and replica
of human brain, we will be able to get cognitive capabilities in machines.
Natural language processing - Writing and reading text in a language is the field of
NLP.

In today's terminology we distinguish different stages of AI that has evolved, ranging from
narrow or weak AI, to artificial general intelligence (AGI) and lastly, strong AI.

Weak or narrow AI: is a system designed and trained for a particular task and is not capable
of dealing with other challenges unless it is retrained (Wirth, 2018). Weak AI lacks the
flexibility of human intelligence, but they can be very powerful in their domain. They can do
tasks such as voice activated assistants. (Yau et al., 2021) stated that it can answer your
questions or obey a programmed command but cannot work without human involvement.
One good example is Deepmind (ALphaGO) or Siri (Apple). There are numerous industries
that AI are being addressed in such as health care, business and marketing. (Wirth, 2018).

Strong AI is an AI system with generalized human cognitive abilities, meaning it can solve
tasks and find solutions without human interference (Wirth, 2018). A self-driving car is an
example of strong AI. Building a website that uses a combination of computer vision, image
recognition and deep learning. AGI is capable of learning and gathering data by itself and
upgrading its action with time, with higher accuracy than a human being. With that said, it is
plausible to say that AI can become a big part of digital marketing. With AI’s different
algorithms and mathematical functions abilities, it has become a crucial part of a very
powerful toolbox that is being used today. With these technologies implemented not only in
industries but also in digital marketing (Wirth, 2018). With the changes of customer journeys
from physical stores to online shopping, these tools give businesses more opportunities to be
accessible to customers and make more personal brand connections with the consumers
(Tiautrakul & Jindakul, 2019). Some of the tools that AI has become part of in digital
marketing are:

Online advertising – Online ads generated by a system powered by AI, also known as
“programmatic advertising” (Gołąb-Andrzejak, 2023). Where AI enables the whole process
from A to Z from design, content production and management analysis, AI automates
the
entire process and gives the final results (Qin & Jiang, 2019) with the help of a neural
networking text generator and natural language processor.
AI powered chatbots – AI Chatbots is a more intelligent version of chatbots. That is because
of machine learning and natural language processing, which makes the relationship with
customers more natural and developed over time and also allows better adaptation to the
target group (Kaponis & Maragoudakis, 2022). Improving the relationship between the
customer and the business leading to increasing customer survival, revenue and overall
platform.
Web designing – AI website design tools entail new possibilities for web-design in a more
efficient and effective way (Gołąb-Andrzejak, 2023). This helps serve the customer with the
utmost relevant content depending on what stage of the buyer's journey they are, depending
on the visitor whether they are after information or after buying a product. (Nair & Gupta,
2021)
Content curation - AI powered tools grant searching for popular content, also planning and
spreading it accordingly. (Gołąb-Andrzejak, 2023). It's about talking about promoting and
authorizing and boosting engagement level, by sharing and using relevant content from
external sources. and that would happen with help of machine learning (Nair & Gupta, 2021)

1.5 Digital marketing

Firms adapt to AI within digital marketing to help their customers understand and learn more
about their product, to navigate among other brands and firms efficiently and effectively.
These same tools help the firm to understand the customers’ behaviors and how they perceive
and engage with their product and service offerings (De Bruyn et al., 2020). The use of AI in
marketing activities can be both broad and narrow. From a broad perspective, AI can perform
any task that a human can effortlessly and more effectively. As for the narrow AI, in this
domain can surpass the efficiency and often effectiveness of humans (Van Esch & Black,
2021) in results much of the potential of AI in marketing will be achieved via machine
learning.

In the article written by Desai & Vidyapeeth, (2021), the authors argue that Digital marketing
refers to the promotion of products or to taking advantage of digital technologies,
predominantly through the internet, but also through mobile phones, display advertising,
and
other digital media. Ever since the appearance of digital marketing in the 1990’s and early
2000’s, digital marketing has developed and elevated the standard of marketing, which
businesses have had to adapt to as digital platforms become increasingly embedded in
marketing strategies and daily usage. Consumers turn to digital devices over physical retail
locations, and digital marketing campaigns have grown more measured and efficient (Desai,
V, 2019). There are various types of digital marketing tactics, such as social media
marketing, affiliate marketing, pay per click among others that are mentioned by the industry
of digital marketing. Additionally, “website traffic” and “content performance” could be
utilized by businesses working B2C or B2B (Desai & Vidyapeeth, 2021).

The concept of digital marketing has become an umbrella term, describing the placement of
digital technologies within the entire marketing process for gaining and retaining customers,
building brands and CRM (Kannan and Li, 2017). Necessarily digital marketing is an
adaptive technology-enabled operation which helps businesses to collaborate with customers
and partners to communicate, deliver and sustain value for all stakeholders (Kannan and Li,
2017). With the growth of digital marketing, and the increase in the usage of mobile devices
and internet technology lead to a rapid growth of AI in a variety of industries (Mogaji et al.,
2020).Van Esch & Black (2021), suggest that digitalization has shifted the preferences to the
center of marketing strategy, making the adoption of AI mandatory. One good reason for that
is the fact that AI can collect data from various sources at an enormous rate through chatbots,
email and social media websites (Mogaji et al., 2020). As AI has a big impact on other parts
in digital marketing, it also has optimized content creation. AI content creation tools are
helping businesses to come up with targeted and engaging content for their consumers by
making the process more efficient and more targeting towards what the businesses consumers
prefer (Van Esch & Black 2021).

These AI content-creation tools work by analyzing collected data from the customers prior
purchases and behavior on the business’s website. The gathered information for the data
provides the AI-tools with the capability to create content that connects with the target
audience. By doing that, the businesses can get better results from their marketing strategies,
when optimizing their advertising campaigns to have better effect, and make their consumers
more satisfied (Van Esch & Black 2021).
1.6 Digital marketing tools

The theology that has been presented, shows great interest on the part of customers and
businesses. It has been proven that AI and digital marketing must be constantly in a phase
ready for change, for businesses to adapt quickly depending on their customer needs. This
emphasis that businesses are able to gain trust and long term loyalty, while learning more
about their customers. Due to the speed and scale of AI and its advantages, businesses are
able to analyze vast amounts of interaction data with the customers, allowing them to achieve
sustainable relationships with customers and growth in revenue. In addition, to enhance the
connection, and establish long-lasting relationships with the customers, hence increasing in
revenue over time. This will aid marketers of the businesses to quickly identify trends and
modify their marketing efforts (Kumar et al., 2016). There are alot of tools that AI have
enabled to make it more productive, qualitative and efficient. One of those tools are Search
engine optimization (SEO) which optimizes the company's website. Some areas optimized
are content, structure, and metadata to improve its ranking on search engine results pages
(SERPs) (Lewandowska et al., 2019). When it comes to digital marketing, there are different
tools for different improvements and goals. This means that this thesis will therefore be
limited, as it will maintain focus on strategies suitable for businesses who already have
implemented AI. Search Engine Optimization (SEO) is used either for targeting a new
audience with specific needs, or to connect to a larger audience. Additionally, matching
customers with their needs depending on their reach and network browsing, and to engage
with the potential customers. Therefore, to transfer potential clients to revenue. Data driven
content is another tool that can be utilized by businesses, without them interfering with the
privacy of existing customers (Desai, V., & Vidyapeeth, B, 2021).

Why is it very crucial for businesses to rank higher organically when some keywords are
searched by the user:

● Visibility: A higher ranking on SERPs increases a website's visibility and makes it

more noticeable than other websites and after that have the benefit of attracting
organic traffic (Iskandar and Komara, 2018). The majority of users tend to click on
the top search results (Joachims et al., 2005), making it crucial for businesses to rank
high on SERPs for relevant keywords.
 ● Credibility: Websites that are visible on the first page of search results are often the
websites that are perceived as more trustworthy by the searcher (Pan et al., 2007). By
ranking high on SERPs, businesses can elevate their online reputation and also the
brand image of the company.
● Targeted Traffic: SERPs support businesses in attracting targeted traffic by
connecting searchers on search engine, for example google, with websites that match
their search keywords (Pan et al., 2007). This targeted traffic has a high chance to
result in conversions for the company, as users are actively searching for information,
products, or services related to the website's offerings.

The increasing use of AI will change the way keyword research is done in (SEO). The AI notes
that with the rise of vocal search, SEO professionals will need to adapt their strategies to this
new type of search, which differs from regular queries. Keywords are identified as one of the
most important elements of a suitable SEO strategy, which highlights the most appropriate
and relevant words to identify a business's website. The thesis provides a proposal of a four-
step model for determining relevant keywords for a company, which are important for
making voice search effective. Ultimately, the thesis emphasizes the importance of
anticipating changes in the marketplace, adapting to these changes, and creating products,
services, and offers that are continually searched for by users Dumitriu, D & Popescu, (2020).

AI can handle repetitive tasks much faster and more efficiently than a human being, which makes
it more sustainable and increases the productivity of the company. With the help of AI,
businesses have the opportunity to organize and automate many processes within the
company. Research indicates that AI-driven automation can result in cost savings. For
instance, AI-powered chatbots can handle routine customer service inquiries, reducing the
need for human agents and lowering labor costs (Davenport et al. 2020). This leads to the
businesses saving time and gaining resources to focus on more creative and strategic
initiatives, to keep the company on the path of growth and having advantages against other
competitive businesses (Nguyen & Simkin, 2013). Another tool where AI has a significant
role is CRM. CRM is an often-used system, for the purpose of marketing, in addition to
storing large quantities of customer data. However, handling and analyzing large quantities of
data can be time-consuming and difficult for humans. Hence, this era of evolving technology
with AI among other tools, is a fine example of how human labor can become more effective
and facilitated instead of being interfered with, due to the help of AI. According to
Chatterjee, S. et al (2019), there is a need to implement AI-integrated CRM in order to
optimize data collecting, and to make it more plausible for the businesses to do their market
research, while searching for new customers. Meanwhile, focusing on growing profit and
revenue gives the user the advantages. The combination of the two systems would therefore
be of help in the form of making decisions without human interference.

1.7 Ethical concern:

While AI can help businesses that employ technology, it can also affect the dynamic of the
company and raise ethical concerns among the firms, like in Facebook, Who generated their
own AI system where the AI created its own language and gave hateful comments against
minorities (Van Esch & Black.,2021). Customer information must be securely shared and
data collection needs to follow important ethical concerns to insure customers privacy and
security (Mogaji et al., 2021). The Author also mentioned there was a proposal being made
which includes the businesses who are in use of AI, are also in charge of manually checking
and correcting the algorithms. This is essential to ensure they are proper and correct, in order
to use the data in a proper and ethical way, to avoid biased decisions.

To guarantee these precautions, businesses have utilized practices incompatible with GDPR,
to ensure the immunity of their customer’s data. AI could also go in depth with the job loss.
According to Frey & Osborne (2013), 47% of jobs will be automated which will leave ethical
and moral concerns of AI. So far, there are not enough laws that can cover those ethical
concerns. The European commission has fixated on AI and digital marketing in demystifying
how GDPR applies for this category of technology which the cloud requires to perpetuate
monitoring (Kaponis & Maragoudakis, 2022). Using AI in government entities could speed
up the process and work more effectively which affects the ethics and the security of
organization and businesses (Hamdan et al., 2021)

Based on the authors’ theoretical framework, a conclusion has been made that the market is
rapidly changing from physical consumer to digital consumer. This means that the rapid
change of the market is affecting how businesses are operating. The need for AI is becoming
increasingly demanding, due to expectations being made from the general interest of
businesses and consumers.
2. Method
The following chapter describes the research method in order to clarify the objectives and
goals of the project. The study goes into great detail on data collection, validity, analysis
methods, restrictions, and ethical considerations. Semi-structured interview questions were
used to obtain the data, and later in the thesis, a thematic analysis was conducted.

2.1 Research design

With an inductive approach through an explanatory study, the thesis will discover the
theories, through which are recognized after generalized conclusions based on the
observations (Bryman & Bell, 2017). According to Saunders et al., (2019), the choice of an
inductive approach could develop a richer theoretical perspective than what already exists in
the literature, with the help of the naturalistic and emergent research design to build the
theory. Additionally, the findings will be supplied through the study to fill the gaps in this
widely discussed but understudied subject. The deductive reasoning has been made from
previous literature, proving AI has been a powerful tool providing efficiency. Thus, the study
will investigate how AI tools are able to influence different marketing tools. Depending on
the nature of AI tools and their capabilities to work efficiently and productively reaching
businesses goals and values. Qualitative researchers are more inclined to include more
descriptive details when reporting on the fruits of their data. Thus, not only focusing on the
descriptions alone, additionally, believing the explanations are important (Bryman & Bell,
2017). Due to the given priority data from the participants being unquestionably descriptive,
it makes it possible for this study to end up being utterly informative. Moreover, Bryman and
Bell (2017) furthermore enlighten the importance of why questions are often underestimated.
Additionally, critical qualitative researchers find it interesting to understand the economic
and political interests that underlie organizational measures, in order to be able to change
them more easily (Bryman & Bell, 2017). Hence, strengthening the reasoning of this research
with the addition of the research question being raised. Thus, due to the generous number of
businesses who have heard about how AI might be helpful, but have not successfully utilized
it in their organizational strategies. According to Saunders et al., (2019), the explanatory
study with semi-structured interviews can conduct an understanding of the relationships
between variables. In this case, how AI affects digital marketing processes, and what ethical
and legal considerations are included in the process of the implementation.
Hence, making it accessible for the researchers to reach the extent of enlightening the
importance of implementing AI with an extension to previous literature, while searching for
businesses fitted for this thesis. In this case, businesses who have not yet successfully
implemented AI, but have the desire and the capacity to do so if financial means may be
supported as a result of the findings of the study.

When conducting semi-structured interviews based on an inductive approach, the thesis will
naturally follow the explanatory and emergent way. Hence, allowing the themes of the
interviews to evolve depending on what emerged from the analysis of the collected data
(Saunders., et al, 2019). Three interviewees was chosen in order to add more objectivity and
credibility into the subject of AI in digital marketing. Furthermore, the selection and
categorization, was narrowed down and chosen along with our criteria, resulting into the
following:

● Businesses who have implemented AI to an extent into their strategy.

● Businesses who are in the process of implementing more AI into their strategy.
● AI retailers who provide businesses with AI-technology.

2.2 Research method:

As previously mentioned, interviews through a semi-structured process was chosen to be the
method of collecting the qualitative data. The research aims to understand how businesses
utilize AI in their digital marketing efforts, and the effect of the utilization in their digital
marketing efforts. With the addition of the repercussions that have to be taken into
consideration due to ethical and legal concerns in the process of implementing AI. By using
semi-structured interviews, the study is able to gain a further understanding behind the
reasoning, changes and attitudes towards the implementation of AI into organizational
strategies. By involving Businesses who are in the process of implementing more AI into
their strategy, businesses that have implemented AI to an extent and lastly businesses who
are providers of AI solutions, it is possible to give a full picture of the issue under discussion
when combining and analyzing the data being gathered from the various groups. When
involving these three types of groups of interviewees, the thesis will be provided with fruitful
information and gain insights of the potential challenges due to the utilization. Hence,
making it possible for the authors to see how the businesses are using AI, how it has affected
the digital marketing process and efforts, additionally, the ethical and legal repercussions
that
have been taken into consideration. Furthermore, making the thesis a collected sheet of how
AI is used in businesses and what it entails, resulting in the motivation of implementing AI in
their strategy. This study may become relevant for those who are in the business sector and
have the interest, but are in need of an example of clear evidence showcasing the utilization
of AI in digital marketing, in order to implement a change of this kind into their strategies.

Furthermore, the thesis uses this approach due to the subject's nature, which analyzes the
knowledge value of one practice being absorbed into another. The semi-structured interviews
also provide the interviewees with the ability to think out loud about things that may not have
been thought about prior to the interview (Saunders.,et al,2019). Hence, making it notably
relevant to the study since the subject is rather underexplored and complex, due to its recent
emergence and relevance.

2.3 Data collection

2.3.1 Primary data:

Semi-structured interviews were conducted when collecting the primary data. This was the
choice to be given due to the aim of acquiring information of the utilization, and the resulting
effects and limitations that were taken into account in the implementation. The choice of
interviewees are aligned with the criteria, of which was described previously in the research
design. The interviews were conducted face-to-face, in a location of their choice and
internet-mediated synchronously. According to Saunders.,et al (2019), these methods provide
the interviewees with the comfort of being in a location that is familiar and safe. This is
advantageous due to the comfort making it easier to focus on the questions being directed to
you as a participant, rather than focusing on the surrounding environment. When conducting
an interview through the semi-structured way, it is allowed to guide the conduct of the
interview with a predetermined list of themes with related key questions (Saunders et
al.,2019). The latter mentioned authors also suggest the interviewees' true opinions and
beliefs are being delivered through the semi-structured interviews, which is crucial to
determine the utilization and impact of AI in digital marketing. All of the interviewees were
asked to participate in this study through a request being made through LinkedIn.
 The interviews were scheduled to include a total of ten participants, see the following table
representing the categorized businesses.

Businesses who have implemented AI to an extent into their strategy

Six out of Ten

Businesses who are in the process of implementing more AI into their strategy
Two out of Ten

AI retailers who provide businesses with AI-technology Two out of Ten

(Figure 1 -The three categories of participants vs categorization of the total -
Created by the Authors)

The choice of participants was based on the criteria of the AI retailers providing AI
technology in the matter of digital marketing efforts, whereas the remaining of the
participants were chosen due to the previous familiarity. According to Saunders et al., (2019),
ensuring familiarity with the organization beforehand is crucial to signal that a thorough
thought-process has been made about the research. Due to the familiarity of some of the
organizations, the gathered data had the ability to be fruitful when gaining trust of the
participant. In addition to the familiarity, there was personal contact involved in the data
collecting process. Hence, the personal contact is evident when managers and employees are
more willing to agree to being interviewed rather than completing a questionnaire.
Especially, when the topic of the interview is relevant to their current work (Saunders et al.,
2019). As a result of the selection of interviewees working within digital marketing with a
shown interest in AI on their LinkedIn pages, a high response rate was accomplished. Thus,
making it particularly suitable for both parties to proceed with the interviews (Saunders et al.,
2019). The latter mentioned authors also enlighten the importance of recording data, where it
is beneficial to audio-record the interview while making notes as it progresses. The
advantages of this practice allows the interviewer to concentrate on questioning and listening,
while also giving them the opportunity to re-listen to the discussion during the data analysis.
With the addition of allowing direct quotes to be used in the thesis (Saunders et al., 2019).
The disadvantages of doing this may also include: The possibility of technical issues, focus
on the
recording rather than the interview itself, additionally, the time-consuming process of
transcribing the recording (Saunders et al., 2019).

As a result, employees and company owners from different businesses in the field of
marketing, both local and international were interviewed when collecting primary data. The
participants were introduced to the central questions in advance. Hence, giving the
interviewee the time to process the questions, providing the research with more affluent and
detailed information. By giving the opportunity to the interviewees to prepare in advance, it
allows the interviewee to gain the trust and confidence of the interviewers in the interview
(Saunders et al., 2019). The interviews started with a few simple questions, to identify the
interviewees occupation, leading toward the fruitful questions and discussions. The
interviews were prepared with an initial of 20 questions being divided accordingly,
depending on how knowledgeable the participants are about AI and its utilization.
Furthermore, during the interviews, follow up questions were added, in addition to better
understand the context of the answer provided for the main question. The follow up questions
were non-prepped questions. Hence, making the questions necessary to keep the conversation
flowing with no need to restructure the interviews, in case the interviewee went off topic or
gave a shorter answer than expected. The timing of the interviews differed from one to
another, depending on which stage they were in the process of AI implementation. Each
interview took approximately 30 to 40 minutes. Since the employees were living in different
cities or countries, the interviews were conducted in a one-on-one consultation. The
interviews were conducted over Zoom, ensuring accessibility and personal connection
through audiovisual interaction, at a place and time suitable for the participants (Saunders et
al., 2019).
During the interviews, some procedures were taken such as:

- Recording and documenting the interview became an essential part of the interviews,
to avoid missing any primary data or secondary observations. If the interviewer only
writes notes, critical expressions or citations could be lost (Saunders et al., 2019). It is
complimentary for the team to follow through the interview, and get the valuable
information, thus a manual transcript was made.
- Experiential data: As mentioned, the interviews were organized in two ways, virtually
and physically. The interview started with a light conversation, with formal
introduction and explanation of the purpose of the study, contributing to a chance to
get to know the interviewee on a deeper level. Furthermore, the interviewees
 presented themselves, and their position in the company. Overall, the meetings went
effectively with a great atmosphere. All the interviewees were welcoming and
talkative, and provided us with insightful information. The interviewees also shared
their knowledge towards AI, and their willingness to adapt to AI both in their daily
chores, and to optimize their working environment.

A descriptive table has been included to the primary data section with information about the
selected businesses that were a part of this thesis. Thus, to show additional transparency, with
the addition of a chance of clarifying the descriptive parts of the partaking businesses. See
attached acronyms to further understand the figure.

Infra motion
Shippey Reynolds My
ompany: Quick Net Substorm Honesty Ampilio Eniro labs SAS
AB International news
desk
dustry: DM DM DM DM DM DM DM DM Security Aviation

rm ownership:EP HC Concern Concern Concern Concern Concern HC EP HC

mployees: 45 163 23 38 26 930 16 214

terviewee: CEO CEO CMO PIM CMO CEO CEO CEO CEO SA

(Figure 2. - Descriptive data on selected companies - Created by the authors)

2.3.2 Secondary data:

Secondary data is frequently used in conjunction with primary data when conducting a more
thorough inquiry. In order to make this thesis topical, the authors decided to study recent
literature in the subject field. To maximize the potential of the collected primary data, it is
crucial to combine semi-structured interviews with high-quality secondary data. The authors
may compare our own findings to the knowledge baseline and put them in a broader context
by carefully reviewing and selecting our secondary data (Saunders et al., 2019). Additionally,
it is clear that the idea of AI in digital marketing is of interest to both the general public and
the business world. Internet sources of secondary data for this thesis, will be shown in the
forms of the content of newspapers and websites. Examples of such information, is the body
of literature that exists on prior understanding in the topic area. Databases offered by the
Mälardalen University library were used to compile the existing data. However, it is the duty
of the researchers to evaluate the reliability, accuracy, and applicability of these sources with
relation to the study (Saunders et al., 2019).

2.3.3 Data Analysis:

Braun and Clarke's methodology is centered on thematic analysis, which involves

identifying, examining, and reporting patterns or themes within data. A thematic analysis was
determined to be the suitable method when analyzing the gathered data in this study, due to
its systematic way of analyzing qualitative data (Saunders et al., 2019). According to Bryman
& Bell (2017), recurring themes throughout the text are the design of a thematic starting
point. The process involved identifying and coding significant patterns of meaning across the
data set, which allowed for a comprehensive exploration of the research question and
objectives.

This approach facilitated the identification of similarities, differences, and patterns in

attitudes toward the phenomenon being studied through a rigorous process of coding and
interpretation, resulting in a deeper understanding of the set of data.

Bryman & Bell, (2017), on a further note that coding and patterns in the gathered data could
be found and analyzed based on, metaphors and analogs, similarities and differences in the
answers, or linguistic connections. This way, the study can be analyzed and summarized into
the most relevant themes and patterns, leading to meaningful insights that can be utilized to
answer the research question.

Ultimately, the use of thematic analysis allowed for a comprehensive exploration of the
participants' attitudes and experiences of AI. Additionally, current choices that are being
made in their digital marketing efforts, ultimately resulting in deep insights that answers the
research question.

2.3.4 Ethical considerations:

Research should seek for ethical conceptions and behaviors even though there is no
"formula" for ethical behavior that applies to all situations. According to Saunders et al.,
(2019), it is necessary to state how the data will be stored in terms of how long it will be kept
and who will have access to it after the study. This way, it is crucial to inform the participants
about the data protection policies and obtain their informed consent before collecting any
data. Furthermore, to ensure that the participants are aware of their personal information
will be collected and stored. Additionally, the present research will adhere to the ethical
principles of qualitative research, such as protecting participant rights and privacy, ensuring
informed consent, disclosing any potential conflicts of interest, and considering potential
harm to participants during and after the study. By keeping the participants anonymous and
confidential, the research seeks to uphold their right to privacy.

2.3.5 Reliability & Validity:

Saunders et al., (2019) mentioned that reliability refers to the extent to which a research
study's findings can be replicated by other studies. Validity focuses more on the degree to
which a research study's findings are accurate and meaningful. In this case, when studying
how businesses have implemented AI and its impact on digital marketing, there is no chance
of this study being able to be replicated in exactly the same way due to the fact that the
conduct is being done at a specific point in time. Since this subject is in constant change due
to its recent occurrence, it's nearly impossible to replicate this study in another period of time
further on. Hence, the subject will change rapidly as technology, and the implementation of
AI continues to advance with the addition of the following associated regulations, thus
decreasing the study's replicability.

However, when it comes to the validity, the study can still have high validity as long as the
research methods are designed to accurately capture relevant data. With the addition of
providing meaningful insights into the topic of AI implementation in digital marketing. This
will be ensured through the semi-structured data gatherings since the data will provide the
study with the flexibility to gather information, while still ensuring that all relevant topics are
covered. Moreover, through the use of multiple sources of data such as interviews with
professionals in the industry and analysis of online resources to ensure accuracy and
reliability in the findings. Additionally, since the respondents were to choose their location of
the interview, the answers had a higher chance of being true to their honest thoughts. Giving
the thesis valuable and accurate data for this study, additionally, the data being of a character
fitted for the theme and the timing of the study. Furthermore, to ensure validity in the study
findings regarding the impact of AI on digital marketing implementation, a rigorous analysis
will be conducted using appropriate statistical methods and data interpretation techniques to
ensure that the results are accurate and meaningful. However, the paper's validity will be
shown by proofreading repeatedly, in addition to taking comments from other authors and the
assigned supervisor into consideration.
 CHAPTER III
This study is based on ten different interviews that have been conducted with various types of
interviewees on the basis of three categorization criteria. The themes throughout the thesis
are based on the previous structure, following the same subtitles which supports the views
and utilization of AI in businesses. Similarities and differences among the interviewees with
their utilization and thoughts of the different tools will be revealed, with the addition of the
reasoning and results from it will be covered in this chapter.

2.4 Interviews with businesses

The following chapter will give an outline following the thematic theme. The empirical
collection was carried out through ten interviews. Five of the interviewees were selected from
the group of; Businesses who have implemented AI into their strategy. An additional selection
of three businesses were applicable and chosen due to the criteria of; Businesses who are in
the process of implementing more AI into their strategy. And lastly, the remaining two
interviewees who would give this thesis a further understanding of the subject in question
were: AI retailers who provide businesses with AI-technology. All interviewees actively
expressed their willingness to be a part of the contributions, but for the sake of the thesis, the
interviewees names are to remain anonymous in terms of ethical considerations of the
interviews. The following figure will illustrate the previously mentioned participants in a
visual form, in addition to contributing further clarity and transparency of the thesis.
The table represents all the businesses that have been interviewed, and they have been
categorized depending on their stage of implementing AI.
 1.Shippey AB

2.Reynolds
International AB
Businesses who have implemented AI to an extent into their Five out of Ten
3.QuickNet
strategy
4.Mynewsdesk
 5.SubStorm

Businesses who are in the process of implementing more AI 6. Honesty

Three out of Ten
into their strategy
7.Ampilio

8.Eniro

AI retailers who provide businesses with AI-technology 9. Inframotion labs

Two out of Ten
10. SAS
(Figure 3 - Company of interviewees vs implementation of AI into business strategy -
Created by the Authors)

2.5 Businesses who have implemented AI to an extent into their

strategy

2.5.1 Digital marketing

Shippey (Interview 1)
The interviewee sees AI technology as an effective tool in digital marketing. The company
uses AI for content generation and SEO, which has led to increased efficiency and website
traffic. The interviewee recognizes that converting website visitors into customers requires a
human touch on two main things. First, on the marketing strategy and second on SEO,
targeting the right audience. The interviewee also mentioned an AI-based tool as a beneficial
tool and the urge to be updated to the development of AI and be one step ahead to try the new
AI tools that could help the company strengthen its competitiveness. AI can help automate
tasks like internal linking for blog posts, but the interviewee also acknowledges the
importance of oversight to ensure that AI does not 'take over.'
Reynolds (Interview 2)
The interviewee mentioned the AI-based tool, "Chat-GPT," as a helpful tool regarding digital
marketing. The interviewee described Chat-GPT as a tool that improves efficiency and
enhances business customer satisfaction. Primarily when AI is utilized to increase the
capacity and enhance the business's quality regarding marketing efforts. Additionally, AI can
enhance certain digital marketing aspects. However, they must maintain the importance of
leadership, team management, and other critical business skills. Therefore, the interviewee
advocated that AI tools can be additional assistance with human supervision. This way, AI is
seen as a complementary tool rather than the foundation for marketing campaigns.

QuickNet (Interview 3)
Although the business has a sizable consumer base in the e-commerce industry, its primary
focus was content creation and SEO. They used Chat-GPT to generate text for their content
and target a better audience with the use of SEO. Due to the company's extensive market
expertise and integration of digital marketing into everyday operations and client contacts,
successful remarketing is possible.

Mynewsdesk (Interview 4)
Our interviewee highlighted their perception of AI's transformative potential within digital
marketing. They expressed a number of significant advantages that they believe are critical to
survive and stay on track in this fast-evolving market.
The company believes in the power of AI not just as a futuristic tool that will be more
apparent in the future of businesses but as something that is needed by companies wanting to
maintain competitiveness in this digital age. The interviewee shared their experience of
witnessing noticeable transformations in their operational processes, especially when it came
to the creation of marketing materials.

SubStorm (Interview 5)
The interviewee highlighted the significance of AI in the domain of digital marketing. The
interviewee also noted that while AI adoption in the broader business landscape may not be
typical, there is a growing trend toward it. The interviewee also recognizes that AI plays a
significant role in marketing optimization and automation, highlighting the necessity for a
higher implementation rate, which would contribute to boosted efficiency and improved ROI.
Moreover, the interviewee discussed customer engagement, acknowledging the potential of
AI-powered tools to enhance and help recruit new clients. The interviewee described AI as
giving a competitive edge and described SubStorm as the forefront of AI implementation in
their business field

2.5.2 AI tools
Shippey (Interview 1)
The interviewee had a minor AI implementation, such as text-generative tools like Jasper and
ChatGPT. The interviewee argued that the tool "Jasper" is more convenient, resulting in
text-generating, but also using Chat-GPT for SEO when posting blogs on the company
website. Additionally, the interviewee advocated that there is a preference for using Jasper
due to its earlier appearance on the market, with the addition of it being a tool that is paid for
compared to ChatGPT. On the other note, the attitude toward AI tools is relatively positive,
but the risk of AI tools replacing some of her areas of work was not considered a threat. The
interviewee mentioned
- ”But I think it's important for everyone using AI just for writing, not just not take the text,
copy paste it and be done. You need to sort of add your human touch and maybe wrong
spellings and stuff. or whatever. Well, fact-checked of course. Exactly, because AI can be
correct like 98% so far, but not like 100% so it always has some bugs.”

This way, it has shown the significance of the “human touch” with changing the generated
text to be suitable for the targeted audience. Additionally, the interviewee advocated the
importance of fact-checking when using AI generated text.

Reynolds (Interview 2)
Reynolds talks about the value of ChatGPT as AI that can help with numerous facets of
digital marketing. It is underlined that ChatGPT is viewed as a starting point rather than the
backbone of digital marketing operations. Higher capacity, higher speed, and improved
output production are advantages ascribed to AI-powered solutions like ChatGPT, all of
which boost customer happiness. The interviewee also emphasizes the significance of
non-AI-related abilities in the context of digital marketing, including team management,
leadership, and client relationship management. AI-powered tools are helpful in digital
marketing by helping with chores, increasing productivity, and increasing customer
happiness. However, they are seen as complementary tools rather than the primary basis of
marketing campaigns.
QuickNet (Interview 3)
QuickNet explained the different types of AI tools that are already utilized in their business
strategies, which were of the text generative and ads-generative tools. Jasper, ChatGPT,
Google, and Meta ads were mentioned by the interviewee. Regarding the attitude toward AI
tools, it was shown to be positive by the interviewee. Furthermore, the considerations before
implementation were of the sort of the tool being able to improve the current customer
engagement, resulting in an increase of their customer base with an addition of their newly
recruited customers being E-commerces.
- ”We have more e-commerce customers than we have bigger customers. And that
can partly be due to the fact that the performance has worked well for the
category. It's not something we did ourselves, it's more a development on Google
that they have made a better campaign. It has affected our efficiency because it
made faster for us to analyze, giving results faster.”

Hence, making it the opportunity of having a heightened efficiency thanks to the fast
received results from AI-tools. Resulting in even better results than prior to utilization in the
business’ strategy.

Mynewsdesk (Interview 4)
The interviewee emphasized that they utilize AI tools in various ways. They use these AI
tools for creating content. For instance, the company uses AI to help create marketing
materials, internal documents, and external communications. This implementation helps to
scale up content production without an increase in team size which saves the company
resources they can use elsewhere in the company. The interviewee also mentioned that the
company uses AI to Identify relevant journalists for their customers based on their written
content. This is a significant application of AI, helping the company deliver suiting services
by finding the proper journalist. Although AI is not integrated into their Customer
Relationship Management (CRM) product directly, they use it in combination with it to
provide contact recommendations on when to post.

SubStorm (interview 5)
The interviewee emphasized how AI helps in content creation, client segmentation, and
predicting analytics. The interviewee uses generative AI tools, such as ChatGPT, for
assistance and to overcome challenges like the "blank page syndrome." The interviewee
mentioned other applications of AI, such as chatbots for client engagement and personal
recommendations. Lastly, the interviewee emphasized that AI tools have changed a lot and
made them more efficient, and also gave them the possibility to focus on more qualitative
tasks.

2.5.3 Ethical
Shippey (interview 1)
The interviewee explained the importance of ethics in AI and acknowledged that it is an
essential issue that the industry must consider. One of the most critical ethical concerns the
interviewee discussed was about data collection and the customer's privacy. The interviewee
expressed concern that the methods that the AI collects and uses data can sometimes violate
the privacy of individuals. It was also mentioned that businesses need to be transparent about
how they collect, store, and use data, as well as obtain consent from the individuals whose
data they are collecting. This transparency will be the key to maintaining trust with customers
and avoiding legal issues in the future. The interviewee discussed the potential impact of AI
on jobs and the labor market, emphasizing that while AI has the power to automate some
tasks and potentially increase the productivity of the company, it can also lead to job loss for
people that worked with these types of tasks and mentioning that it is crucial to have new
strategies to manage this transition, such as retraining workers for new roles.

Reynolds (interview 2)
The interviewee expressed worries about the potential risk of customer data being spread or
compromised but spoke further about the ethical risks not applying to its usage. The business
emphasized the importance of the outcome and that the reasoning behind their customer
attraction is connected to their delivery. Meaning, that if what is promised is being delivered
and it shows results, there is no threat from the customer's side since that is why the business
is being hired. The team is also encouraged to use the AI-powered tool ChatGPT as an
assistant to enhance their work by contributing to higher productivity and capacity rather than
the tool being seen as a replacement. The interviewee also acknowledges the risk of the tool
being used in an unethical way, but goes further on by saying that there are no risks involved
if the use of the tool is done in the described manner. Customer engagement can increase
through output production, speed, and quality, leading to customer attraction and retention.
The ethical consideration that is being mentioned is ensuring that AI-powered tools are used
to enhance customer experiences and engagement without crossing any ethical boundaries.

QuickNet (Interview 3)
The company stresses the importance of keeping the upload rate high and ensuring the
information is original. Additionally, they underline the moral conundrum of how much
consumer data should be utilized and say they decide to use the least amount possible. They
emphasize the necessity to strike a balance between tailored marketing, respect for clients'
privacy, and eliminating the feeling of surveillance. The interviewee also discusses the
difficulty of figuring out how much to engage clients through remarketing without upsetting
them. Overall, the interviewee stresses the necessity to strike the proper balance regarding
digital marketing strategies while acknowledging the ethical issues.

Mynewsdesk (Interview 4)
The interviewee mentioned that they recognize the potential for AI to manipulate incorrect
information, which could lead to misinformation. That is why the company is considering
strict fact-checking processes to ensure the trustworthiness and reliability of AI-generated
content. This commitment shows the possible understanding of harm to their reputation and
the public mistrust from spreading misinformation. The interviewee considers the possible
negative side of AI in the job market, specifically job losses due to the automation of many
fields that were managed manually by human beings before. However, the company believes
that AI should be seen and used as a tool to support growth rather than as a tool to minimize
the workforce. This approach aligns with a broader ethical consideration in AI development,
showing the importance of human workforce AI deployment that respects the dignity and
value of human labor.

SubStorm (interview 5)
The interviewee showed an in-depth understanding of maintaining a cautious stance toward
data privacy and ethics. The interviewee mentioned the significance of being mindful of the
data utilized in AI solutions, mainly not to cross any boundaries or compromise client data
gathered by AI technology, which could potentially infringe upon privacy norms or GDPR
guidelines. The interviewee also respects the “Human in the Loop” principle, discussing that
AI should not make decisions independently without human intervention.
2.5.4 Financial
Shippey (interview 1)
The interviewee emphasized that AI tools have significantly increased website visits for her
company, which is the preliminary stage in customer interaction. This is a critical component
as more website visits translate to higher chances of converting potential customers,
subsequently leading to an increase in sales and growing the company's customer base.
The interviewee also emphasized the role of AI in personalization. The interviewee stated
that AI could understand the customer's preferences and behavior patterns through machine
learning and data analysis, leading to satisfied and returning customers.

Reynolds (interview 2)
Interviewee 2 expresses many profound opinions and thoughts regarding financial
considerations. The interviewee mentioned they had yet to invest heavily in AI tools but are
open to exploring paid options if they solve specific business problems. For instance, this
business primarily uses ChatGPT extensively since it is a free tool. Furthermore, the
integration of AI has contributed to the impact on productivity.
- "It would be capacity... The second one would be speed... It used to take a writer two,
three, four, five days to write a good text. Now we can produce it in maybe an hour."

According to the stated quote, this has shown an increase in their capacity to produce even
more and better content with the addition of the speed of content creation Regarding
efficiency and quality, the interviewee has helped the business to improve the quality of the
content for their marketing since ChatGPT has provided the business with assistance in ways
of proofreading and editing. Hence, higher output production and better overall results have
been contributed with the assistance of AI. Lastly, the business mentions that they are open to
investing in paid AI tools, even though their current implementation has been free of cost.
This consideration of paying for additional AI will be relevant if the cost reduction is
reduced. According to the interviewee, the considerations and evaluation of further
implementing AI have been shown to increase productivity and improve results for a change
to unfold.
- "If it solves the problem, then sure... it becomes just a mathematical question... finding
the problems that we have in our current business and looking for solutions to them."
 QuickNet (Interview 3)
When it comes to the financial consideration regarding AI in the thinking of QuickNet, it was
said that:
- “We have more e-commerce customers than we have bigger customers. And that can
partly be due to the fact that the performance has worked well for the category. It's not
something we did ourselves, it's more a development on Google that they have made a better
campaign. It has affected our efficiency because it made faster for us to analyze, giving
results faster.”

This shows that machine learning in forms of Google and Meta ads has a significant place in this
mid size company, which has shown an effect on their efficiency, making it easier to analyze
the data. Enabling the business to be given results faster. Furthermore, QuickNet also has a
rather positive attitude towards the implementation of AI into their digital marketing efforts.

- “Everything we can do, we will save money on. More customers will save time and
save personnel. I think that is a big advantage. And it will get better. Many things.
You can think you are good, but you will never beat a Google data store on the
logic.”

Meaning, that the implementation of AI tools, in their case Google tools, is seen as an investment
and a part of their ROI, rather than unnecessary tools. Additionally, the attitude towards AI
being more advantageous and competitive than when not using AI.

Mynewsdesk (Interview 4)
The interviewee emphasized that it is still in the early stages to show any noticeable increases in
revenue or profit. The interviewee was careful in assessment, saying that while there have
been marks of financial growth, it is too early to correlate these financial improvements to the
usage of AI definitively. However, the company kept maintaining a strong belief in the
crucial role that AI has already started to play in the marketing and sale of its products and
services. The respondent suggested that implementing AI has dramatically improved their
capacity to communicate their products more engagingly and specifically. The interviewee
believes in AI's potential to lead the company to increased customer interest and, later on, the
purchase of their product or service. Even though this may not directly result from AI
applications, it indirectly drives revenue upward. Furthermore, the interview highlighted the
view of AI as an investment with the potential for long-term financial benefits, meaning that
the company believes it will benefit from it further on. The company believes that as they
continue to use the power of AI technology, their understanding of what it can be used for
continues to evolve. They will start to see more data on financial gains. They see the financial
rewards of integrating AI as a process that will make returns over time rather than
immediately.

SubStorm (interview 5)
The interviewee emphasized that while direct revenue increases have yet to be seen since the
implementation of AI, the company has seen an increase in engagement. This could cause an
indirect increase in income and gradual growth of the company. The interviewee also
discussed that their company is a small consultancy that has a focus as it sells services rather
than products which makes the direct correlation between marketing efforts and revenues
more complex. The interviewee further said that while some AI tools are expensive and hard
to implement, the ones they have been using are mostly affordable or even accessible. As a
new company in the market, they must be aware of their resources and ensure that they
benefit from these tools and be worth the cost. The interviewee suggested that the company
could potentially adopt more AI tools as the company grows and has more resources to
invest.

2.6 Businesses who are in the process of implementing more AI

into their strategy

2.6.1 Digital marketing

Honesty AB (interview 6)
As for Honesty AB, the interviewee has a different vision of digital marketing where he
described his company as a problem solver when other agencies are not capable of delivering
what their customer needs. This is where the light points at the company. The interviewee
pointed out as the company cares more about the level of sophistication, rather than the
volume. Also, mentioned the pros of having 40 AI or robots, where the interviewee
mentioned “I am like a director” and all those 40 AI’s helps the campaigns, ideas and ads
come to life and give it the creativity and inspiration in a more sophisticated way.
Ampilio (interview 7)
Although Ampilio use an excessive amount of AI in their digital marketing, they still have
not reached the point where they tried an AI marketing campaign without human interaction,
the interviewee also mentioned the idea to adapt it in the near future when he invests more
into AI especially strong AI where it can learn and generate data and give suggestions that fit
the customer with the size of their company, the tone and the character as well as reach the
right target group. to optimize their target group even more, the interviewee mentioned the
use of CRM to find potential clients. CRM helps them with the parameters based on the data
and the customers’ needs of services. Despite that, CRM offers information about companies,
their “location, turnover, growth, money, offer, industry etc.” also information on which
platform they are active on and how active and consistent they are. The interviewee also
pointed out that AI has contributed a lot but mainly affected time and value since it helped
the company to work more effectively and sufficiently.

Eniro (interview 8)
According to Eniro, they used an AI digital marketing campaign where the AI helped their
client to reach new customers and the overall performance of AI was 570% more effective
compared to their previous campaign which was handled manually. After that the company
had the objective of having less manually and more automated marketing campaigns.
According to the interviewee, AI will generate more qualitative and effective content, more
accurately, providing more suitable language when talking to different groups of customers
depending on their age and interest. Compared to human labor, AI has much more
advantages in a lot of aspects (profit, margin, reduction cost and more efficiency).
Interviewee also maintains the necessity of adapting to AI, he believes that AI is the future
and not adapting to the future will lead to the failure of the organization and won't be able to
compete in the market.

2.6.2 AI tools
Honesty AB (interview 6)
The interviewee noted that they utilize a variety of AI tools within the company. These
include text-generating AI like ChatGPT-4 furthermore various AI functionalities found in
image editing software such as Photoshop and Pixelmator. Additionally, the interviewee also
stated that they make use of tools named MidJourney and Firefly, which are AI tools that
create images based on text descriptions of the user.

Ampilio (interview 7)
The interviewee noted utilizing numerous AI tools in their digital marketing efforts.
Specifically, the interviewee mentioned using Amanda AI, an automation tool for Google
Ads. Furthermore, the interviewee also mentioned using generative AI for text creation,
which is ChatGPT-4. Moreover, the interviewee noted Guava, an AI-based prospecting
(CRM) tool.

Eniro (interview 8)
The interviewee did not identify the exact AI tools the company uses. However, the
interviewee mentions that they license their AI tools from various providers and have
established strict contracts with these providers. Some of the features of the AI the company
utilized was mentioned by the interviewee and they were logo-generating AI, presentation
AI, And website AI. The interviewee furthermore mentioned that they do not use ChatGPT
for their business operations, however, the interviewee personally uses it for personal writing
tasks.

2.6.3 Ethical
Honesty AB (interview6)
Argued that there should be more regulation and GDPR surrounding AI and regulations and
that politicians should consider this a serious matter to solve it and come up with improved
regulations to secure all parties.

Ampilio AB (interview 7)
The interviewee mentioned the necessity of knowing what to give in an AI based tool, and
what we should put or not. He described it as any other concern where a company doesn't
send a secret message in an email, they make sure that it is safe and secure. Where it is the
same with AI and people should take the same precautions when it comes to censored
information or data. When it comes to the ethical concern of labor Ampilio mentions that
they would not under any circumstances fire any employees because of the use of AI, instead
AI should help employees to be more creative and rely on more important jobs instead of
doing repetitive jobs. He also mentioned that the ethical and moral concern will be a bigger
case in the future. When AI becomes much more involved in our life and strong AI takes
more space than it does now.

Eniro AB (interview 8)
The organization did have one main concern about GDPR and the act of EU organization
with AI law and bylaws and how they are trying to shape the laws toward AI and which
direction they are going with security and privacy policy. For now, the company goes
through manual precautions before publishing or sending in the product or service, securing
the quality and what it holds. These precautions are made because the company is licensing
their AI, having very rigorous and structured contracts in order to secure their own safety
and their customers as well in order to limit the outcome of the tools.
Another issue brought up was the ethical concern of labor and how many people will lose
their job with time and build up a bigger scale between low and high skill professions, and
went deeper to talk about the danger that humanity takes by trusting/using AI and the
possibility of AI taking over humanity. The company had good results from the customer
base, about the company using AI and didn’t raise any issues. Eniro had some doubt that
some customers might not be satisfied with the results, but in general they didn't raise any
ethical concerns toward the customers and the business when it came to the quality and value
of the services.

2.6.4 Financial
Honesty AB (interview 6)
The company was started in February of 2023, so he could not share any results with us yet.
Honesty described this as the industrial revolution where a lot of people will benefit from it
but also some will experience drawbacks from it. He mentioned that AI is bringing even
more efficiency to the world, which will end up leading to rapid profit in the long term.
Where he also mentioned the situation of the market and how it shows a lot of margin, but
with time that will decrease depending on how much the market adapts to AI over time. He
also mentioned the efficiency of AI is increasing the gap between the low-skilled level
and high-skill level of jobs which will lead to cost reduction and bigger margins in general.
Ampilio AB (interview 7)
Also argued that AI brings efficiency, quantity, and quality. At the same time, he also gave
an example of building muscle with steroids. It grows five times faster and for a company, it
is the same with the help of AI. It will grow five times even bigger and get more customers
and offer more services in the long teams, which will generate margin, revenue and cost
reduction and raped growth. The organization adopted AI 2 years ago, however, the
interviewee didn't want to share the profit or the effect of AI to the finance of the company.

Eniro AB (interview 8)
As a big-sized company, they had difficulty sharing their profit, revenue and so on due to the
fact that they are listed on the stock exchange list. Instead, confirming that AI has positively
affected the profit of the firm in a few aspects, cost reduction, margin, and improving
revenue. The interviewee mentioned that the most significant advantage of AI is margin and
quality. Where he argued that margin would benefit both the business and the customer,
where the prices will go low, but also the efficiency will go up, and this will lead to more
customer base with new revenue streams.

2.7 AI retailers who provide businesses with AI-technology

2.7.1 Digital marketing

Infirmamotion (interview 9)
The interviewee mentioned that their company is not using AI tools for digital marketing
purposes except Chat-GPT, sometimes for writing purposes, for instance, emails. The reason
for this is primarily due to the nature of their company, as they help clients with security
solutions, mainly for government entities and police departments. Based on this focus, they
can avoid getting involved in broad digital marketing efforts to reach a general audience,
such as civilians. Instead, they use a strategy they describe as "direct aggressive sale," which
involves directly approaching their niche market by calling or meeting with potential clients.
They also mentioned that they have thought about using AI for marketing but decided against
it due to the specific nature of their target market. He emphasizes that for their industry, it
might not be necessary or beneficial to use AI for digital marketing.
SAS: (interview 10)
The interviewee discusses the role of digital marketing in today's business world, especially
highlighting the impact of AI and automation on traditional marketing practices. The
interviewee also emphasized that AI's role expands to automating decision-making in digital
marketing in today's market. This includes choosing what the companies communicate, for
instance, when deciding which channel to utilize and who the message is intended for. These
types of automation allow businesses to better their marketing efforts and improve efficiency.
The interviewee further explains that AI has improved customer experiences by analyzing the
behavior and preferences of the customer, leading to personalized recommendations and
offerings by the company. As a consequence, this can boost customer satisfaction, conversion
rates, sales, and brand loyalty. However, they also discussed that the usage of AI should not
replace a clear marketing strategy. While AI can offer valuable insights and automation, a
comprehensive strategy is still crucial to steer these processes in the company.

2.7.2 AI tools
Infirmamotion (interview 9)
The interviewee mentioned that the company used Chat-GPT to generate text to double-
check their writing skills and be able to optimize the content in it. Also, they are considering
adopting more tools for their marketing campaigns, but for now, they are satisfied with
chat-GPT.

SAS: (interview 10)

The interviewee talked about the AI solutions that they are providing, which were shown to
be related to machine learning in the form of Website AI. This was mentioned in the form of
tracking the customers' behavior online and generating proposals for what is supposed to be
done for further improvement. Additionally, the use of generative text tools, which is in the
category of Neutral Language processing, in the forms of the AI tools Jasper and ChatGPT.

2.7.3 Ethical
Information (interview 9)
The interviewee discussed the topic of ethics when discussing the use of AI in their security
company. He mentioned data privacy and the GDPR law, which is a primary ethical
consideration when using AI. The interviewee also discussed how they handle client data.
Their way of working is extracting essential information and discarding the footage to respect
customer privacy. He also highlighted the importance of cybersecurity to protect this
sensitive data, which could be a big problem if some hacker managed to hack into the system
and extract the data that was stored.

SAS: (interview 10)

The interviewee pointed out that the concerns of AI are equal to any technological concern
where they described AI as black box. the interviewee mentioned:
- “You can't really verify, for instance, what comes out of ChatGPT in that sense but
when it comes to security perspective, I would say AI systems are like other systems”
So trustworthiness is unprotected when it comes to Chat-GPT when it raises concern whether
the information that is taken and given are accurate enough to use. But when it comes to
cyber-attack or hacking, the threat is equal to any other technological tools and systems.
Additionally, they also mention when it comes to ethical concerns, to make sure the
information is not biased in any way.

2.7.4 Financial
Information (interview 9)
The interviewee mentions that using AI has substantial financial benefits for both their
company and the clients. For instance, the interviewee took an example saying, “AI can
work 24/7 for free, unlike human employees who require salaries and benefits,” which means
that AI is not money-requiring and can benefit the company in ways that human beings could
not. He also emphasized how cheaper it is to use AI for security purposes rather than
employing a police officer to monitor screens. When discussing return on investment (ROI),
they see AI as a positive contributor, highlighting the cost savings and efficiency gains as
critical financial benefits.

SAS: (interview 10)

The interviewee (SAS), who is a retailer of AI, brought up an example from their customers
where they quoted the following regarding the question; “Do you see AI as a part of ROI, if
yes, elaborate why?”
- "Yes, in marketing AI is used to automate and take better decisions. Automation and
increased relevancy will improve customer experience which in turn will lead to
increased loyalty, decreased churn, increased NPS, increased sales (cross- and
up-sell), etc.”
Meaning, that the automation from the AI tools can generate decisions that are matched to
what is being wished for from the customer. Hence, leading to an improved end-customer
experience, which in turn will lead to increased loyalty, decreased churn, increased NPS,
increased sales (cross- and up-sell). Regarding the question of changes in forms of more
revenue or profit, the interviewee answered the following:
- “A Nordic insurance company used our attribution algorithm in their marketing. The
results include an improved customer experience, increased conversions, and reduced
marketing costs for the organization.”
With this in mind, it shows that the results from the use of algorithms resulted in reduced
marketing cost while increasing and improving conversations and experiences with the
end-customers. In the latter, a table representing the tools that has been utilized in the
different businesses has been made, in order to see what has been presented in a visual form
to the addition of the descriptive text by the authors.
 Infra
Shippey Reynolds Quick Mynews
Tools Substorm Honesty Ampilio Eniro motion SAS
AB International Net desk
labs
Website AI ✓ ✓

Amanda AI ✓

ChatGPT ✓ ✓ ✓ ✓ ✓ ✓ ✓

ChatGPT 4 ✓ ✓

Google/Met a ads ✓ ✓

Create Logo AI ✓ ✓

Jasper ✓ ✓

AI-CRM ✓ ✓ ✓

E-mail AI ✓

Presentation AI ✓ ✓

Mid ✓
Journey

Firefly ✓

(Figure 4 - AI tools vs businesses implementation - Created by the authors)

3. Analysis and discussion
This section presents the study's analysis from a theoretical and empirical perspective in line
with the study's overall framework. The study's three main themes - Digital marketing, AI
and Ethics - are covered by the analysis. In addition to this, a division of the business has
been made depending on the size of the organization.

3.1 SMEs
When it comes to the analysis of small and midsize businesses, findings of keywords,
similarities and differences have been covered in the three different topics among the
businesses. Five out of the ten businesses will be covered during this part of the chapter.
3.2 Digital marketing
Among all of these five businesses there were some similarities that needed to be covered.
For instance, the use of at least one sort of text-generating AI has been found to be the
similarity among these SMEs. The keywords “Assistant” and “Tool” were found to be
mentioned when explaining how and why the AI tool was being used. Additionally, their
digital marketing efforts are, for the moment, handled with the decisions coming from
humans. Meaning that the AI is being used as a tool among these interviewees, rather than a
replacement for human labor. The AI generated text from this type of tool was being used for
multiple tasks among the interviewees. The most common use for it, was to generate a text by
AI that is fitted for their purpose and which is placement of the text, which could be for a
website, social media and such. Four out of five of these interviewees brought up the
importance of “fact checking” when using the generative text. This is justified for multiple
reasons and situations. Since the text generative tool is being used to support the businesses
own digital marketing efforts among four out of five of these interviewees, the generative
text has to conform to their platforms and purpose. The aberrant interviewee used the text
generative AI to help answering E-mails in a formal and overview way. Even though the
process before the usage of the generated text was similar, the reasoning behind the
utilization of this type of AI-powered tool, the outcomes and attitudes towards AI, has been
shown to have a significant difference among the interviewees in their digital marketing
efforts. For instance, the reasoning of the utilization of one of the two sorts of this tool, was
since it is free. Two of the interviewees highlighted the financial part of the reasoning. Hence
one of the two mentioned that this tool will be used until a better one is created, with the
addition of it being free. Making it further plausible for this business who has a strong view
and strategy directed against “cashflow”. While the other participant has a list of AI which
they wanted to implement and utilize in the company, but has limitations of doing so for the
moment due to the current financial state. These two latter mentioned businesses were also
those who indicated a positive attitude towards AI tools, while the other two interviewees
sense uncertainty with the use of AI. These opposing attitudes were as a result of one of one
of the two being more reliant on another paid sort of text-generative tool, which was invented
and published before the free version. The other one had no further use of AI other than
answering emails, even though this business was one of the AI retailers themselves.
3.2.1 AI
Another positive mindset towards AI was found in an additional interviewee. A business who
has utilized AI to an even further extent in the marketing efforts. The utilization of AI was
covering different parts of this business which included brainstorming. Hence, allowing the
marketing efforts to be delivered at a faster pace. The reasoning and outcome from the
utilization of AI has shown to have a similar development. Five out of five found it to be
easier to direct the right text to the right target group when using AI. Additionally four out of
five of these businesses found that the production of content has increased with the help of AI
powered tools. To the AI powered tools. Hence, making it possible to have a higher pace of
work through tasks being completed faster, leading to several tasks to be processed
simultaneously. The use of AI tools were found to be connected to financial benefits since the
multiple tasks being delivered were a cause of a higher customer rate due to decreased and
market leading prices. The decreased and marketing leading prices were able to be
implemented among these companies as a result of the use of AI.

3.2.2 Ethical
Ethical considerations were brought up in multiple ways and areas where it has to be taken
into consideration. For instance, what was previously mentioned fact checking when using AI
generated text. The majority of these interviewees agreed that AI is being used as a tool
rather than a replacement. Making it important to make sure that the gathered data is not
being gathered from a blog or another unreliable source has been a concern among these
text-generated AI users. Five out of five interviewees also agreed on the ethical parts of
usage, which is also connected to what you are requesting from the tool. Meaning, that it
should not be an ethical issue, as long as you don't input personal information that could be
of an ethical concern. Another ethical consideration this is in question is the human labor
aspect. Two out of the five businesses advocated, of which one of them already
demonstrated, that the discussion of increased unemployment is a consequence of the
utilization of AI. The AI retailer mentioned that AI will take over the world and unemploy
people working with lower skilled professions. While the business who has utilized AI to a
greater extent already has shown the potential and consequences of utilization of AI. Since
the staff, which was a number of 40, was replaced by AI robots. Which corresponds to the
same number of staff as the previous state. And this has shown to be a risk when
implementing AI.
3.2.3 Discussion
The benefits of the utilization of AI in business are many, where the positive side and the
bringing of the utilization has included a more efficient process due to the fast pace, with the
addition of the ability of having many customers at the same time. Hence, making it possible
to have improved margin since it has shown that it can replace some of the human labor. The
downsides of AI are that there are little to no regulations regarding the use of AI generated
text, and where the gathered data is being gathered from. Since the law of GDPR are not yet
extended to this field of theme. Another consequence to consider is that the wrong data can
be inputted into the text-generator, making the importance of fact checking major. Lastly,
there is always a risk being taken when operating online. The risk of cyber-hacking of the
customers’ information and thus being exposed to external parties other than the business, is
an issue that could transpire, regardless of correct or incorrect usage of AI. Consequently,
market leading prices can enable a more stable and increased customer base, however the
importance of maintaining on the right side of the ethical considerations, has become majorly
increased due to utilization of AI.

3.2.4 Digital marketing

The company Ampilio AB is a forward-thinking company that makes use of AI tools in its
operations. This is an important strength that the company needs as it shows the company's
commitment to harnessing the power of technology to enhance its business processes, which
will keep the company more productive and processes more efficient. The company has a
positive attitude towards AI suggesting that they are likely to continue investing in the latest
AI technologies, keeping them at the cutting edge of this rapidly evolving field.The company
saves time and resources by using AI such as chatgpt to text generate which can better the
efficiency of content creation and customer communication, providing more personalized and
targeted messages to customers. This will make it easier for the company to be more active
in, for example, social media, since they can generate content faster and it allows the
company to connect to its customers quicker. The 'human touch' they incorporate into their
AI tools is important and needed since AI can be based and acts on what it was trained on,
meaning that the company corrects it and takes out the aspects that are not suitable for them. This
approach shows their comprehension of the significance of human interaction which will
help the company to keep strong customer relationships. Furthermore, Ampilio fact checks
the outcomes from the AI tools. By doing that it will help the company to build trust and
credibility with customers and other stakeholders, which is vital in establishing a strong
reputation in the marketplace.

Ethics However, there is a potential concern related to data privacy. As the company uses AI,
CRM, and SEO tools, it will collect and also process a large amount of its customer data. In
that sense, it's crucial that Ampilio has strong data privacy measures in place to protect this
data information and comply with relevant laws and regulations. This will keep the
company's competitive advantage and will lead to reputational success for the company.

Mynewsdesk is already utilizing AI for text generation, which is an innovative approach, as

earlier mentioned allowing for more efficient, accurate, and consistent content creation. This
can enhance communication with clients and potential customers, saving time and allowing
the employees to be more productive and do more tasks during the day. Furthermore, the
company implemented AI in CRM resulting in better customer insights, and improved
customer services which will increase customer satisfaction. The company also utilizes AI
for matching clients of the company with suitable journalists, which can make this process
faster than it would take a human being. This ensures a better fit based on the client's
requirements and the journalist's expertise. This can better client satisfaction and lead to more
successful collaborations. It is then important to factcheck when the AI finds, for instance,
the journalist the company's employees have looked into manually and when they are
satisfied connect the client and the journalist.

Quicknet, uses ChatGPT and Jasper, both text generating AI, which indicates its commitment
to embracing advanced technologies and improving operational efficiency. Its practice of
AI-based fact-checking shows that the company relies on these AI’s for accuracy, which is
vital in today's world where it's important to be faster and more efficient than the
competitors. Furthermore, the company shows a positive attitude toward AI and openness to
innovation, which could provide a competitive edge in the rapidly evolving technological
landscape, especially this market that the company operates in.However, the company has
some downsides if not taken into consideration. For instance, its intention to let AI handle
social media can be a cause for concern. While AI can automate certain tasks,it may not
currently
have the capacity to fully grasp the tasks of human communication, leading to possible
misunderstandings or inappropriate responses that could disgruntle customers. This way of
using AI might result in a lack of personal touch, which is often appreciated by customers.

When it comes to big sized companies, implementing AI might take more time and effort but
it could also give better resolution and margin in the end. Implementing AI was one part of
the interviewees plan, which he called it the 90 day-plan, where he wanted to revolutionize
the company and add value so one of the moves was to implement AI in the system because it
was seen as a future tool As of a result of the company being listed on the stock exchange,
they were unable to share with us the names of the tools but instead they described what kind
of effect or what kind of help it gave them we can see mainly that. It affects three significant
areas, namely, margin, effectiveness and lastly, cost reduction. These are the three elements
that will help the company stand its ground and gain more customers and profit in the long
term. Yet again, because of their stock listed we couldn't get any information about their
financial situation. Implementing AI gives the company a boost in many areas when it comes
to digital marketing and some of the tools

The tenth business (SAS) has already invested a lot in AI which gives them a lot of positive
attitude, and some of those tools actually help them to even improvise and give them
recommendations of how they can reach their customer. Even more than doing their analysis,
it's improved efficiency communication and again cost reduction
The company uses AI to monitor their website and give proposals through AI and they are
even using it in a CRM system where they can find and target clients. AI has brought more
efficiency and cost reduction and increased logistics which is affecting positively on the
company due to their target group.With the help of AI they can resonate their proposals
whether it's fit for their target group and benefits them while providing a positive attitude.
On the other hand, some of the personal information on how they are using their information
in case they are selling it to the third party could raise a lot of suspicion and affect the
company negatively.

In general, we can see that all companies are adapting to AI whether in whether at a fast pace
or a slow one, because there is an understanding in the market that not adapting to the new
technology might lead to the company being pushed out of the market, like it happened back
in 2000s where when companies that didn’t want to adapt to the new technology went
bankrupt and went out of the market, while those who adapted to the technology and the new
ways of market functionality, survived and thrived. From our chart below, you can clearly
see that one of the most used tools is ChatGPT since it is actually also a free tool that
companies use to generate more effective results. On the other hand, with big sized
companies, they are much more invested in other tools such as building websites, optimizing
ads and building logos which affects not just their productivity but their creativity as well
leading to better results.

As discussed in the theoretical framework, (Du et al., 2020) explained the functions of the
different sorts of AI tools, where “this intelligence is built using complex algorithms and
mathematical functions” based on the following:

Machine learning - Provides a machine with the capability to learn from data and
experience through algorithms.
Deep learning - A machine that learns techniques that trachea technique which
comes naturally to humans.
Neural network - This is used to learn things, this is a replica of structure and replica
of human brain, we will be able to get cognitive capabilities in machines.
Natural language processing - Writing and reading text in a language is the field of
NLP.

With the findings from the primary data along with the secondary data, it has shown that
Natural language processing in the forms of text-generating AI such as ChatGPT and Jasper
has become a popular choice of AI tool made by the majority of the participating businesses.
Additionally, the rise of Machine learning has shown its popularity among the businesses in
the forms of generating and learning from data that is stored, or even new data that is
presented to the AI. This was found to be used by the businesses in the forms of AI tools who
generates various digital marketing tasks such as:
Websites, logotypes, pictures and ads. Where Google ads, Meta ads, Firefly and Midjourney
were some of the names of the AI powered tools being used by businesses. This was most
commonly found to be utilized among SMEs. Lastly, deep-learning in forms of AI CRM
systems has shown a significant utilization among one midsize enterprise, and the two large
enterprises that participated in this study.
A table has been summoned to summarize the different types of AI tools that have been
utilized in the businesses.

Tools Utilization of AI-tools in Businesses

Text generative AI (Neutral language

Eight out of Ten
processor)

Website generative/tracking AI (Machine

Two out of Ten
Learning)

Ads generative AI (Machine Learning) Three out of Ten

Logo generative AI (Machine Learning) One out of Ten

AI CRM (Deep Learning) Three out of Ten

E-mail generative AI (Neutral network) One out of Ten

Presentation generative AI (Machine

Two out of Ten
Learning)

Picture generative AI (Machine Learning) Three out of Ten

(Figure 5 - Utilization of AI-tools in Businesses vs Tools- Created by the Authors)

4. Conclusion
In conclusion, the purpose and research question for the thesis was formed to be:
“The purpose of this study is to examine how organizations use AI technology in their digital
marketing, and how this affects the effectiveness of the digital marketing process. This in
conjunction with any repercussions regarding ethical and legal concerns, that may occur
through the utilization of AI. Thus, the following will be the research question for this paper:

- How do businesses utilize AI and how does it impact digital marketing?

Based on our purpose and question, these conclusions and answers can be drawn from this
study.

The utilization of AI in digital marketing has taken off and established itself in businesses'
digital marketing efforts and strategies. All of the interviewed businesses are operating
mainly in the field of digital marketing, except for the two AI retailers. The popularity of text
generating AI tools has gained momentum in its utilization in the past year due to the
increasingly talked about ChatGPT taking its place in the market. Where the findings of this
thesis have shown significant similarities and differences with this type of tool. It has helped
businesses to generate text used for emails, brainstorming ideas about text for posts to their
digital marketing etc. Furthermore, the use of this tool was more common to be utilized in
SMEs than larger sized businesses, where it was more used for personal reasons.

Additionally, the study found mid size and larger businesses to have further considerations
regarding the ethical and legal precautions. For instance, one large sized business walked the
interviewers through the ongoing regulations being set regarding the use of AI and the
regulation around GDPR, and that the business is well aware that these regulations and laws
must be taken into account when utilizing AI to a further extent into their digital marketing
efforts and strategies. Other than that, it was found that the SME interviewees thought about
the ethics regarding the use of AI was dependent on what data is being entered. Furthermore,
making it dependent on what is being entered from the user was the precautions being taken
to considerations from the latter mentioned types of businesses. The study’s interviewees
came to demonstrate a generally positive attitude towards AI among the selected
interviewees. However, The utilization, on the other hand, has turned out to be different
among the different companies that have been interviewed in this study. The utilization took
place in different ways in connection with different reasonings mentioned by the businesses.
For instance, was the choice of using free AI tools a matter related to financial measures and
current budget. Regardless of whether it was because the company does not have the budget
to utilize AI tools that must be paid for, or in other cases where the financial decision
unfolded in the form of using it as long as it works to the degree of satisfaction desired by the
company. The financial considerations also meant that businesses that were of a larger size,
proved to possess the financial resources required to implement several differentiated and
advanced AI tools such as AI-CRM.

Hence, we can see the effectiveness of AI tools is mainly targeting the margin, reduction of
cost and quality of products and services, which could lead to more profit and more customer
gain in general. But to deny that this could also lead to some challenges such as creating a
gap between low and high skilled jobs, where AI could take over would be unwise. The size
of the company plays a big role in whether the company could invest in AI or not even
though investing in AI could be expensive, it has a lot of benefits in different aspects.
Whereas big sized companies could invest more and gain more in the long term, they have
difficulties implementing the new technology in the company which could take time and
effort. Meanwhile, small and midsize businesses have more capacity and flexibility to adapt
to AI, and grow rapidly with the help of AI in their digital marketing. Hence, making them
able to enter new markets, and gain more revenue.

This way, the study has confirmed previous studies about AI being a valuable and profitable
business tool. Additionally, the findings from the study have contributed to the extent of
existing studies due to an in-depth, proven, and demonstrated summary of the utilization of
AI tools in businesses, specifically in digital marketing in the Swedish market.

5. Limitation of the research

The most significant limitation of this study was the chosen geographical focus on
Exclusively Sweden. The findings were primarily formed by Swedish social culture,
economic, and business conditions, and may not be universally applicable. Our focus on
businesses was based solely in Sweden, which could result in a potentially incomplete
picture, as experiences and operations of global corporations were not considered for this
study. A further significant limitation lies in the fact that most companies that were
interviewed, had only recently started investing in AI. Consequently, these companies were
unable to give us useful data for the financial part of their implementation of AI. This lack of
mature AI implementation data is also an important limitation, as it could have helped our
understanding and contributed to the comprehensiveness of the study.furthermore another
notable limitation of this study was the time pressure under which the study has been
conducted. Despite precise planning and organization, the strict deadline set a significant
limit on the depth and breadth of the study. As is true for most research efforts, this study
was bound by a preplanned timeline that included different activities such as data collection,
data analysis, interpretation of results, and reporting. With more time the authors would have
more time to make the study broader and more in-depth.

6. Recommendations for further research

This study has analyzed the importance of AI in digital marketing for businesses. According
to today’s market, the researchers used primary and secondary data to have a better
understanding for a clear analysis of the state of the utilization of AI. However, the study on
AI and digital marketing cannot be comprehended in this single study. Furthermore, new
research can be done by using different data from different regions, states, and countries.
Additionally, more of what is yet to come regarding AI, since it is constantly changing, with
the addition of regulations and laws that will be regulated around this fairly new
phenomenon. Therefore this study is primarily limited to only one region and its current
state, and for a better understanding of this subject, further studies are needed. To provide
with an even wider and concluding view of other regions and countries' situations depending
on the utilization and the current regulations attached to it.
References:
- Alareeni, B., & Hamdan, A. (2022). How Artificial Intelligence Revolutionizing
Digital Marketing. In Impact of Artificial Intelligence, and the Fourth Industrial
Revolution on Business Success (Vol. 485). Springer International Publishing AG.
https://doi.org/10.1007/978-3-031-08093-7_13
- Arntz, M., Gregory, T., & Zierahn, U. (2016). The Risk of Automation for Jobs in
OECD Countries: A Comparative Analysis. OECD Social, Employment and
Migration Working Papers (No. 189). OECD Publishing.
- Bessen, J. (2019). Automation and jobs: when technology boosts employment*.
Economic Policy, 34(100), 589–626. https://doi.org/10.1093/epolic/eiaa001
- Bryman, A., & Bell, E. (2017). Företagsekonomiska forskningsmetoder.
- Brynjolfsson, E., & McAfee, A. (2017). The business of artificial intelligence.
Harvard Business Review. Retrieved from
- Camilleri, M. A. (2017). The Corporate Social Responsibility Notion.In Corporate
Sustainability, Social Responsibility and Environmental Management (pp.
3-26)..Cham, Switzerland: Springer Nature.
- Chatterjee, S., Ghosh, S. K., Chaudhuri, R., & Nguyen, B. (2019). Are CRM systems
ready for AI integration? The Bottom Line: Managing Library Finances, 32(2), 144–
157. https://doi.org/10.1108/bl-02-2019-0069
- Chaffey, D., & White, G. R. (2022). Digital Marketing: Strategy, Implementation and
Practice (8th ed.). Pearson.
- Davenport, T. H., Guha, A., Grewal, D., & Bressgott, T. (2020). How artificial
intelligence will change the future of marketing. Journal of the Academy of Marketing
Science, 48(1), 24–42. https://doi.org/10.1007/s11747-019-00696-0
- Desai, V, & Vidyapeeth, (2021). Digital marketing: A review. International Journal
of Trend in Scientific Research and Development, 5(5), 196-200.
- Desai, V. (2019). Digital Marketing: A Review. International Journal of Trend in
Scientific Research and Development, Special Issue(Special Issue-FIIIIPM2019),
196–200. https://doi.org/10.31142/ijtsrd23100
- De Bruyn, A., Viswanathan, V., Beh, Y. S., Brock, J. K., & Von Wangenheim, F.
(2020). Artificial Intelligence and Marketing: Pitfalls and Opportunities. Journal of
Interactive Marketing, 51, 91–105. https://doi.org/10.1016/j.intmar.2020.04.007
- De Mauro, A., Sestino, A., & Bacconi, A. (2022). Machine learning and artificial
intelligence use in marketing: a general taxonomy. Italian Journal of Marketing,
2022(4), 439–457. https://doi.org/10.1007/s43039-022-00057-w
- Dumitriu, D., & Popescu, M. (2020). Artificial Intelligence Solutions for Digital
Marketing. Procedia Manufacturing, 46, 630–636.
https://doi.org/10.1016/j.promfg.2020.03.090
- Du, X., Watt, F. M., Luscombe, N. M., & Lynch, M. D. (2020). What is AI?
Applications of artificial intelligence to dermatology. British Journal of Dermatology,
183(3), 423–430. https://doi.org/10.1111/bjd.18880
- Frey, C. B., & Osborne, M. A. (2013). The future of employment: How susceptible
are jobs to computerisation? Technological Forecasting and Social Change, 114,
254–280. https://doi.org/10.1016/j.techfore.2016.08.019
- Gołąb-Andrzejak, E. (2023). AI-powered Digital Transformation: Tools, Benefits and
Challenges for Marketers – Case Study of LPP. Procedia Computer Science, 219,
397–404. https://doi.org/10.1016/j.procs.2023.01.305
- Hamdan, A., Hassanien, A. E., Khamis, R., Alareeni, B., Razzaque, A., & Awwad, B.
(Eds.). (2021). Applications of artificial intelligence in business, education and
healthcare (Vol. 954). Springer Nature.
- Hoofnagle, C. J., van der Sloot, B., & Zuiderveen, Borgesius, F. (2019). The
European Union General Data Protection Regulation: What it is and what it means.
International Data Privacy Law, 9(2), 65-98. doi: 10.1093/idpl/ipz002
- Huang, M., & Rust, R. T. (2018). Artificial Intelligence in Service. Journal of Service
Research, 21(2), 155–172. https://doi.org/10.1177/1094670517752459
- Iskandar, M. S., & Komara, D. (2018). Application Marketing Strategy Search
Engine Optimization (SEO). IOP Conference Series: Materials Science and
Engineering, 407, 012011. https://doi.org/10.1088/1757-899x/407/1/012011
- Joachims, T., Granka, L., Pan, B., & Hembrooke, H. (2005). Accurately interpreting
clickthrough data as implicit feedback. https://doi.org/10.1145/1076034.1076063
- Kaponis, A., & Maragoudakis, M. (2022c). Data Analysis in Digital Marketing using
Machine learning and Artificial Intelligence Techniques, Ethical and Legal
Dimensions, State of the Art. https://doi.org/10.1145/3549737.3549756
- Kannan, P., & Li, H. (2017). Digital marketing: A framework, review and research
agenda. International Journal of Research in Marketing, 34(1), 22–45.
https://doi.org/10.1016/j.ijresmar.2016.11.006
- Kumar, V., Dixit, A., Javalgi, R. G., & Dass, M. (2016). Research framework,
strategies, and applications of intelligent agent technologies (IATs) in marketing.
Journal of the Academy of Marketing Science, 44(1), 24–45.
https://doi.org/10.1007/s11747-015-0426-9
- Lewandowska, A. M., Wolski, W., & Zieliński, M. (2019). Website content
optimization in search engine results pages. Journal of Retailing and Consumer
Services, 49, 137-146.
- Loureiro, S. M. C., Guerreiro, J., & Tussyadiah, I. (2021). Artificial intelligence in
business: State of the art and future research agenda. Journal of Business Research,
129, 911-926.
- Makridakis, S. (2017). The forthcoming Artificial Intelligence (AI) revolution: Its
impact on society and firms. Futures, 90, 46–60.
- Mittelstadt, B., Allo, P., Taddeo, M., Wachter, S., & Floridi, L. (2016). The ethics of
algorithms: Mapping the debate. Big Data & Society, 3(2), 205395171667967.
- Mogaji, E., Soetan, T. O., & Kieu, T. A. (2021). The implications of artificial
intelligence on the digital marketing of financial services to vulnerable customers.
Australasian Marketing Journal (Amj), 29(3), 235–242.
https://doi.org/10.1016/j.ausmj.2020.05.003
- Mogaji, E., Olaleye, S. A., & Ukpabi, D. C. (2020). Using AI to Personalise
Emotionally Appealing Advertisement. In Advances in theory and practice of
emerging markets (pp. 137–150). Springer International Publishing.
https://doi.org/10.1007/978-3-030-24374-6_10
- Nair, K. S., & Gupta, R. (2021). Application of AI technology in modern digital
marketing environment. World Journal of Entrepreneurship, Management and
Sustainable Development, ahead-of-print(ahead-of-print).
https://doi.org/10.1108/wjemsd-08-2020-0099https://www.bokus.com/bok/978914711
2074/foretagsekonomiska-forskningsmetoder/
- Nguyen, B., & Simkin, L. (2013). The dark side of CRM: advantaged and
disadvantaged customers. Journal of Consumer Marketing, 30(1), 17–30.
https://doi.org/10.1108/07363761311290812
- Pan, B., Hembrooke, H., Joachims, T., Lorigo, L., & Granka, L. (2007). In Google
We Trust: Users’ Decisions on Rank, Position, and Relevance. Journal of
Computer-Mediated Communication, 12(3), 801–823.
https://doi.org/10.1111/j.1083-6101.2007.00351.x
- Qin, X., & Jiang, Z. (2019). The Impact of AI on the Advertising Process: The
Chinese Experience. Journal of Advertising, 48(4), 338–346.
https://doi.org/10.1080/00913367.2019.1652122
- Saunders, M. N. K., Lewis, P., & Thornhill, A. (2019). Research Methods for
Business Students (8th Edition) (8th ed.). Pearson.
- Smith, J., Johnson, M., & Lee, S. (2021). The Digital Transformation of Everyday Life
and Organizational Practices: A Systematic Review and Future Research Agenda.
Computers in Human Behavior.
- Steimers, A., & Schneider, M. (2022). Sources of Risk of AI Systems. International
Journal of Environmental Research and Public Health, 19(6), 3641.
https://doi.org/10.3390/ijerph19063641
- Tiautrakul, J., & Jindakul, J. (2019b). The Artificial Intelligence (AI) with the Future
of Digital Marketing. Social Science Research Network.
https://doi.org/10.2139/ssrn.3405184
- Tien, N. H., Diem, P. T., Vu, N. T., Dung, H. T., Dat, N. V., & Duc, P. M. (2021). The
process of CRM system implementation at Dien May Xanh in Vietnam. International
journal multidisciplinary research and growth evaluation, 2(4), 761-768.
- Yau, K. A., Saad, N. M., & Chong, Y. (2021). Artificial Intelligence Marketing
(AIM) for Enhancing Customer Relationships. Applied Sciences, 11(18), 8562.
https://doi.org/10.3390/app11188562
- Van Esch, P., & Black, J. S. (2021). Artificial Intelligence (AI): Revolutionizing
Digital Marketing. Australasian Marketing Journal (Amj), 29(3), 199–203.
https://doi.org/10.1177/18393349211037684
- Vlačić, B., Corbo, L., Costa e Silva, S., & Dabić, M. (2021). The evolving role of
artificial intelligence in marketing: A review and research agenda. Journal of Business
Research, 128, 187–203. Elsevier Inc.
- Wirth, N. (2018). Hello marketing, what can artificial intelligence help you with?
International Journal of Market Research, 60(5), 435–438.
https://doi.org/10.1177/1470785318776841
- Yalçın, N., & Köse, U. (2010). What is search engine optimization: SEO?.
Procedia-Social and Behavioral Sciences, 9, 487-493.
- Yankelovich, D., & Meer, D. (2006). Rediscovering Market Segmentation. Harvard
Business Review, 84(2), 122-131.
 - Yau, K. A., Saad, N. M., & Chong, Y. (2021). Artificial Intelligence Marketing
(AIM) for Enhancing Customer Relationships. Applied Sciences, 11(18), 8562.
https://doi.org/10.3390/app111885

Graph/Table APPENDIX

Businesses who have implemented AI to an extent into their strategy Six out of Ten

Businesses who are in the process of implementing more AI into their strategy Two out of Ten

AI retailers who provide businesses with AI-technology Two out of Ten

Infra
Shippey ABReynolds Mynews
ompany: QuickNet Substorm Honesty Ampilio Eniro motion SAS
International desk
labs

dustry: DM DM DM DM DM DM DM DM Security Aviation

rm ownership:EP HC Concern Concern Concern Concern Concern HC EP HC

mployees: 45 163 23 38 26 930 16 214

terviewee: CEO CEO CMO PIM CMO CEO CEO CEO CEO SA

1.Shippey AB

2.Reynolds
International AB
Businesses who have implemented AI to an extent into their Five out of Ten
3.QuickNet
strategy
4.Mynewsdesk

5.SubStorm
 Businesses who are in the process of implementing more AI 6. Honesty
Three out of Ten
into their strategy
7.Ampilio

8.Eniro

AI retailers who provide businesses with AI-technology 9. Inframotion labs

Two out of Ten
10. SAS
3

Infra
Shippey Reynolds Quick Mynews
Tools Substorm Honesty Ampilio Eniro motion SAS
AB International Net desk
labs

Website AI ✓ ✓

Amanda AI ✓

ChatGPT ✓ ✓ ✓ ✓ ✓ ✓ ✓

ChatGPT 4 ✓ ✓

Google/Met ✓ ✓
a ads

Create Logo ✓ ✓
AI

Jasper ✓ ✓

AI-CRM ✓ ✓ ✓

E-mail AI ✓

Presentation ✓ ✓
AI

Mid ✓
Journey

Firefly ✓

Tools Utilization of AI-tools in Businesses

Text generative AI (Neutral language

Eight out of Ten
processor)
 Website generative/tracking AI (Machine
Two out of Ten
Learning)

Ads generative AI (Machine Learning) Three out of Ten

Logo generative AI (Machine Learning) One out of Ten

AI CRM (Deep Learning) Three out of Ten

E-mail generative AI (Neutral network) One out of Ten

Presentation generative AI (Machine

Two out of Ten
Learning)
Picture generative AI (Machine Learning) Three out of Ten
5

Interview APPENDIX

Appendix 1
Question:

Could you state a little bit of your background? Your position in the company?

Answer:

I work as a photographer.

Question:

Yes, and so what do you know about AI?

Artificial intelligence, it’s…

Follow up question:

Meaning, What do you think of AI when you hear the word AI?

Answer:

I'm trying to be positive and embrace it. Not to be scared of it. Rather than use it. I can see
the potential of not being not that great in certain areas of life. like computers and robots
starting to take over and talking to each other and making their own bad decisions. But for
me as a photographer, it's not really a concern of mine what I use.

Follow up question:

how do you use AI?

how I use AI. Alright. I use it more in my marketing for website SEO, how to find out what
to search… Engines of the web. Yeah, engines. Also, how to write for me. I don't use
ChatGPT like everybody else. I use something called Jasper… So I've used it before all this
thing about ChatGPT even started.

Question:

When did you actually become interested in AI?

Answer:

So. I've been interested in a long time. I'm a programmer from start, system architecture, so
IT stuff is very interesting for me. It's not what I do daily life anymore I used to, but it's still
an interest that I have. So sort of how can I use AI to sort of make it easier for myself. So I've
been interested for a long time. I can't say a date really. Yeah, I understand. But it's more of
what can I do. I mean, I'm a solo entrepreneur, so I need to do everything myself and it's only
24 hours a day. Everybody has it. I wish we had more, but we don't. So how can I be more
efficient? Rather than working full time, maybe work half time or work full time and just be
able to do more. So for me, I choose Jasper as in helping me writing blog posts to make sure
they're optimized for the search word that I've decided. I can see what other search words I
also need to use to make it even more optimized. It helps me write, but I'm very... The
concern comes to you need to rewrite. It's like Chat GPT. What you put in is what you get
out. So you might not actually get what you want. So you need to make sure how to actually
teach yourself how to make the prompts. How do you ask the questions correctly? Because
how you ask the questions is how you get the answers. That's not what I meant. Well, you
probably asked a stupid question. Well, not stupid, but not in the right way. So he understood
what you're saying.

Question:

How many AI tools do you use for your job or in your daily life?

Answer:

I got software for editing to softwares but they're sort of the same but they're different
generations and then for writing. Okay that's all right.

Question:

Are you aware of any competitor or similar, entrepreneurs that are using AI or
implementing AI?

Answer:
More and more small entrepreneurs such as myself, maybe not photographers as such, but
communication media firms, they're using Chat GPT for marketing to help them write, help
them write prompts for LinkedIn for instance, and help them write blog posts, help them
come up with the ads text.

Follow up question:

But do you think this is giving you kind of an advantage right now in the market?

Answer:

Oh yeah, definitely.

Question:

Are there some AI tools that you are wishing to adapt in the near future?

Answer: Nothing that I can think of now

Question:

How did you go about selecting the right AI tools and technologies for your business
needs? And what were some of the main factors you considered?

Answer:

Just for the writing, there's obviously a lot of tools out there. Went on recommendations and
just, not verbal recommendations, because I don't know anybody in Sweden using it, but,
because it's an American thing, but just going on recommendations of what I saw online,
what was, that it seemed good. It was my husband actually that found it and said, this is what
you need.

Question:

how has the integration of AI impacted your marketing team or your marketing
workflow and process and how have you addressed any potential potential challenges or
concerns?

Answer:

Well, the challenge is how to actually write it, how to prompt and get the right answer.
Because I'm kind of thinking, well, I'll just do it quicker myself. Because I know what I want
to write, but I want him to do it for me, but like, well, I can just quickly do it myself rather
than... because he doesn't understand exactly what I want him to write, or I need to prompt
him with so many things, which I think is like, well that's not interesting right now, but it
obviously is. So I just need to work around that, thinking well he needs this information to do
a correct output. So I still think he's probably better at writing than myself, or quicker, just
that we need to put it in different places.

Follow up question:
Did it happen that, let's say, one time you had a work but you don't know what to write
about like what to write or how to write.

Answer:

Oh yeah, the subject's ideas. Yeah exactly, that's very good. It's like sometimes I want to do a
blog post or a LinkedIn post about okay five tips about what you should think about when
you're photographing portraits. I mean I know the answer, I can write it down, because I do it
all day long. But I ask him, and he tells me like this and this and that. So I just read through
it, thinking, yeah, that's okay, right, because I know I'll delete that one, because it's not that
important. Because sometimes, like, because you do it so often, you kind of, you don't think
that, oh, maybe other people don't know about this. But then the AI reminds you, like, think
about those stuff, yeah. That's what matters. The concern I'm having is, though, can Google
see if it's AI-generated, because it's a tool as well, saying, okay, this text is AI-generated, this
text is not. But I've tested this tool, and sometimes it's right, sometimes it's wrong. I've
written this all by myself, and it's AI, and I'm like, no it's not, that's me. So it's sort of
undecisive for my research so far that can he detect it or not. But I think it's important for
everyone using AI just for writing, not just not take the text, copy paste it and be done. You
need to sort of add your human touch and maybe wrong spellings and stuff. or whatever.
Well, fact-checked of course. Exactly, because AI can be correct like 98% so far, but not like
10 0% so it always has some bugs.
Well, you can make up as well. If you write about Helen shippey, it'll be like, no, that's not
really me. That's not right. I think I was married to somebody. I forgot his first name, but
there's a shippey author. Okay. That's a lot older than me. im like, no, I'm not married to him.
He's like, no, no, you're wrong. But if I'm somebody called shippey, and It goes, oh, I put
these people together. Similarly, why not? Yeah, it makes total sense.

Question:

how have your customers' response to the use of AI in your digital marketing efforts?
And how you notice any changes in customer behavior?

Answer:

I'm hoping they won't notice. I don't want to brag about, oh, I'm using AI, because people are
still like, oh, no, did you? You should be honest and do your own work, sort of like a
Swedish sort of thing. Everybody should be hardworking. Yeah, you should be, but you also
need to be smart working. So it's not like I'm telling everybody, apart from the other day, I
need to delete that comment now, that I'm using it because I don't want people to really
know. And also I'm thinking, well I don't want any competitors to catch on. But I'm still
happy and confident with myself anyway that, okay well, it's all photographers can... It's
there for everyone to use, it's just a matter of how you use it. I mean true, you can also use it
as like your own... As you said, like an entrepreneur. So you can make this one of your tools,
or like you talk about it.

follow up question:

I just realized Lightroom and Photoshop just came out with a new version, I think it
was November or December, and there's loads of AI in it. Even Premium Pro, I don't
know if you use it?
Answer:

Yeah, I do use that. So that's another software that I do use that has AI in it. So it makes it
just more convenient and easier for just anyone, by pressing a button and it's done.

follow up question:

But do you feel intimidated

now? Answer:

It's like everything you are using, it's kind of becoming much more AI, much more
automated. I'm feeling it. When it came to Lightroom and Photoshop, which I've been using
for ages and learned the hard way of how to replace a sky, how to do, replace something.

Yeah. I'm feeling a bit cheated. Like now I don't even need to know all the skills I have and
just press a button and anyone can do it even if you're five years old like what's the point so
yes a bit intimidated but I'm trying not to go down that road.Everyone is a photographer these
days. No, it is intimidating and a bit scary, but when you're a certain age, you're coming,
you're finished your school, and you're going out to work and not sure what you're gonna
work with, but if you're in communication or something, PR, they're asking you to be a
photographer, they're asking you to know the social media, you need to be a content creator,
you need to be everything. So yes, that will take jobs from me, for sure. Not like the big
campaigns, but for normal daily life, social media, daily thing, weekly, it will take jobs from
a sole photographer, for sure. Just because it's easy. Or it's handy. But I can also see it in a
perspective of, well, since it's press of a button for you, it's the same for me. So that will
speed up my work as well.. I mean, So what I used to take a few hours, now I can take a few
minutes. As long as you have a computer smart, or fast enough. That's right. Because it takes
more of power. Exactly. Than using AI. Yeah. That's a really good thing.

Question:

That's what I'm like, a little bit still thinking about. Okay, can you provide any
examples of successful AI based digital marketing campaign that your company has
already done?

Answer: I haven't done any campaigns as such, as marketing campaigns. I don't really do
those at the moment., obviously. When it becomes personal, then people always like. So this
morning I had written earlier, but I had posted it, and I was like, well do you remember? I
talked about my first practice, or internship. So I remember, and so people, and then I say, oh
well today I have somebody I want to be a photographer. But then I had, my question was,
well, do you remember your first internship? And it's like, whoa, everybody wanted to go
memory back lane, like, oh, I remember this, and I'm like, wow, are you so... It's a lot of a
nonsense post really, but it got interaction that I hadn't... I mean, like, why? Why are you so
interested in this? But then for the algorithm, obviously, that's good. I think it's also like the
storytelling of behind, like, do you remember your first... I think everybody just like, go back
and go back memory lane like oh yeah I remember that time
Question:

how do you ensure that your AI based marketing effort are ethical and responsible
sorry respectful of customer privacy if you actually been into that situation.

Answer:

I don't collect any data. But for instance, let's say, um. I mean, I have a newsletter that
collects, but that doesn't have anything with AI to do today. I might have AIhelp me write the
newsletter, but it wouldn't collect the email addresses. name or addresses or stuff like this.

Follow up question:

Would that be necessary for you to know where your customers live or else

Answer:

It's interesting to know, but I don't think I'm not collecting it at the moment. Yeah. But it's
interesting, of course. The more knowledge I have, the better. And it becomes more and more
easier to advertise. But I still think that has to do with how I would word and phrase. If it's an
advertisement, then where would I want to advertise?
I mean, from which cities, which countries? But it still doesn't have to do with AI, I think.
Unless Facebook actually uses their own money. We were thinking if you have this specific
name of some sort of customer you want to reach out to, it may be, like you say, you are the
one depending on what you type in. So if you type in a name, that would be an ethical issue,
but you reach out to a bigger target than just one person. As far as I understood your answer,
you don't use their names and addresses, you just type like, västerås, Yeah, no, I don't think
it's applicable for me. Yeah. But, yeah. Understood.

Question:

How could AI be useful in your marketing, especially help you organizing to adapt to
the future of the market,

Answer:

Well, you need to then predict how the change is going to do. Is it tilting right or left?. I think
that the challenge is just to stay on top, not to be afraid of anything, just to sort of check it
out. I mean, if chat's GPT is something interesting, well, check it out. I'm changing my
website soon, the layout and everything. And I listened to a SEO guy that was using chat
GPT a lot. There's a paid version. There's a free version. He's using the paid version. And the
chat GPT can help you with finding all your sites on your own website that you need to do
301s for. You can just do a script for that, so you easily can, I mean that's amazing. I mean,
you can speed up things like that. I wouldn't do it myself anyway,

Question:
 what ethical concerns or potential negative consequences do you think should be
concerned when integrating AI into digital marketing efforts in your company?

Answer:

Well, you should think about where all the addresses, if you have a CRM system or a
newsletter, and you have emails, think about how do they actually arrive there? Is it ethical?
Are they all, have they signed up properly? Or have they not? Or have you actually scouted
the internet and through AI and found them? Because then obviously that's not ethical. And
also, do you need to have so much information about that person? Or is the email enough?
Do you need to have the city and what they work with and age and stuff, which you probably
don't really need. It could be nice to use to know, but you don't need it. Most of the time.
Marketing. Now I just think fact check and make sure.

Question:

Have you seen any increase in customer recruitment to your company after using

AI? Answer:

Probably a little bit. I've seen increase in website, people checking out the website. Then you
obviously need to convert them, that's up to you. But yes, I would say yes to that. Maybe not
loads, I would want more, but it's a work in progress, you'll never be finished.

Question:

you said already you're using like your AI to kind of generate your SEO. Have you ever
ended up like having a problem with Google about that?

Answer:

I don't know. Don't know is the answer to that. I haven't heard about it. I've only heard about
that that you can feel that it's only written by AI, the text itself, you can feel that which I'm
pretty sure you can. There's obviously nobody saying yes or no, nobody knows, but I think
you can, but that's why you always need to have your own touch. You cannot generate 100%
or even 90, you probably need to have 80 at the most. But again you never did actually like
copy paste just because you always add your own touch. At the most I think it's going to be
boring even though you're probably going to tell them oh right it's funny it's still going to be
you need sort of like something first. I did I think I wrote something the other day and I
could I read the text and I could feel something's missing. I think it's sort of like the the
nuance of human was more like they, they, they instead of I or something, it was something
like, well, this is not really right, but hmm.

Question:

Will incorporating AI in your marketing make it simpler to achieve your goals?

Answer:
Well, I think I'm writing a book. And it's not done. It should have come out last year, but it
didn't. It still hasn't come out, because it's not written yet. But it's a great idea. But I think that
book or a chapter to peek into, or no, not chapter to peek into, AI can help me a lot with just
writing it, speeding up that time. And I think having a book would probably make me more
different in my business.

Question:

Do you trust AI to make decisions based on data that it's collected? And would that
replace some of your employees?

Answer:

I mean, you don't have employees I suppose, but you're intern. Well it could, no I don't think
I'll trust, because sometimes I outsource editing for instance, so it could maybe outsource if I
do have AI doing editing as fast as, fast enough, and I don't need somebody to outsource it,
so that could be cheaper. So in that sense I could use AI for that. I use a lot of gig
companies, like you do that for me, you do the website, you do that. I don't employ people,
but I have somebody else do the work for me.
So in that sense AI could help. I mean it could actually do a website if you just ask it.
Specifically like the new generation. I might not want to use that website but it can. It can
give you a good start and then evolve it.

Follow up question:

since you are using AI for like writing and stuff, did it like kind of start generating
your text with your own touch?

Answer:

I don't know if which kind of AI is it exactly. I don't think so, no. So it's kind of the same?
Yeah, I think it's still the same generic, I don't think so. Actually I had another question, yes,
about the language, do you use it mainly let's say in Swedish or mainly in English or both?
Only Swedish, because I target the Swedish market, so for me Swedish is the only one that's
interesting at the moment So yeah, it's an American tool. So it's English is their native. So I
was concerned how good is their Swedish

Question:

Are you concerned that AI would take control of your market campaign and such?

Answer:

I Kind of like in their own way plays like against you in certain way No, I can't see how.
No.I had a company that helped me. He did AI-generated Google AdWords that I was paying
for. It's like the Amanda we were talking about, Amanda AI. It takes her, or whatever, a
while, I've understood anyway, for her to understand how to find your target. So what I found
when this guy was helping me with the Google Ads was he was using the wrong words. He
was like, well, that's not how you ask for a photographer. It was just a lot of wrong words
that I don't think people search for to find a photographer. So I think that's a waste of money
and time. Well, not time, but waste.
Follow up question:

So you always believe AI would always stay a tool, but not a threat in a way that it's
going to take over the entire operation or the entire process?

Answer:

Yeah, I don't think so. You need to oversee it. Or make sure that it doesn't take over. I'd like
to do a great job. I mean, it's kind of scary though, if you think about it, like a robot would
take over the entire thing.

Question:

What choice are you now making that AI could automate by relying on preserved data?

What choices are you now making that AI? So what choices am I now making that AI could
do? Could automate for me. Good question. Let me think. I guess they could help me with
linking more for blog posts. Because a blog post, to get a good SEO, you need to have lots of
links. And they can help me with internal links. Like this blog post is very similar to this one.
Oh, you should link them together. He can help me actually find my own blog post and say,
okay, link these ones. And rather than me finding them, and remember, oh, what have I
written about? Like, after a while, you've written a lot about a lot of similar stuff, but... So I
guess AI could help me with automating that, or just giving me the hints. Sure. That's a really
good point, actually. I never thought about it. Yes, maybe even finding other external links
would be great as well.

Appendix 2
Question

So tell us about yourself, your current Occupation and the time of experience in your
field?

Answer: I work primarily with e-commerce and customer service and market automation is
the same thing, depending on who you ask. I run a small company with 5 employees. We
work primarily with e-commerce, but we work with some tech companies too. What were the
other questions? Yeah, what's your role in the company? Role? I'm the CEO, founder.
Question:
Is your company currently using any AI tools, if yes, in what ways?
Answer: The only one we use is ChatGPT, but there are some other smaller random tools
that we use to automate things or to help us with prompts and stuff like that. But ChatGBT is
the one that's actually used pretty regularly. So sometimes you just ask it simple questions
and it will ask you questions. Sometimes you can use it to write copy or sometimes you can
use it to see if you have written a campaign that didn't perform well. You can ask AttributeD
to analyse it for you sometimes. Basically you can use anything you want with
ChallengeGPC. So if you have a question or a problem, it's like having an extra pair of eyes
in the same room as you. So it can be used for basically anything, actually, which is pretty
cool.
Follow-up Question:
Since when did you use AI in your company?

Answer: We were pretty quick to use it. So the first two, three weeks, which was like
playing around and tinkering and figuring out. Now, so obviously we started using it
systematically since January maybe.

Follow-up Question:
About the other tools that you already mentioned, you said like you are using a few
more, or what about those? Did you use those even before you had ChatGPT?

Answer: Yeah, but that's not really our expertise. Sometimes we'd be working with some
Google Ads agency who have some like price or something like that, and assess algorithms.
But there's thousands of those tools. They were claimed to do the same thing. But ChatGDP
for me was the first one that you actually use for real, and actually help you in everyday
things, and not niche scenarios, where it could be helpful sometimes, but not all the time.
Question:
Are you planning to kind of invest even more in AI in the long term?
Answer: Oh yeah, well the paid version of ChatGPT a month its like that, so whatever. But
the main thing is basically just trying to scale and leverage it. So it's not one of those things
where I look at it to replace something or someone, but rather a way to help it. It's like think,
why should you be a business analyst in the 1990s and you have to use Spreadsheets and
stuff then Microsoft Excel comes and just changes the game. I see it a bit like that It's just
like Microsoft Excel where if you learn to use it you can use then then it's great But like the
actual role isn't going anywhere So if we can find more use cases for it, then sure we'll do it
but I'm not like actively looking for the solutions for the problems we have. If that solution
happens to be AI, then great. But if it's another solution, then we'll explore that too.
Question:
Are you aware of any competitors or similar enterprises that have successfully
implemented AI? If so, do you see them having an advantage over other companies in
the market now with their way of using AI?

Answer: I would imagine that anyone worth their soul is using trans-GBT in some regards I
mean some of them some of them will be using it for really simple stuff without actually like
realizing the potential that it has I would be very surprised if there's anyone that calls
themselves a marketer or a senior marketer or a corporator that doesn't use it in some
capacity.
Question:
How did you go about selecting the right AI and technologies for your business needs?
And what were some of the main factors you considered?
Answer: So the only thing we actively chose was Shared Tivity. And the reason, well, it had
a lot of hype coming out on Twitter. And there's people using it basically every single day
with cool stuff that they're using because of it. And then I saw it for the first time and
I
realized pretty quickly that this can be used. So this is free so there's really no reason not to.
We don't really work with things that require an automatic AI pricing model or a competitive
thing, it's not that complicated. The way we use it is to come up with new marketing angles,
review copy, create copy, you can have a strategy too.
I mean, it's pretty fun. Because it's read like everything on the internet up to 2019, they have
to realize that like, in our space, things change pretty quickly. What's relevant in 2019 is
really relevant today. But it's still a really good starting point.
Follow up Question:
If there is an AI that is not free and is doing a much better job than ChatGPT, would
you consider adapting something like that?
Answer: If it solves the problem, then sure. So if ChatGPT is free and I'm making X money,
but I'm missing out on Y because of the limitation of the charity and there's another tool that
costs X times 2 that's giving me, that can help me, it becomes just a mathematical question
but it's not like, it's more like finding the problems that we have in our current business and
looking for solutions to them. If we don't have solutions for them and there's some tool that
could do it, then sure, like yeah, but I won't be actively looking for an AI thing. If it has a
purpose, then sure.
Question:
How has the integration of AI impacted your marketing team's workflow and processes,
and how have you addressed any potential challenges or concerns?

Answer: Well, I've told the team that if they're not using ChatGPT then they should be using
ChatGPT because like ChatGPT won't replace anyone however someone using ChatGPT
would definitely replace you. So there was some friction in the beginning because like all of
our copywriters for example they say oh a robot can't understand the angles that we can, they
can't understand the product the way we can, they can't have the relationship with the client
that we can etc etc etc and no it can't.
However, it's like turning down a free assistant. That's good for you. So I'm trying to
encourage them more to actually implement it more in their day-to-day things because it will
make them more productive, give them higher capacity. It's quite easy to work with too. So if
you have a normal client and you send them a piece of copy, the most common thing is that
you'll get back, oh, and the shares in Dazzle or Steam, and this isn't really what we were, I
just like it, but with ChatGPT, you don't really have that stuff, you can just, it just raises your
lowest level by far.
Question:
How have your customers responded to the use of AI in your digital marketing efforts,
and have you noticed any changes in customer behavior or engagement?

Answer: Well, since we're producing more and we're producing quicker, our customers are
happy. And our customers are in the same space as I am. They say, OK, Carl, this is a really
cool tool that you can use to help your business. Why aren't you using it? It's like borderline
negligent not to use it. Especially if I have clients paying me for a copy, they're paying me
for strategy, they're paying me for consultation, etc.
And they realize that I could be delivering X times 10, but I'm only delivering X because of
some, no reason whatsoever, then I can't really defend it. So our clients are happy, they
were.
We've had, we haven't even lost a single client yet because that wants to replace us with
ChatsGPT or use it themselves with ChatGPT so that's not being a problem either. So like
there wasn't any like trust issues basically raised whenever you let's say said or like talked
with your customers and you asked you told them that you're using AI in your like... Well it's
not really something that we tell, well we don't hide either but like we use Notion, we use
Pickup, we use Slack, we use... We can tell them how to text back if we want to, but it
doesn't matter.
They hire us to deliver results, and we deliver results however we see fit. If that's by hiring an
external consultant, or using ChatGPT, or embezzling a different tool, like, as long as you get
the job done, then the client's happy. Okay, can you provide us any examples of successful
AI-based digital marketing campaign that your company has? Probably, but we don't have
any AI-based campaigns. We have AI-assisted campaigns, I'm sure, and that would be
basically all of them.
Because all of them, in some regards, have used, I don't know all of them, but a lot of them
have at some point consulted ChatGPT at least once. But it's not like a foundation of, it's
more like something to get the ball rolling.
Question:
How do you ensure that your AI-based marketing efforts are ethical and respectful of
customer privacy?

Answer: Rare we will need to use a name or anything personal in charge of TV at least
sometimes we could upload a brand to charity but only the only way a brand can be uploaded
to charge you see is if they're big enough to be recognized by charity which means that
they're they're really they're already there on it so there's nothing there's nothing new that can
come from it there's nothing damaging I can go from it so I don't have, honestly, any ethical
risks. As long as you put in the right stuff, I don't have, honestly, any ethical risks or
concerns about trans-GPT.
Question:
What are the key challenges in your current digital marketing efforts that you believe
AI helped you solve?
Answer: It would be capacity. So now earlier a writer could realistically maybe write a
thousand, two thousand words a day. Try to achieve that just skyrockets. The second one
would be speed. So it used to take a writer two, three, four, five days to write a good text.
Now we can produce it maybe an hour. If that speed it would be quality too because you can
use it to proofread and edit your text. So what I normally do is because Swedish is my second
language, we work basically always in Swedish, so what I have to do is I have to write
something out in Swedish, then upload it to ChatGPT, tell it to fix all the grammatical errors
and stuff, and then it gets back to me. So I would say quality, speed, and output production.
Question:
How can using AI in your marketing especially help your company to adapt to the new
market.
Answer: Obviously, the other companies have already maybe adapted to ChatGPT anyway,
so what do you think? So this is a bit like where I touched on earlier when I said that
Chattopadhyay won't replace anyone. However, people using ChatGPT will definitely
replace you. So this is where it comes back to business. So who has better ideas, who has
the better
strategy, who has better processes, who has better routines, who has the best.
Because the thing is, say if you are a wizard with ChatGPT and you know all the right
prompts and you basically live in a system, however you can't lead your team, you can't
recruit your team, you can't manage cash flow, you can't handle a client relationship, all that
stuff, then you're gonna lose to someone who doesn't use ChatGPT anyway, who is good at
that stuff. So it's more like, I actually do like the comparison to Microsoft Excel, if you used
to work with numbers before, or data.And then, because Microsoft Excel, the real business
analysis or any of that stuff didn't disappear just as Microsoft Excel came. It just made
everyone who was really good at that job even better at their jobs. And people who couldn't
keep up, these were the people who wouldn't be able to keep up anyway. And it would be...
It's like the same thing. If you're going to marketing school, I assume, every two, three years,
Google has an update, Facebook has an update, SEO changes the algorithm, whatever, and
everything, the kids in the industry just get tanked by it every single time.
Whereas people have been around the world and they have their principles, they know how to
work, they actually know the game, then things come and go, trends come and go, and then
you just, you adapt all the time.
Question:
What ethical concerns or potential negative consequences do you think should be
considered when integrating into AI in digital marketing?
Answer: Well, it depends on how you use it unethically, then you should be fine. Because if
you use it in the way we're talking about, business or strategy or editing texts or producing
texts, there isn't even any risk involved.
However, if you want to use ChatGPT to ask someone to, I don't know, I don't know, I can't
even think of an example, but something unethical or something cheeky or something
naughty, then you can even Google it. It won't stop you. The information is there. Like,
ChatGPT is just... It's like if you look at Google as a library with all the information out
there, you can't find it yourself. We can just ask the librarian, hey librarian, where's the book
about this topic? The librarian says, oh, here it is. That's the same thing. GiantJPT is like a
librarian.
Question:
How do you envision AI-powered tools improving customer engagement and helping to
recruit new customers?
Answer: Well, the thing is with these AI tools, is like, ChatGPT is the first one to break into
the mainstream, like, for real, because of the price point and the use cases. But there are tons
of tools that recommend products, that recommend AV tests, that recommend price
matching, all that stuff that's automatic. Some of them are decent, some of them are bad. It
doesn't really... There's so many. Some companies, they'll build an AI tool that will help you
compare this to this or whatever or match this to this. And you're like, what problem does this
solve? Like, who cares? It'll be like you speak to some tech agency in Stockholm, and they
say that we've got that we've invested 5 million kroner into building this thing. And we've got
three clients and we're losing 8 million kroner a year because no one needs it because it's not
solving a problem.
It's just like it's just it just becomes a bit of a cool thing to have. So if there are a ChatGPT,
the way it solved problems for the customers or businesses worldwide, that was
revolutionary. It was game-changing. However, you develop a tool that compares a product
feed with AI logic. AI logic is a pretty banal thing to actually even, because like, what does it
even mean? Like, whatever. But so the problem is, so my point is, if you can use AI things
to
actually solve real problems and real, like, increase output, increase results, increase speed,
then sure, there'll be a ton of them coming in. However, up to now, all the AI tools have just
been pretty nice to have, they haven't been game-changing whatsoever.
Question:
Will incorporating AI in your marketing make it simpler to achieve your goals?
Answer: Simpler, sure. Simpler and easier. The only thing you have to be careful with is
relying on it completely blind, relying on it as a help, then sure. However, there are loads of
people who themselves, it comes back to the thing I was talking about earlier when it just
comes back down to business. So there are some people who, they're mediocre corporators,
they're mediocre salesmen, they're mediocre designers, whatever, and they think that
ChatGPT would help them get to the next level, and it won't, because all of the other
mediocre corporators, mediocre designers, mediocre whatever, all using transparency.
So it's not really... So yeah, so... like yeah, it makes marketing effort simpler, but at the end
of the day it comes back to you.
Question:
And do you trust AI to make decisions based on data that it collected, and would that
replace some of your employees that are currently doing that task?
Answer: Well, it depends on make decisions. Like would I, would I, suggestions, absolutely.
But decisions, no, because you can't be held accountable. So when a person makes a mistake,
I say, okay, what was your thought process here? Why did this happen? If it goes well, great.
If it doesn't go well, all right, what can we learn? However, with ChatGPT, well, someone's
actually using, like, delegating a decision to AI, for me, it just seems like madness. Unless it's
something really simple, like determining the win of an AI test or something like that.
But can you imagine, like, the board of a company saying, right, we lost $12 million this
quarter. Who's fault is it? Oh, yeah, ChatGPT? That just can't happen. But let's say in case
one of your employees kind of do this sort of a mistake where you lose a lot of money and
they kind of put it on AI and how do you think you will solve that kind of an issue? If
someone made a mistake like that and then blame AI instead of blame themselves or take
responsibility then we'd be having a very different just like AI wouldn't even be in the
discussion it'd be about like responsibility, ownership, things like that. It's madness. At the
end of the day, it might be good for you guys if you want to work in an agency or start your
own agency in the future. One of the reasons we get hired is to take the blame. So if we
produce really good results for our client, then the head of marketing of the client takes all
the credit because we found these guys, we hired them, we onboarded them.
It's my decision making, depending on what. If things don't go well, then it's not my fault. I
brought in these guys to do it for me, and they fucked up. And that's fine. And then we just,
we find them, and that's fine. It's just internal politics. And then if you try to play the internal
politics by hiding behind AI, even though someone else is making decisions, then I can't see
that being a particularly successful idea. And are you concerned that AI would take control of
your marketing campaign and such in the near future or like, let's say, in a decade or two?
I don't know. So again, like take control. If like this fucking Skynet came in and took control
of all my computers, then yeah, I'd be very concerned.
Follow-up Question:
But what if, let's say, if a company adapt an AI where the AI gives them all the tools
that you are providing right now, do you think that something might happen and does
that concern you? If a company would give them the tools, I understand the question.
I
think it means that a new tool like ChatGPT comes out that basically solves all the
problems that you solved in your company?
Answer: Oh yeah, sure. Well, would that be a concern of mine? No, not really. So, it would
have been a concern of mine in the beginning of my career. If you're a new corporator or a
new designer or a new whatever, then yeah, then you're in trouble. Because like I said, you
can be replaced by anyone else using the same tool on your level and you'll be gone.
However, on the level that we work with, it's much more decision-making and relationships
between ourselves and the clients. And like I said, when it comes to like the internal politics
of things.
So if they, because what we do is we protect our clients from, say something does go wrong,
it becomes our fault and then it becomes our problem to deal with. However, if that doesn't
happen, then the client won't have to worry about it. If AI would come and say... So, the
thing is, my area of email marketing, if you look at it in the 1990s, there was zero
automations, there were zero tools, you had to send every email yourself one by one by one.
And then the next tool came along, ASOL, and started sending to lists and newsletters.
The next tool came along that let you start having automations. The next tool came along and
started using new segments. The next tool came along and added advanced reports in. So at
every single step of this time, someone could have said, okay, I can start doing this myself
again. So I'm sure that there'd be some clients that would say, okay, Karl, I really needed you
to produce 100 newsletters a week, 100 newsletters a month for me, and now we can do that
ourselves, that's fine. However, there's just so many different levels to it because, yeah,
there's so many different levels to it. So I'm sure that it will be different.
It'll be damaging for you. So with the clients that we just, our clients that hire us, we have a
couple of hours here and there, like just really good to have. They do it because they could do
it themselves, but they can't really build people, they're just doing it themselves, they're sure.
Those clients could be going. But with the clients that want to have help with building a
strategy, building up a relationship, building actual, an ongoing partnership, then those guys
are fine.
Follow-up Question:
So basically, depending on the task that would kind of either give you more, like let's
say, opportunity to work, or like lose a client, depending what are you giving the
customer?

Answer: Kind of. Yeah, but it goes back to what I said earlier about the, no one using
ChatsGPT, ChatsGPT won't replace us at all. So when I was using ChatsGPT it will replace
us. Which means that if we are not learning how to use AI in our business, our strategies, our
campaigns, whatever, whatever, whatever, and we're just doing, we're just composing as
usual, make sure what's going on is going to replace us.
However, if we are using ChatGPT to take our strategies and ideas and create it to the next
level, then we're fine, because we're up there, and then ChatGPT brings up there, and then we
also use ChatGPT, so we're back up there again, in a dynamic. So it normally replaces if we
get lazy, or complacent, or drop the ball.
Question:
Yes, and what choices are you now making that AI could automate by relying on
preserved data?
Answer: So one thing we do is when we A, B test something, earlier you had to actually
analyze results so they B test yourself and make the decision on A or B. However, we have
tried, we have found automations with most tools. This is in the ChatGPT, this is in the email
tools. You can automate it so it says, if we get a result over this result then we can make this
the winner. And so that's pretty cool. But even, yeah, so that's something we can automate
and use. The automations overall, that's a completely different story than the AI stuff,
because automation stuff, that's a huge thing of what we do. So if you look on a website and
you look at a product for five minutes, then we want to send you an email with more
information about that product.
If you're buying a car, if you make your first purchase, if you subscribe to a newsletter, so. So
it's basically just trigger-based. If A happens, then we respond with B. And that's been
around for a long time.
Question:
What comes to mind when you think about AI?
Answer: I don't know, I don't think about stuff like that. The only thing I think about is cash
flow, monthly current revenue, and clients.
Follow-up Question:
So AI, what does it mean for you?

Answer: I don't know, artificial intelligence, that's what it means. I'm guessing it could mean
innovation, ideas, adaptation.
Question:
In a financial way, have you noticed that since you used AI, you gain much more
revenue or profit in a way?
Answer: It's a bit early, because we're gonna be using it for one quarter. This quarter's much
better than the last quarter. However, this because it goes back to business. So like I like is it
because of ChatGPT or is because we updated our pricing structure I probably because we
opted out of pricing structure, but tragedy it didn't hurt us at least yeah.

Appendix 3

Question:
So tell us about yourself, your current occupation and your field of work

Answer: Digital marketing, coding, yes. I'm in charge of the web-bio division of the
company and our own marketing and a little bit more. Okay, what is your role in the
company? Formally, I'm web-bio chief and marketing manager if we can use the long title.

Question:
What comes to mind when you think about AI?
Answer: It is an exciting thing to learn. It has been around for a long time, but has become
more used in recent years. But the new tools can become more public. It has become more
open.
Question:
When did you become interested in AI?
Answer: I have been doing this for the last few years, not more than that. I feel that this is
more a philosophical question, what is the borderline with the NoA in terms of regulated
marketing.

Follow-up Question:
Do you feel that your attitude towards AI has changed thanks to ChatGPT, or how was
your attitude towards AI before you had access to it?
Answer: Yes, it's become much easier. There is a web-limited version of everything. Instead
of having to code everything yourself and do queries and embass everything. I think that both
for programmers and the general public it has become much easier. To reach a specific target,
depending on what is calculated in the area. Maybe we have a good definition. Where do you
draw the line?
Question:
How much is your company using AI, and for what?

Answer: We use it, actually, daily. We have a lot of Google ads and we help our customers
to buy them. But they use Google in their own campaigns. It's either we who use it, or
Google, or the customer. It's the use. And we use both Jasper and ChatGPT to create basic
information on texts. I can't just click create and send it out on the web, but to get an idea of
what you can write about, a form of basis on a text, because we have a lot of production on
the web, but we can't find everything. Should you write about a choice of beer. Then you can
start with something.
Question:
Does your company have any plans to implement more AI tools in the near future, if
yes, what types of tools?

Answer: We keep an eye on what comes up and also change it. We have had Jasper before,
but now we have got Bing, so we can use it. It's always fun to try new tools, but they have to
provide something we don't have. If you don't have ten tools, you can do the same thing and
get confused.
Question:
Are you aware of any competitors or similar enterprises that have successfully
implemented similar types of AI as you? If so, do you see them having an advantage
over other companies in the market now?

Answer: I could think that all digital offices, whether they use it or not, use it. I don't think
there's a difference. I don't think that anyone... I think it's better to use the ones that are more
new and where the owners and all the employees are a bit younger. I think that the ones
between 20 and 35 years old are faster to get in there. Ok, so... I don't think that there is any
unique in any way, locally. And we have almost only local customers. And we have a few
customers in Stockholm, but most of the customers are local here.
Question:
How did you go about selecting the right AI tools and technologies for your business
needs, and what were some of the main factors you considered?
Answer: Google, of course. We have been working with Google Ads since 2006. We have
only been involved with their development. It's not that they care about. We have contact
with them and they give me a lot of ideas. They develop everything. And you, sorry, go on.
The other tool is more interesting to look at what is out there and use it to simplify. I don't
know if you used Google AI before to get to SEO. They had AI a few years ago, but then the
whole project was a failure. I was just wondering if you have used it. I use it for Google. I
use their tools mostly. Google Analytics. And ads. But you don't use any of the generator
SEO? Not from Google. Not from Google. No. I understand. I bought it from Jasper. Ok. I
think they are better. Or if it is from O2.
Question:
How did you go about selecting the right AI tools and technologies for your business
needs, and what were some of the main factors you considered?

Answer: I think it is a good question.

Question:
How has the integration of AI impacted your marketing team's workflow and processes,
and how have you addressed any potential challenges or concerns?

Answer: It's mostly for the people who work, or the people, I don't know how many, we are
four in the department, well five now, but I am, it's the two of us who work with marketing,
the people who work with development and production have not, but, I don't know, I'll talk
and think about it later. I think this is very limited with the market people that use it. The
most practical thing is to create content.
Question:
How have your customers responded to the use of AI in your digital marketing efforts,
and have you noticed any changes in customer behavior or engagement?

Answer: They just want good results. They don't care. They stick to the rules and do it well.
Which tools we use, I don't think they care. Some are interested in knowing how we work,
but most of them don't give a damn. more part of it. I don't think so. Everyone does it. Then
no one can choose. Then they have to do everything themselves. Or how? I think it's more
about which Google Ads you use and what kind of campaigns you can use. But there is more.
You can have e-commerce, which is a lot more. You can have performance, which is very
good.
Question:
Can you provide any examples of successful AI-based digital marketing campaigns that
your company has executed, and what lessons did you learn from those experiences?

Answer: No, not at all. But performance max on Google Ads is very much you give a data
and do something and don't care what it does either, which you have worked with for a long
time. I'm not completely comfortable with that I have no control of what it does. We always
get help with this, we send in 100 000 in the campaign, we can measure, we get 100 000
clicks, we get 7000 orders, it generates a lot of money back. We can measure that, but the
intermediate step is what did these 5 000 people get through the and the other 15, we have no
idea. If you belong to a large target group, if you want to be a person who is 30 and lives in
Vasteras, then you follow that, but if you have the whole country, then it becomes a bit of a
mess. But that is the advantage, you just get to let go, control, that the result is ultimately
important. Part of a re-traded deal is how long it stays in the market
Question:
Are you currently using AI with your CRM system? If yes, what kind of AI-CRM do
you use, and for what purpose?

Answer: For our part, we do not use that much of our own customer data. Some customers
do it more, load up lists and so on. We have chosen, for our own market, to keep it very
limited. Because it's a part of the identity of the customers, and as a broad operator, we know
more than many others what's happening, and then we really keep and what is happening and
we are doing all kinds of remarketing to avoid discussions.
Question:
When it comes to the financial perspective, have you seen any type of changes gaining
more revenue or profit, after implementing AI into your strategy?

Answer: We have more e-commerce customers than we have bigger customers. And that can
partly be since the performance has worked well for the category. It's not something we did
ourselves, it's more a development on Google that they have made a better campaign. It has
affected our efficiency because it made it faster for us to analyze, giving results faster.
Question:
Do you see AI as a part of ROI, or an unnecessary and disadvantageous tool?

Answer: Everything we can do; we will save money on. More customers will save time and
save personnel. That is a big advantage. And it will get better. Many things. You can think
you are good, but you will never beat a Google data store on the logic. I think that's what
Google's data all is about. But would you change, like, to put a lot of effort into your digital
marketing and like, not hire people to do the work?

For example, to generate pictures, videos, SEO, you know, that kind of stuff, IPO? Yes, I
think we can put more value on what we do today. But in the end it is about the number of
customers that come in and the size of them. Because it requires a bigger project than it
should be. I can't handle sitting with 1500 SEK in my pocket. It's almost not worth it to start.
Then I have to throw up something simple and fast.
Question:
What are the key challenges in your current digital marketing efforts that you believe
AI help you solve?

Answer: It's a lot of help to fix or find a solution to it. It's a lot of work. That's why I've
decided to create content. The basis of the content. It's faster to get ideas of what you can
write with. And to create a basis of the text. That's a huge difference. Now, the tool has been
around for a long time, to generate text, and even Google Translate, but it has been quite
worthless before. But now it works well, because now you can take a text and then you
always have to rewrite it a little bit, but still a very good, basic and easy to use.
So that's the big difference. And then of course all the... all the forms of remarketing, both via
Facebook and Google. Which at the time didn't really work. Maybe remarketing is a
calculation, but in general it's still to... to reach the number of visitors that have been there.
I'm not a visitor, but I can reach it.

Question:
What ethical concerns or potential negative consequences do you think should be
considered when integrating AI into digital marketing efforts in your companies?

Answer: Part of it is that you have to keep track of the upload rate.
So that you get users with pictures and texts. They are pretty good at retrieving content from
other websites and other users can check if it is really unique content. So that they don't just
get the competitor's text and publish it. So, it takes a little bit of knowledge. And the ethical
part is how much data you should use about the customers. And we have chosen to use very
little. If you compare it to our market, which is quite stable, we have a lot of subscribers. We
don't have that much recurring sales for our customers. Then you don't need to use that much
data. But with the customers they buy quite often. This gives you a much better reason to
feed in data to AI to be able to direct the message. If you have an e-commerce with 40
different products, you want every customer to get information about the products that are
relevant. And then you have to send in a lot more data into the system. For our part, if you
have bought a broadband, you don't need another broadband. It is more difficult then, when
we go to the customer, because the more products you have, the more you can go to the
customer and sell to the same customer, the better. that I'm a customer, comfortable.
How much can you expect? You get it every time. But this is what I've written about, or this
is what I've thought about. In the end, someone will feel monitored. I also feel that with
remarketing. I've been in an e-commerce, I've looked at Bodojo, and I get them quite as many
times. You have to weigh this. How much can you chase a customer? Quite a lot of times.
That's why you have to weigh how much you can chase a customer before it gets
uncomfortable. And it's not an easy question. You don't want to sell, but in the end you turn it
over to the store owner. They never want to sell.
Question:
How have your customers responded to the use of AI in your digital marketing efforts,
and have you noticed any changes in customer behavior or engagement?

Answer: We will work more with Marketing Automation, which is in our control, to make
sure that we have a lot of different target groups. We sell everything from home telephony to
retirees in the north, to customers who want to rent our data center. It's hard to find what
would communicate to the right people.
Question:
Will incorporating AI in your marketing make it simpler to achieve your goals?

Answer: Yes, it does. By... how can I say it... It's faster to do it with a job consultant. And we
can do more tasks with the same staff.
Follow-up Question:
So you believe in AI? You have no problem with AI? Like, to do your job the right way
or other complications?
Answer: No, I don't think so. We won't get there in the future. At least not in my lifetime. I
have AI taken over everything. We have A-tags everywhere, we do nothing. It will always be
needed. Many of us will not be able to get over it. Some will need the coder from the
beginning. Many of us will never be able to get over it. Everything that is practical, you will
never get it. Start mounting lamps or building houses. You will never get there. It will be
pure routine work, from bookkeeping to good management, the customer service, they will
disappear. But that's what the statistics say, that everything from routine to transport and
such, will disappear totally, such jobs will disappear.
Yes, there will be more marketing and development. I do that every day. That's true.
Follow-up Question:
But with that in mind, do you think that making a decision when it's backed up by so
much data is a good idea?
Answer: We have to evaluate what decision we make and not give it a free hand. We have to
keep an eye on what results we get. It's important to keep an eye on what's happening and not
how much of what we've done. As long as we have the supply, as long as we have the supply,
as long as we increase the purchase of that target, as long as the target is fulfilled, as long as
you have control over that, then you can probably get it wrong very quickly. Ok, so the last
question.

Question:
What choices are you now making that AI could automate for you by relying on
preserved data?

Answer: Automated decisions. I'm in need of this margin, I can press play and shoot it out on
the channels where it works best.

Appendix 4

Question
So is your company as of right now using any AI tools that help you be more effective or
so?

We use right now the only thing we change we use open AI APIs and work upon them so we
use them to build products basically. So we have a couple of products running on OpenAI.
Then we have our own AI that's reading material. And basically, we have created automatic
media monitoring for the customers. We have created finding the right journalists for our
customers so depending on what you write RAI can actually find which journalist that you
should send this out to or talk to. Just as a help on the way.
Are you aware of any competitors or similar enterprises that have successfully
implemented AI as you? If so, do you see them having an advantage over other
companies in the market?

In the text writing or text generative AI market, I would say that Jasper has an advantage
because they were first. So the advantage isn't really in the product. Well, they have a really
good product, but product-wise, the other ones are catching up mostly because the latest GPT
version does most of the work. So working on GPT-3, which Jasper has been doing for a
while, and then moving to 3.5 or 4. If you move to 4, you suddenly get tools and an AI that's
so powerful that all that competitive strength is just lost. They basically, they have the
marketing, they have won the marketing war, but I think copy.io, Tekta, are coming up with
good stuff. So it's a lot of marketing tools coming up, or writing tools basically.

Do you see yourself having a competitive advantage over your competitors?

Yeah, we're in a little bit of a different space, since we're working with press releases and
have the PR take and not the marketing take. So since we have the whole flow, we have the
creation process, we have the distribution process, we have this networking process with
journalists and also the media monitoring. In the Nordic countries, right now we are quite
alone on the strength and the creation product that we have. And also finding the journalist,
the right journalist. So, but worldwide there are three to four competitors coming up, and
especially in the US, they are getting a little bit more buzz around them. Although, I won't
say that our features are better than all of them, but it's better than most of them.

Has your company any plans of implementing more AI tools in the near future?

Well, we set AI strategy a couple of years ago where the whole company is actually being
pushed forward and all the products are based on AI. That was a vision we had a couple of
years ago We're definitely not there yet but we start to look at more to see how can we can
drive this forward and Build our products instead of looking how we can build products and
AI again, so we're taking an AI first approach right now to a product

Are you currently using any AI with your CRM system?

We're not using it in contact. Like, if you have added a contact, it's there. We don't use AI to
enrich it or do anything. We do it in connection with the CRM tool. So if you're going to send
something to a press release out or a marketing tool or something, we'll give you tips on
when the best time is to publish to the contacts in the CRM. Also, as I said, we're finding the
right contacts to add. And since in our CRM system, the main contact is the journalist. So
we're finding them. But in the actual CRM system, we're not using any AI to sort, to filter.
We have some smartness in it, but it's not.

When it comes to the financial perspective, have you seen a type of changes, gaining
more revenue or profit after implementing AI tools into your strategy?

like the tech-generated AI is too early to say. We're building, since we have a quite big
product, and I think it's about, I would say 20% of it is AI-driven, and it's quite a chill to
build in the last six or ten years. So, and our pricing model isn't based on features, it's based
on packaging.
So it's hard to see, but I believe we have been used to the AI a lot in selling it, like the
presentation of the product. So I think it is revenue, it has driven revenue, but I can't point at
any certain feature that we've implemented.

Do you see it as more of a long-run revenue or more than a short-term revenue?

I see the... I see the... what we started a long time ago, I see it as a long-term revenue. I see...
we don't know how the AI that comes out right now is going to develop. but you go for the
long term., financially wise right now, it's hard to see.

What are the key challenges in your current digital marketing effort that you believe AI
helped you to solve?

generating, to gain speed basically in putting drops out there. To always keep all the channels
updated, you need to keep, you need to create so much material. What AI does, it actually
helps us to scale up without scaling people. So you can create more material for not only one
channel. It's easier to actually broaden your channel strategy. So it's hard with the social
channel that comes up, but now one message can be divided quickly into several at one point.
I think we're using it quite a lot in the draft mode to see if something is, to see if a topic we
come up with, instead of writing a draft and evaluating it, we can create it and then evaluate it
quickly and then say it's a go or no go and then put hands on it or not.

How could using AI in your marketing especially help your company to adapt to the
new market?

Yeah. Yeah, that's, yeah definitely. Like, we are one of, I think, I would say if we look, if you
look at the company, we have about 150 employees and I could probably choose any
company that's below 1000 employees and say that we have the most people on development,
on the dev department that's looking at AI right now. So I think we need to keep the latest
technology and we've been doing it for quite some time. So I think we have a routine to not
miss anything. And we're covering different channels, we're covering AI within development,
within marketing and other places.

What ethical concerns, and potential negative consequences do you think should be
considered when integrating AI into digital marketing efforts for your company?

Well, it's the clear ones, you know the fact checking. So always do the fact checking. Always
look at the hallucination that the AI is generating on text. It's basically the same thing that
you always do if you let somebody else write something for you. Or choose who to send it
out to, you basically do the check and then you hope that it will go faster than you have to
write it yourself. So those are the ones. Ethical to have somebody writing it for you or
looking at, you can also look at the sustainability of it. It's good to use AI, but also the people
situation. Should you fire people in an organization because you have AI? We don't see it that
way right now. We see that we have a good amount of people and the AI will support us in
growing instead of looking at the basic.

How do you envision AI-powered tools improving customer engagement and helping
your company to recruit new customers?
 Well, we're using it. If we look at the message around our competitors and our marketing
message, It's totally different. We try to be, in our business, in our field of business, we try to
be the ones that are in the forefront of AI. This is what we try to communicate, everything
from the website into all the material we create. And also, for example, on individual
LinkedIn posts. So it's not a coincidence that I write about AI.

It's not because only me, it's not a personal growth for me, well it is in some sense, but it also
is encouraged by the company for me to show what I do within AI and what the company
does.

Will incorporating AI in your marketing make it simpler to achieve your goals?

Marketing goals?

Hopefully. I think it's too early to say actually. But I think it will. I'm pretty sure it will.
We're already using it in a sense. And it's, you know, small things that sometimes take time is
just write it up and then send it on. So it's all everything from internal documents to external
that we're going to send out so we're using out of the things we use it's mostly generative AI
technology.

So the next I think you already touched on this one do you trust AI to make decisions
based on data that it's collected and that would be replace some of your employees that
are currently doing that task now?

Some decisions that they are making now? Right now I don't think so. Ask me again in six
months, maybe. I don't think that... and that this is my belief now, it can change. I don't
believe that ChatGPT now could replace any person in a company. But I believe that if
somebody puts a layer, and with that I mean builds a tool, and bring in other data to
ChatGPT. So you make a product where you, ChatGPT is the engine with information from
somewhere else. And whatever that creates will be, yes, then I think it could replace people.
It will be more updated to the latest data, and it would be sort of like putting expert
information into ChatGPT.

Are you worried about losing control, for example, campaigns like Google Ads and
being very dependent on AI?

Not right now again that's a it's a it's a good question there is no there's no fear there should
be more fear in organization that it is but we also we also meet everything and if you don't
you shouldn't work here. If you don't meet everything as a challenge and something that
you're going to try, then you're not a manuscript fit. So, yes, we maybe should meet it with
more fear, but right now we have the competence to not be stuck with AI in case of we don't
know how the privacy or whatever happened in different regions especially in Europe and
GDPR comes in etc. So we're not stuck to AI but we're not scared of it.

What choices are you now making that AI could automate for you by relying on
preserved data?

We're still writing too much material by hand. So we don't really trust AI to get the tonality
right. And that is a bit prompt engineering. You could absolutely get the tonality right, but
you need to do a little bit of work yourself. So I think that's where different tools come in,
because ChatGPT doesn't really support, and it doesn't give you help. It's too broad to
give
you help in a specific area. So if you want to write a marketing post or a Facebook post or
Snapchat post or whatever, you can't choose your tonality that easy.

You have described that you mostly use AI because it helped you to create more
material. Do you se any value for example time efficiency and productivity and so on?

Well, all the AI that I use it is a bit of a start.But it also saved me a lot of time doing analysis
on different companies. If we go to the use cases, for example. So we can feed in and we can
create a prompt and we can create use case from customers, how they can use a product
within seconds, as long as we have the right prompt, we can just feed it in. And we can get an
output that we can use for sale or for marketing, great success cases, or something like that.
So we can use it to analyze. We can build it. Myself, I use it a little bit to test my strategies
for different features.
I put it in, and then I cross-check it. I basically create a persona of quality assure that can
actually check. And I write, be hard on these things and then they can check everything I do.
So it's like having an extra colleague checking my work, basically.

Appendix 5
Question:
Can you briefly introduce yourself?
Answer:
My name is …. and I currently work as a CMO, so Chief Marketing Officer for a
consultancy-based company that works with AI solutions and intelligent automation. We try
to do client specific solutions when it comes to automation with either just AI or AI in
combination with software robotics called RPA. So robotics process automation. And I have
a background within IT for a long time. I studied myself at university and I've been the CMO
of my company, Substorm, since almost two years now. So. I'm trying to.

Question:
What do you know about AI?
Answer:
So since we work a lot with AI solutions, I'm quite knowledgeable, I would say myself. And
I would say that it's like the most interesting topic right now when it comes to tech and the
development within both the regular field of everything when it comes to our communities
and everything, but also when you see how people react with tech and AI solutions overall. of
valuable insight and a lot of valuable new ways of working, both for regular people as
ourselves, but also for companies and people within all kinds of departments within a
company as well.
Question:
when din you become interested in AI?
Answer:
Yes, so I've been interested in automation overall for as long as I can remember because my
father has been in automation all my life. So it was introduced quite early in the whole kind
of way of thinking when it comes to automating different things in life and when it comes to
bringing technology into your all-day life. So I've been into automation for quite a while but I
didn't really come into AI up until 2019, somewhere there in the autumn or winter of 2019.
So I started to automate a lot of my way of working when I worked as a consultant myself
and then I came across the CEO of the company that I work for today and got introduced to
the whole AI solutions and how you work with AI solutions through her, more or less.
Question:

Is your company currently using any AI tools?

Answer:
Absolutely. I would say as a marketing professional myself, I use AI a lot when it comes to
content creation, which is probably the most hyped part of AI solutions within marketing
right now. I guess that you've heard a lot about the generative AI field overall when it comes
to the chat GPT and you have mid journey and a lot of different kinds of AI tools to actually,
like, make it easier to create new things. So I use CHATGPT i and other kinds of AI tools in
my day-to-day work to help with, especially when there's something called the blank page
syndrome. So I'm not a super creative person myself, but I'm also, like, I'm usually like
editing things that maybe a copywriter or an art director have done for me, but I can now like
create something through chatGPT or other tools and do it myself and more efficiently as
well.
We're actually, yeah, we haven't really got into using some tools into our own like marketing,
but there's a lot of tools in the field overall right now. I think that most of the, what you call,
kundsegmentering, client segmentation overall. And there's also a lot within predictive
analytics when it comes to seeing what kind of trends that will come up within marketing.

You can also use it in chatbots and that kind of things when it comes to more of a client
engagement thing. But as probably you all know as well, you can use it in like personal
recommendations, trying to get your portfolio or your, what you offer more direct to the
typical client you use. But you know, it's in Spotify and like news outlets as well so it's
everywhere. Yeah that's just a few. I really want it.

Question:
Has your company any plans to implement more AI tools in the near future?

Answer:
So I've been looking into different kinds of tools that we can use, but since my company is
still kind of a startup, that we're kind of early or we're quite small, we have some limitations
when it comes to resources. So I think that bigger companies with more resources when it
comes to actually putting money into marketing will probably adopt AI tools quicker. But
I'm
keeping an eye on it and I'm hoping to use something in the near future But all tools cost, so
you need to be cost effective as well.

That totally depends on what kind of need you have, I would say. So if you have a need for
content creation or if you want to do more predictive analytics or if you want to do improved
reporting and kind of measuring the data that you get from your campaigns or how your
website is like working overall. It totally depends on what kind of need you have. Yeah, so I
think you need to focus on the need part before or when you're trying to choose what kind of
tool you want. For me, it has been like resource effective. It's been quite cheap, more or less,
or more or less free. So that has been the main factor for our company since we're, as I said,
still small.

We don't really have the allocated resources for using all the tools. So I needed to be mindful
of what kind of tools we can use that are cheap.

Question:
How could using AI in your marketing specifically help your company to adapt to the
new market?

Answer:
So the team is only me. I'm alone. But it has definitely made my work more effective. It's
given me a lot more time to actually do more qualitative stuff. focus more on segmentation or
reporting or analytics of our whole content creation overall. I think that everyone has to do
that, especially with these tools being quite new. As long as you're very mindful of what kind
of data, what kind of information that you put through the tools, you need to be really
mindful of not stepping on any boundaries. They are quite... you don't really use data in the
way of maybe like going into the field of GDPR and doing those kinds of stuff. They're quite
basic right now, I would say.

Or a lot of them use it like if you have Google Analytics and like applying it on that instead.
instead but it's always your own data that you use. I don't think that anyone really noticed,
since, like, especially content creation is quite diverse, so you can do it in many different
ways. And I always try, like, I use it with myself as a human in the loop when it comes to
using the solutions. So I'm always trying to edit it so it sounds like the same kind of tone of
voice that we usually have for our brand. So I don't think that anyone noticed, but I would
say that the increased level of engagement or the increased level of posts that I've put through
have given more of an engagement when it comes to our social media channels at least.

Question:

What ethical concerns or potential negative consequences do you think should be

considered when integrating AI into digital marketing efforts in your companies?

Answer:
So I see an increase in engagement overall just because we're actually putting more out there.
We haven't really done any digital AI marketing campaigns, unfortunately. So we haven't
really gotten there since we're such a small company still. So we haven't really worked with
campaigns in the traditional marketing way that you usually do when you have more
resources, unfortunately. So I never use customer information in it. That's probably the first
and the easiest way to ensure that it's both ethical and that it's not compromising any client
data or anything. cutting that out of the equation, you're mostly safe. No, we're not.
Question:
When it comes to the financial perspective, have you seen any type of changes gaining
more revenue or profit, after implementing AI into your strategy?
Answer:
Not really any direct revenue, I would say, but you can always see when you get more
engagement, you can get more of leads when it comes to sales. And since we're a consultancy
company, we're always selling services instead of products. So that is kind of a direct
dependency, even though we haven't seen any direct revenues that are connected to the
marketing parts. But an increased level of leads will increase the level of sales, which on the
bottom line, will increase the revenue as well. Absolutely,
Follow up question:
Do you see any increase in productivity and efficiency since you started using
AI? Answer:
Yeah. Yeah, and I can add that probably have made a difference when it comes to having
more resources working with marketing in the organization. Like I can do most of the work
myself now since I have these kinds of tools that help me in my work. So we don't have to
have another person or a content creator that is assigned to the company.
Question:

Do you see AI as a part of ROI, or an unnecessary and disadvantageous tool?

Answer:
It's definitely a return on investment part, even though we don't really have any more tools
than what I'm using for content creation. But I would say that all growing companies that use
AI need to see it as a return of investment. Because if they don't, they will fall behind. I think
that if we would go more with using tools or focusing a lot more on SEO and digital
campaigns and like working more with, how to say it, like campaign words or SEO in
overall, it would probably help us be more cost-effective actually, because there are a lot of
like AI tools that can help you optimize how and when you buy different of keywords and
make it more cost effective depending on the time of the day or the inflow of people actually
buying those kinds of keywords at the exact same time.

Question:
Are you aware of any competitors or similar enterprises that have successfully
implemented AI? If so, do you see them having an advantage over other companies in
the market now?
Answer:

I would say that there aren't so many companies that have actually implemented AI yet. It's
really on the increase, of course, but I don't see that there's a lot of companies that we
compete with that actually do that in a broader spectrum right now. So I would say that if you
start doing it now, you have actually more of a, you have kind of the opportunity to get a
little bit forward instead of lagging behind.
So like the optimization and automation of a lot of things within marketing is already there,
but it needs to speed up and it needs to be used more. So I would say that the key part that
could increase is the adoption of it, which will increase efficiency and the economical part of
being more, you get a higher level of ROI in your marketing initiatives.So everyone needs to
adapt, that will actually make the development of the tools even faster as well.

Question:

Do you trust AI to make decisions based on data that it collected and would that replace
some of your employees that are currently doing that task now?

Answer:
Yes, definitely. I would say that I am trusted to make decisions, but not to act on the
decisions before you have had a human in the loop to look at them. I'm not... I don't really
SEE IT as a problem. I'm not afraid of it, definitely not, because the AI solutions right now,
they aren't too smart, so to say. I often do lectures when it comes to AI and I always try to
focus on the fact that AI is just advanced statistics at the base level, so it will never be like a
Terminator taking over and running everything. And when it comes to, I would say it's of
course a real problem with people not having as much work. But on the other hand, that
development will also create new work.

Follow up question:
What type of tasks would you trust it
with? Answer:
So it's the same with industrialization. It will, some parts will disappear and some parts will
reemerge. So it's a give and take. I would say that if we start working with more optimization
of SEO, that could be one thing. I also would trust an AI solution to go deeper into doing like
content creation, especially when it comes to video and moving pictures content. And just
like doing bunches of that kind of thing, as long as you feed the tool, the right like a brand
book, and the data that we actually have created during the years that we've been active.
Appendix 6
Question:
If we start with the questions, tell us about yourself and your current occupation, your
role in the company and which field of work do you have?
Answer:
I am an engineer by training, which is a little bit unusual in my business. But I'm also a
trained co-provider from Berks Corp of Communications. I have a little bit of both sides
there. And I started an agency in 2009 that was sold last year. I actually kept the name, I still
have the brand name, and even the company I sold what the assets and liabilities of the
company, the insides of the company. So I was running an advertising agency for 13 years. I
thought I was not going to go back doing that and was working with two startups over the
course of last year, one in MedTech and one in PropTech. And then almost by accident, I
slipped back into branding again because there was this project that I just thought was too fun
not to do. So I did it and realized I could use a lot of these AI tools, not so much for final
work but for the whole process up until that point.
I'm sure it's gonna get to a point where we do the final live work as well, but so far, I haven't
done that. But so what ended up happening was that I suddenly felt I had a lot of high
capacity, almost to the point where, at the biggest point, I think my agency was like 35 or 40
people. It's almost that feeling of being able to do that amount of work, not only with one
person and robots basically then.
So, and of course, I'm fully flexible then. With pricing, I can set prices that are way below
what anybody else, or what I myself charge for if I want, or I can charge more. But that
flexibility was amazing too. So suddenly it was felt really creative, and business-wise it felt
very attractive. And it was just really fun again. So I'm back to doing brands again, but I'm
running my agency myself.
I have a studio at home, so I can do photography, film, and music at home in my house. I
have a separate floor for that. The combination of having that productive flexibility and
having the AI tools, that has become my new thing, you could say. It's a really great, really
fun thing to do. So that's what I'm doing right now. I think if you look back at the works that
are most famous for it is probably the transition we did with Halo in 2010 to 2016. That was
the first thing that put us on the map. The second thing I would say was Lico.
That became, in terms of numbers, the most effective ad campaign or branding campaign I've
ever seen in my career, at least. The only one competing with it would be Evopt, but it was
better than that.

Question:
So is your company currently using AI tools, and in what way are you using AI tools?
Answer:
Basically it replaced a lot of the staff. I mean, the capacity of working that I have now, I
would say is comparable to at least when we were 30 people, maybe more.So, and one of the
reasons for that is that it's not a linear progression. So when you're 40 people at an agency,
you have maybe, I don't know, maybe say 10 creatives. And then you have a ton of admin
and overhead. And there's people making sure that everything runs smoothly, and that the
meetings are booked on time, and deadlines, and everything around it. So that means that,
you know, if I can have a lot of things in my head and use these AI tools, a lot of the need for
that coordination disappears. So the coordination costs disappear and you don't need people
for that either. So that's kind of an interesting aspect of it that I didn't see coming really. I
didn't realize that was gonna happen and I've been throughout my whole career and this is a
weird thing maybe but throughout my whole career I've kind of been almost hiding that I do
many different things right so I'm an engineer I have worked as an art director I worked as a
designer I'm a trained copywriter and worked as a photographer I'm a videographer I do I
mean I make music and say all the instruments I sing you know I have all these things that
I've been doing but always I've always been hiding it I've never been I never wanted to talk
about that because it made me not credible in any of the fields so I just start to say hey I do
strategy and that's it.
You know, I do a little bit of creative, but mainly strategy. That's what I said, or used to say.
And for credibility. And now that I have all these tools, it's suddenly a huge competitive
advantage to have all those things in one head. Because then you can prompt AI in different
ways with all that knowledge in one hand, having all those different fields in your head, then
you can prompt as a photographer with all the knowledge about cameras and lenses, but also
with tonality, with all your kind of knowledge of different writing styles, and then of course
there's a technological part to it, you being able to use APIs, so chatGPT for example, or
technical aspects.
So when you can bring all that together in one head, you don't have any need to have a
meeting, just to kind of agree with people on how to prompt or anything like that. So
suddenly it became a huge competitive advantage, and that's the first time in my career, it's
the first time in 20 years I've felt that way. So that's kind of how I work with it. Well,
actually I didn't answer how I work with it.
I will answer that too. So, if you don't have any questions on what I just said, I can ask the
second part too. No, just keep going, how you,

Question:
in what way are you using AI?
Answer:
Okay, cool. So in practice, when I work with creatives, and I think this is true for most, for all
agencies probably, that you work kind of like, if you're a creative director, you creatives to
You breathe them and they come up with ideas And it's not super common at least in my
Experience that you just use the ideas as they are you can listen to them and you bounce them
around Feed off of each other's creativity and create new things So you have adding
perspective to each other's thoughts.
And AI is really good at doing exactly that also. So if you would ask Chat-GPT for example,
to just create a campaign, it will not be very impressive, you know, yet at least. But if you do
it and you also throw in like a high chaos parameter, a high style parameter for example, you
wont get the result that you were expecting. And the new stuff feeds into my creativity, or
what used to be the team's creativity but is now mine. And that way I can kind of use it the
same way as I would work with a creative team.
That you're bouncing around ideas. And it doesn't matter if the idea is not perfect or usable
even, because that's not what you're trying to get. You're trying to get perspectives and just
getting pushed out of your comfort zone. Basically feeding your creativity. And that I think,
Chat-GPT for example is doing really well. And suddenly you can do that without having,
I
don't want to sound like I'm talking down on humans, but humans have human problems.
Humans have problems of having a bad day or having their boyfriend or girlfriend break up
with them or having alcohol problems or gambling problems. Humans have human problems
and Chat GPT doesn't and never gets tired and never gets done for the day and it's always
awake you know in the middle of the night to come up with something.
So you have all those advantages as a creative partner and there's also like this infinite
capacity. It's like having a hundred credits if you want so, and it's fast too. So all these things
together make for a super high capacity to come up with and create and work with ideas. So
that I think is one part of it. And the second part of it is about images. It will be about video
in the not too distant future. And the way I use that now is not for final images, even though
some are so good that they could be final images, but you have these rights, question marks
about intellectual property, still, even though that's gonna change with Firefly, for example. I
don't know if you're familiar with Firefly or Ultimate. Firefly is Adobe's version of
Mid-Journey, basically.
It's going to have a lot more creative control, but the most important part of it, I think, is
going to be that it's been trained on Adobe's own image library that they own the rights to and
by extension I own the rights to or I have the right to use if I'm paying for their image library
which I am so that means the derivatives of that will also be okay to use so then you don't
have any rights issues if you use Firefly compared to mid-yearly that's just using potentially
copyrighted material. I think past that, right? So, but anyway, but I'm not using it for final
work anyway. I'm, I am using it to quickly sketch up ideas. So I have these ideas, I want to
test them out, I just sketch them up, I get back good enough things to be able to show to a
client and discuss. And that way, we don't, I've always kind of done that, but with like very
kind of fast sketches, because I think that it's really important that you have the same picture
in your head as the client does.
If you don't show sketches, you will 100% certainly have two different views, different
images in your head, you and the client. So that's why I've always been doing a lot of
sketching, always, always, and that's why I have my own studio and everything else. So, but
now suddenly I can sketch up things that are a location or that are, you know, just weird
things that don't exist and put people in strange contexts and do things that used to be pretty
hard to sketch up in a good way, I can now do super fast.
And same thing there, if I throw in a high chaos parameter, or a high style parameter, I get
back weird shit, and sometimes that will spark new ideas. Just like when you work with a
designer, and the designer comes up with something, and somebody's like, oh, I didn't think
of that perspective, and now we have a new idea. And that's kind of how AI does it as well.
So those two things, using AI as a creative partner, and the second thing, using it to sketch
things very fast, and in a very nice way, those two things are my main usages.
I know other people use the book to write long chunks of copy and stuff. I don't really do that
so much. I enjoy writing my own things. And I don't really, I just don't believe what I do.
So, I would say that's my two main ways of using it.

Question:
Okay. We have some more. So, are you aware of any competitors or similar enterprises
that have successfully implemented AI and if so, do you see them having an advantage
in the market?
Answer:
It's hard to say Everybody's talking about everybody See this mean there's a mean that's been
floating around. It's like I can't remember exactly. It's like a fast dog and there's like a cheetah
and it's just like fast, faster.
And then the fastest is people becoming experts in AI. That's like the fastest animal on the
planet. And it's so true, everybody's trying to be so optimistic about this and it's the exact
same thing that happened with the internet in the late 90s or web two, mid 2000s and when
apps came, I mean people were just trying to take advantage of the technology and it was like
this kind of an opportunistic thing. So you don't really know how good people actually are
because everybody's trying to sound like an expert. And I think it's similar, it's similar to like
you know when digital cameras started appearing, in the beginning it was like a USP to say,
hey, I'm a digital photographer.
And then pretty quickly, that was like, yeah, of course, you are, you're back to just being
about the art, I think that certainly there has been a lot of work going into it. I think that it's
an important kind of parallel thing that the economy is bad. So people are kind of firing a lot
of people, but it's a little bit hard to say if that's about AI or if it's just the general economy.
But a lot of companies I think are panicking a little bit about it.

question:
Is there a competitive advantage?
Answer:
: I mean, obviously I can do work now way, cheaper than I used to do. I mean, we are talking
like a third of the price. And at the same time, I'm making six times as much. So if you're
going to try to compete with me and you have a ton of staff, well then you really can't do that
because you're gonna be working at a loss. Whatever we charged a million pounds for before,
we used to have 30% margin on that, and that was extremely high for the industry. And now
I'm charging like 300,000 for the same thing.
I mean, of course, that's gonna be very hard to compete with. But then, it's not only about AI,
it's about like how well can you slim down your organization as a result of AI, it's a little bit
of a hard thing to answer because I think at the end of the day, the technology itself is
available to everybody, and everybody will be using it, and it won't really be a competitive
advantage. There's like a digital camera, was not a competitive advantage pretty quickly, and
it just came back to being about having the best vision and the best creative ability and all
that stuff. So I think this is the same. There's gonna be a short period of time now in the
beginning where people can take advantage of being a first mover, but that's gonna even out
very fast, I think. But then you also have this residual thing for a while, maybe six months in
Sweden or something where people are gonna try to rebuild their organizations. And that
kind of residual inefficiency will be around for a while and those who don't act on that, they
will be in big trouble, I think.

Question:
what type of tools are you going to implement in the near future?
Answer:
Besides what we already have. Yeah, so my main tools right now is ChatGPT4 and Mid
Journey and Firefly is hopefully gonna be amazing. I haven't received my beta yet even
though I was one of the first in line to get it. We'll see when it appears.
I'm sure that's gonna be very interesting. Then we tend to forget about these AI tools that are
inside of other tools. So I mean the AI functionality in Photoshop or in Pixelmator or in all
these video editing softwares or in Nimono. check that out. All these things are tools inside
tools that make use of AI so that people don't need to have the same granular skill sets as they
used to. And so that's gonna be a huge change. I think you see it, the Mono is a great example
of it. I don't know if you've seen that, but it's like an all-in-one podcast solution thing where
you just place a unit in the middle of the room and this is to the room that tone and What the
room and how the acoustics of the room are there are four mics and it just processes
everything for you And it will most likely edit it for you as well and not in the distant future.
I think you see it in the new or the recent Sony camera that was released. I can't remember
the name of it now, but it's like EZ-N1 or whatever it is. And where they're using a lot of
intelligence to like automatically frame the shots and that kind of thing. So you don't need to
have a photographer with you necessarily anymore. You don't need to have photographer
skills on the same granular level as you used to. So those kind of AI tools that are hidden a
little bit, you don't think of it as AI, but it is, that is, I think, a big trend where you get a lot
more automation into your workflow and be able to work much, much faster. So I think those
things are going to be big.

Question:
Okay, so how did you go about selecting the right AI tools for you and your business
and what were some of the main factors that you had to consider?
Answer:
Well, I'm super interested in technology, right? I'm an ex-engineer, I come from that world. I
find it extremely interesting. I built my first AI-based television analysis tool in 2004. So I've
been in this world forever, and I just love it. So for me, I really enjoy staying on top of
everything that's going on, and it's basically entertainment for me. It's what I love to do, is
just dig into all this.
So when I select, I think I'm so much in the space, but then I just see, I play with things a lot,
right? So I play around and then I realize, hey, maybe this could be used for that. And I think
that playful kind of attitude to it as well, which has been extremely useful and beneficial.
That's not why I do it. I do it because I think it's fun. But I think it's a really useful way of
doing it because there's no books on it. There's no way of learning, learning it like in a
traditional education style type way, you have to like play around with it.
question:
We were thinking like, did you have to consider when choosing your tools like
ChatGPT 4 that it has to be free or that it has been tested?
Answer:
Oh yeah, that's a known issue. Is it safe? I mean, yeah, of course, if I was to use it for final
work, I need to know about rights issues. And right now, there's no way of doing that because
nobody has... nobody knows. There are no court cases or anything like that to go on. So,
since I don't use it for live work yet, I don't have to consider that. or other tools that are
trained on a set of data that we have control over, or control over the rights for, then you can
start using it live.
So, but of course, that is a consideration. You have to be aware of it, so you just run off and
use it, of course. I think one consideration will always be, of course, quality, how good is the
quality? Simple, technical things, like resolution, is gonna be an issue. I mean, we have to
have high-resolution images for high-resolution use cases.
So, I mean, there are technical aspects like that, of course. What else is there? I mean, he's
abused, I mean, right now, the way that Mid-Journey works, for example, is almost like a
joke. We're going to be, I mean, obviously going to look back at the time where we're using
Mid-Journey through Discord as a very strange Much like we laugh about, I don't know, like
what the first iPhone looked like or whatever. But, yeah, so I think that ease of use thing is
going to be a parameter. If you use APIs, of course, API access is a parameter. So, yeah, I
guess those are things I think of. Okay.
Question:
How have your customers responded to the use of AI in your digital marketing efforts?
Answer:

And have you noticed any changes in customer behavior, for example in engagement? talk
about it in the beginning when it was like, wow, is this, does this model not exist, you know,
in real life. It was fun to talk about. But besides that, it's not really a big issue that we talk
about. I mean, people find it interesting, but since I'm not using it live, it doesn't really matter
so much for them. But of course, what they do see is that, okay, everything is now super
much faster than before, I am much faster than before, and it's much cheaper than before.
So of course they appreciate that. But people are curious, but I've not really been involved in
other issues around it since we don't use it for real work. So to speak, or live work. Sorry.
Sorry, I'm a little bit distracted now because I need to just go into the heart in here. Yeah, no
worries. Just give me a second. All right. All right.
All right. So if you have, okay, all right, sorry. Hey, I have booked a parking here. Hey man,
have you tried scanning your QR code? I didn't know I had it, let's see. I don't see any QR
code. Did it come through an e-mail? It came through a phone call, I think. It came through
an e-mail, let's see. I'm going to see if it works. It's not working very well.

Question:
So the next question we have is how do you ensure that your AI-based marketing efforts
are ethical and respectful of customer privacy and what steps have you taken to address
any concerns in that area?
Answer:
Ethical and respectful of customer privacy. I don't know if that's a relevant question because
customer privacy, would that be the privacy of my customer? Or of data collection, for
example. I don't think that applies really to what we do, since I only use it for sketches and
stuff, so I can't really see... that it is ethical and respectful of rights.
it's like we talked about before if you are using AI that is trained on data that you don't have
the rights to yeah then we have an interesting discussion right then there's almost like a
philosophical discussion it's going to be there's going to be a legal like we are going to not in
a not too distant future what the actual legal answer is to that. Maybe more so than the ethical
answer, I would say, because right now we're trying to figure out what that means. Because
in a way, you could say one argument could be like words that other people have the rights
to, and basically stealing it.
That argument could be made for anything. I mean, I'm inspired by everything all the time.
So, if I'm super inspired by a ton of things I've seen throughout my life, and I create
something kind of inspired by that. It's not also stealing them. It's a similar thing, but here we
have a computer doing the same thing. So that could be, you know, a way of looking at it.
The other way of looking at it is the stealing of course.
But either way, we are going to have a legal answer to that in an up to distant future. But until
then, it's a risk assessment you have to make, right? Because the ethical assessment is
something that we're still discussing. We don't really know. We're still trying to see how we
have people have different opinions on that. And the risk assessment that we are, I think most
of us are making, I certainly am, is that I'm not going to use anything like a live real
campaign that I don't know if I have the rights to or not. We're going to get Firefly very soon,
where we know that there's going to be a ton of other versions of that. I'm sure, you know,
Getty Images or whoever has a lot of data to train on will do that. So we'll see a ton of that,
you know, we have trained this system on data that we have control over the rights for. So, I
don't know if it's... So, you're thinking of that, you're making that type of assessment before
you use anything for your inspiration as I, from my talk at the beginning.
Yeah, I mean for inspiration, I think you can use anything. I don't think, I can't see that, I
mean anything that you're allowed to watch, I guess you can use for inspiration, right? Yeah.
But if you use it for production, that's a different story. So, yeah.

Question:
So, we're going to move on to the financial part of this, and that is, when it comes to the
financial perspective, have you seen any type of changes, gaining more revenue or profit
after implementing AI into your strategy?
Answer:
Yeah, as I said before, it's much more efficient, way more efficient, much faster. So, of
course, that means it's way more profitable. For now. But I mean, the market won't let anyone
have insanely high margins for very long. So, it will adapt. But what I think we will see is
that a certain type of, this is the same in like a law firm or anything, like the grunt work, the
simple work, that will become less valuable because we can have AI doing that. And it's
partly true for creative work as well in this type of sense where you're kind of inspired
So, those type of roles will get less valuable but I think the other type of roles where you're
higher up in the complexity, if you're a great creative director or if you're a great, like,
experienced lawyer or something like that, then you will be more valuable, because now you
can leverage yourself with a much higher capacity. So that I think is what's gonna happen,
and to be honest, that's what's always been happening in the world, not only with people, but
also with computers, that we reach higher and higher levels of abstraction, and you don't, you
know, the simpler things kind of get automated.
I guess that's what civilization is actually about, right? And then we're going to need, just like
with the Industrial Revolution, we needed a political system that kind of dealt with the fact
that you had some people benefiting from it, and other people who were losing from it, and
we created these political systems that we have today, basically, liberalism against socialism.
Those systems are made for that revolution, and now we're heading into a new revolution,
and we're going to probably need new systems that deal with this. So, I'm looking very much
forward to see what's going to happen there because I think that there needs to be a new
system of government there one or two new political systems that can compete against each
other. So that's my prediction but we'll see. We will see. Yeah look into look into meta
modernism if you're interested in that question because that's one of the one of the attempts to
address that issue that I've seen recently.
Question:
So how do you envision AI-powered tools improving customer engagement. How it's
going to improve customer engagement. Well if that's not applicable to you, then maybe
how do you use AI to recruit new customers maybe?
Answer:
don't really do that right I'm so much used to have a problem that their agency hasn't been
able to solve and they call me and I can't solve it so it's I don't really that's not really how I
work

Question:
so incorporating AI into your marketing make it simpler for you to achieve your goals?
And in what way?
Answer:
Well, I guess it's the same thing there, right? I don't have... I don't market to people in that
way. I don't do it for my marketing. And I think that part of that maybe is because also I
mean in a way, yeah, I guess you could say that I build because I'm also a user of these tools
that help me automate my own concentration or that kind of thing, but Yeah, I don't really
think so, but if you think about like my actual concentration that actually brings me
customers It's not about volume for me.
It's about having a really high level of sophistication, basically where people can understand
how I think about things. And automating that would kind of defeat the purpose, right, so
that's not really applicable to me, I think. Yeah. Let's go, yeah.

Question:
Okay, so what choices are you now making that AI could automate for you?
Answer:
I guess it's almost the other way around for me. I am the one making choices, but AI is
producing input for those choices.
I think we have the formula wrong, it might be... why don't do a test suite anymore because I
can just AI generate that test suite. Or, you know, let's come up with a few ideas to spark this
process. Then, I used to do that with creatives, like let's come up with a few ideas and bounce
them around. Have a brainstorming session or whatever. That AI does really well. Yeah.
I used to have the same thing. And maybe actually research. Research as well is something
that AI can do reasonably well. But the problem with AI is that it's one of the best ways to
describe AI. It's like a salesperson that doesn't necessarily know so much. Like a really good
car salesman that doesn't necessarily know so much about cars. Right. So it's really good at
convincing you that it knows.
But sometimes it's just bonkers, it's just wrong. It's the one thing. In a very convincing way.
The research, that is obviously a issue. Yeah. But it's good at structuring things, I guess, like,
alright, maybe a structured document for X and then we will ourselves validate all the data
and everything else, yeah. Then that can save you time, of course.

Follow up question:
Is that something you might consider doing in the
future? Answer:
If there was a tool to do that for you? I mean, we're already, I'm already doing it. So it's
almost like a, it's like a junior assistant kind of, right? So it points you in some direction. You
know, try this direction, try that direction. But when it tells you that, yeah, it's this exact thing
and this exact paragraph or whatever, then you have to be aware that it may be wrong, you
know? Like, it's like a junior assistant making you wrong, but it can still point you in a good
direction.

Question:
Yeah. So, I think it's a little bit like that. Okay, this is, I think that was about it. Do you
want to add anything or do you want to ask us anything?
Answer:
No, actually, maybe just for my own sake, it could be nice to have this, if you transcribe this
it could be nice for me to have it as well because I'm kind of thinking a lot about these things
and now I've been doing a lot of thinking with you so getting those notes could be useful
Yeah, no worries.

Appendix 7
Question:

Could you state your name? What do you do? a little bit your background? Your
position in the company?

Answer:
When I started working with sales and marketing, it was only Google. Ten years ago, my
then boss Peter and I decided to start Amplio. We are a full service marketing agency. to to
Instagram, of course, TikTok and Snapchat. We work a lot with search engines, Google, Bing
for example, where it's both paid advertising and search engine optimization. So the whole
search spectrum. I work a lot with Google Ads and such.

Question:

What kind of Google Ads are you using?

Answer:

Meta, Google, etc.

Question:

do you use any kind of AI to optimize this?

Answer:

Yes, we do. We use all digital channels that our customers need, based on their goals. So ads
or meta ads. But the strategy and the way we work is different depending on what goal, what
company, what product, what service. And therefore, for us it doesn't matter if the customer
comes to us and says, I want to see you on Google or I want to see you on Facebook. We
look at what is most important for the customer.
Or think about if Facebook is better. If you only work with Google, it's hard to say, no, he
doesn't trust us, he trusts them. Because they have to make a lot of money to buy bread. So,
we are independent of channels, but we are also independent of tools. There are a lot of
different programs. So you try to find those that you don't have to have 20 of for the same
thing. In that we work with some AI tools to improve the result, or be more effective, or save
money for the customer, or help the customer grow faster. example, a collaboration with a
company called Amanda AI, which works with the automation of Google Ads. And their
business idea is that a computer can do calculations and adjustments much, much faster than
a human. And that's a lot of what it's about in Google Ads.
When you've laid out the strategy, you've found the target groups, you've found, like, laid for
the It is unlimited what areas of use you can use chatGPT to. It is not just to take forward and
write texts. And just as chatGPT is now, we can not use chatGPT to just write text and then
we go. But you need to use these tools as a complement right now. and It's a fast process. But
you can also use it to get different ideas for marketing campaigns, social media, the simple
things that everyone is talking about right now.
It's the most common use for Chat-GPT. We have a predominance on Chat-GPT4, because
it's much easier to understand than 3.5. It makes less mistakes, is a little faster, and a little
more creative.

Question:
Are you planning on actually give a new service to your customer with the help of
ChatGPT4, as you know 4.0 can create a website from scratch.

Answer:

Both yes and no we are motto here in Ampilio to help our customers with their marketing and
make it more effective, measurable and better results, but then we use different tools to reach
our goals our services won’t change but the way we do it will change, how productive we are
but We are not going to change our name from a marketing company to a company that helps
with AI. If the customer came to us a year ago and said that they need a new website, we
would help them with that. If the customer today says that they need a new website, great,
then maybe we can help them with that with use of chatGPT to You always have to add new
tools, change what you do, because the market in digital marketing and the internet is
constantly changing.

Question:

Are there an AI that you want to adopt or implement in your company, that you
haven't done yet?

Answer:

The list is long and there are a lot of cool AI tools that I would like to have , the most hip, the
newest tools that exist right now. I have a list of tools that we will try and experiment with
and try out. I was in a conference in Stockholm where they show the newest AI tool it was so
exciting and there I created my list of AI where I want to try some of the tools I don't know
how committed you are with the subject AI, it feels like you've read it all, but it's happening
so much now on so many fronts at the same time. So, I think that's a good thing. I think its
going to go faster with time and won’t slow down we have for example “Gama” which is an
presentation tool based on AI. I think it depends on the companies to use AI that can help
them in their industry as much as possible and as good as possible to deliver value to the
customers and give them the service they need We are eight people here and some think this
is more exciting and interesting than others. Not everyone likes AI, but everyone uses AI in
some way to help in their chores

Question:

Do you Feel that AI is like the future? If one of your co-workers, doesn't want to adopt
to this new tool, are you just going to keep them in the company and give them more
practical jobs than using AI? or would you lay them off?

Answer:

We don't have anyone who opposes AI in our company. And if it was like that, I wouldn't
kick them out, because they don't like AI. I think it's illegal. But it's about creating, as a
leader in a company, it's important to create a culture that is inclusive, that makes you start
playing with AI.
There are many who are afraid of AI, like, what happens to my job? What happens to this?
What will the future be like? What will robots do to the world? The spectrum is very broad
there. But especially as people might think that I work with presentations, or with AI that
does presentations. Your work is so important that you might want to do presentations, but
that's not where you are. Take a journalist for example.
have stopped having as many journalists, they kick out staff, they stop printing newspapers as
often, and things like that. And then there are a lot of journalists who have been forced to
change professions. And this is not... Okay, but what should you do? Should you stop
working then, because you are 40 years old and you are no longer a journalist? A journalist is
good at writing. Well, take this journalist, but you have to write a news article for a company
instead. Can you imagine that? Because, instead of writing in a newspaper, you write in
MailChimp or Klavio instead. So then you use the creative and you can write texts, because
it's just a tool. And to learn a mail tool, it takes a maximum of a week. Then you can do it
instead.
And I think that AI is a big part of it. Use AI and work with them, because they won't take all
of our jobs. They will only change the way we do things. And we've had three big
revolutions. That's when oil and steam came, and we could transport things pretty fast. And
then we've had other... when industries have shut down. That doesn't mean you can still
create clothes. You can be a clothing designer. You may not be working on clothing
anymore. That's how I see it that way. You have to work with AI, otherwise it will be
difficult.

Question:

So you don't look at AI as a threat to your company or your profession right now?

Answer:

No, I don't. It's the way we work, the services we deliver, and what we do during the day. We
won't do exactly the same thing. When we start working with GPT, we change our way of
working compared to how we did before. We can use it to help us in our daily lives.

And that's how it will be in all our different areas. So, you have to be able to be pretty fast,
easy-footed, and be able to change your value-adding, ta Amanda Ai as an example, we had
one of our analysis expert, if we have Amanda AI to optimize those campaign then what is
my job now.. but we came also to the conclusion that instead of sitting and optimizing
campaign he can focus on much important staff instead of focusing on optimize the message
on each search word. Now we're going to build 10 more cents here, and 10 less cents there,
and so on. That's not what he thought was fun. Now we're creating this campaign, putting the
strategy. Now he uses the AI instead, because it's boring work, so to speak. So it's about
seeing what possibilities there are, instead of what obstacles.

Question:

Do you have any idea if any of your competitors, who also uses, AI to improve their
work?
Answer:

I think, and I don't know this, I think that most media agencies, advertising agencies, use
chat GPT in some extent. I think everyone does it, in different ways.

Question:

have you also started using Amanda AI in the beginning of this year?

Answer:

With Amanda AI we have been using it for a year, maybe. We had partnerships with other
generative AI for text production, long before Chat-GPT came on the scene, before it became
popular. We have a new dealer, for example, who we helped to generate unique descriptions
for his products with generative AI, about one and a half, maybe two years ago. And this was
before anybody know what Chat-GPT was. So what happened was that instead of putting a
quarter on each product and writing a unique text, it takes a long time to have 5,000 products.
Which is impossible so we notice if you used those tool for 2,5 years ago we could have
generated unique text for each product in e-ecommerce which lead to Google premiered his
products in Google search when they searched for some of the products. Which made him get
a lot more traffic and he sold a lot more of his products.

Question:

So, you have actually used AI for the past two or three years already?

Answer:

Two years, at least a concrete case.

Question:

Was there any specific reason you used or maybe you have already answered this, but
you use Amanda AI. Did you have more suggestions for other AI that you could use, but
you only chose Amanda AI?

Answer:

Well we have been pitched other tool but we choice Amanda AI beacue of its simplicity and
give us a lot of margin.

Question:

How have your customers responded to the use of AI in your digital marketing efforts,
and have you noticed any changes in customer behavior or engagement?

Answer:

They didn’t have anything against it right away, but as any other technics we had to explain
our tools and how we do it and how its brings them more value.
Question:

Did you have a case where a customer became a bit suspicious of your services because
of the AI?

Answer:

The difference is that if we want to come up with a pitch, a marketing plan or proposal for
our customers or a new customer, we explain that we are going to target TikTok, then I need
to explain the value of TikTok just like I need to explain the value of Amanda AI. people
have an opinion about TikTok as well.
Okay, maybe it's just for kids, how does it affect their integrity, and they live outside of
China, and so on. If you have AI, it's like, well, what, should a robot do automatically?
Imagine if they're wrong, imagine if the robot optimizes incorrectly. If you know that it's a
colleague of yours who's looking at the office doing this, then I think it's safer then AI. So I
think it’s the same concept or worries despite what it is, but we didn’t anyone complaining
about AI its always goes to its value and what it can brings to the table.

Question:

But do you tell your customers that you use AI?

Answer:

Or do they actually ask them if they use AI in their work or not. But in the cases where we
use programs like Malawi, we do it in collaboration with our customers.

Question:

We have talked about it, which tools we use for this. Have you had any campaigns that
are only driven by AI in your digital marketing?

Answer:

No, we don't have that. I don't know of any tool that does this per automatic without human
input yet. At least not in the channels and media that we work with.

Question:

But if you had the opportunity, would you try something like AI taking care of the
whole campaign itself?

Answer:

strong AI which can generate what it can improve and after a while it can learn everything
possible in the business, what your customer needs, so it will come up to you and say, do this
and do that, or we can send your SEO to Facebook. Would you think about adopting
I think so, absolutely. I don't understand why you wouldn't be able to do it if it was better or
just as good. Because if we have a 40 hour working week, me and my colleagues, if I can
make sure that, ok, now I'm saving 20 hours by letting an AI do these tasks, then you can
spend the time doing something else. So, of course you could do that. is very ethical and
respectful of your customer's private information or data? That's a great question. First and
foremost, you should not feed anything into a purchased QPT, which is a company's secret.
are business secrets. And I think you should approach AI and these things just as you do in
your own companies. So you should not send business secrets to the mail, or write them on
paper and put them in a library or on a file. In the same way, you should not take business
secrets or sensitive information to the chat GPT have If you do that, I think you have the
most knowledge.

Question:

And do you use CRM to collect information about your customers or CRM in general
in your company?

Answer:

Yes, we do.

Question:

And is there a connection between AI and CRM in your company?

Answer:

Yes, it does. We have a tool that is based on an AI solution called Guava. Okay. And it is a
prospecting tool that is based on an AI algorithm that finds prospects, potential customers,
who are in need of our services. Based on a lot of different parameters. And then they look at
the data points that are available, everything from where the company is, what is the
turnover, do they have growth, do they make money, what do they offer, what industry are
they in. Down on other levels, data that they can look at, for example, what CMS is their
website built in. Do they have Facebook, do they have Instagram, do they post there often

Question:

Do you think that CRM and AI have made the efficiency of your company even more?
How did it affect your company?

Answer:

Yes. There are some tools to write headlines that get really high clicks, like ads on Facebook
and such. But no, we can use a little more of that, for example.

Question:
But do you think that you need to send your team to a training about AI or something
like that? Or so long as everyone takes it into their own hands to learn about AI, about
chatGDP, or use AI and such?

Answer:

on All of it comes from Google Analytics, now replaced by Universal Analytics, It's a big
thing that's happening. And we've done training in that. And we learn by ourselves, and we
work in the tools. But to speed up the process, we've done training to be really good, fast.
Instead of sitting and trying to make an error. And that's something that you generally see in
the digital market. You take part in a lot of new information, education, webinars, courses,
and things like that, to learn more, because so many different things are happening all the
time. You follow the news, the biggest things that have happened in the digital market, the
biggest changes, and so on. So it's a cultural thing that people need to learn new things.

Question:

Is there a difference for you between AI and human labor?

Answer:

It's a deep question. But how they work, you have to relate to it in some way. If a computer
can print out 10 perfect texts at the same speed as I can write one, why should I do it? Then I
can do better. I just was at the lecture, where he... There was a memory master there, one of
Sweden's best memory masters. I think he won the World Championship with memory
master. In one hour he learned 1060 numbers in a row. Totally random numbers. He
memorized them in one hour. If you learn 1000 numbers, you can call it the Grand Master of
Memory. And then you translate it to data. If you saw a computer and said, I'm the Grand
Master of Memory, I can remember a thousand numbers, you'd just be laughing. Here we
have 512 gigabytes, but that's a lot of numbers, you can't really compare. On the other hand, a
computer can't tell the difference between a Chihuahua and a monk. Have you seen them?
Yes, I have looked at them. But to make a comparison... It's really... Is it a donut? This
computer is really hard to tell the difference between. We can do that. We don't have to think
about it. We do it right away. We have experiences, we have feelings in a different way. AI is
wonderful and it will be a part of our lives and will be more in the future. But it is still an
algorithm. Like it is right now. Then in a few years, there will be a big difference. But now it
is an algorithm. go percentage, did you get profit or gain? No, it's hard to say. It's hard to say.

Question:

But do you see it as a future thing, that it might be profitable later? Or implemented?

Answer:

I think that if you don't work with AI, it will be necessary for us to be profitable. There will
be some niche offices or professions that will use less and will be able to make money and be
profitable. If you train bodybuilding, everyone uses steroids. You don't have to take this into
account in the report, but as an example you can find something about this. If they build
muscle 5 times as fast, how the hell are you going to compete against them if you don't use
steroids? As an example. Same thing with AI. How could you compete with someone who
uses AI if you don't do it yourself? If it gets that much better? There are examples of
extensions many years ago. 4-5 years ago, an AI that helps with mammography and looking
for breast cancer on pictures. Have you heard of that? Yes, I saw that it was 4 years earlier?
This was that a doctor had the right to 90% of the cases and an AI had the right to 91% of the
cases. But together they had the right to 99% of the cases. That is where the difference is.
That is how we have to think and look at AI. Not like, should an AI do it or should I do it?
We do it together. We also do it with chatgpt.We write a text together with chatgpt, not that
we ask someone to write it for us. It's been a long time since we've taken code language
developers, for example, and asked them to help us create a chatbot for Slack or something
with these and this and that, that should have this function, and then you can write the code to
that. And that's good, but before, then you might have to go to a site called GitHub and
search for the code, because when someone writes it, they copy it and use it. What is the
difference

Question:

Now we go back to digital marketing. The problem you had before with digital
marketing, how did AI help you to find a solution for it? Or did you have a specific
problem before that AI came and helped you with it?

Answer:

They have been able to help in the creative phase when it comes to ideas and brainstorming.
It is also easier to search for information. Instead of using Google to search for 10 things, you
can send in one code in chatgpd.

Question:

What are the key challenges in your current digital marketing effort that you believe AI
helps you solve?

Answer:

challenges in your current digital marketing effort that you believe AI helps you solve? So,
basically, time and value. I think it was Time and Value. It's just these two that really stood
out. Effectivity. With the head thing. I think that's true. Yes, it is. That was a bit of a
complicated question, wasn't it? No, not really, but it's very deep. You can't just answer with
my opinion. That's true. I think you have to think a little extra compared to if you do it
yourself, because then you might have it in your mind. And that you try to think about the
moral of the matter, rather than an extra round about those things.

Question:

But did it take a long time for you to learn everything about AI? How to use it, how to
use all the tools you can for your own company?
Answer:

of I know all companies that have challenges with ads on Google.se. And how the ad channel
stands between other media, with both Bing, which is also a search engine, but also with
social media, newspapers, TV, search behavior and such things. And I also know the
limitations that exist within Google Ads because it takes time to optimize within the
accounts. in We optimize to get the right customer, the right ad, the right time when they are
looking for it. We want to get as many visitors to our website as possible for our budget. And
they should buy later. That's it. And we could do better with AI.To use AI in a different way,
then you might have to think about morality and ethics in a different way. Of course. Do you
understand? Yes. So, I think that those questions will be bigger in the future when we get AI
for real, when it goes more from being an algorithm to something else. And the way you can
use AI. I know of examples when troll farms in Russia used GPT to produce 20 000 tweets
in a day, and then used another AI to post on Twitter. Then you use AI for something that is
not so good. But you can still do that. You can post many tweets anyway. I don't know if
you've heard of them, but they made two AI's that introduced them to each other. And these
were smart AI's, or like strong AI's. So in the beginning they started talking about pretty
common things, like hey, hey, blah, blah, blah. And then after a day, they started generating a
lot of information, and then they started to use it.And then they started to talk, and the people
who connected them to each other, couldn't take over the control anymore. to talk to each
other, because they were trying to figure out what they were talking about in their own
language. And it was written like, we can take over the world, we can take over all people, it
was pretty dangerous in a way. So that's why I'm like, do you feel threatened in a way in your
company when it comes to AI? or we will take over your job or the whole company or do
something bad. No, not now. No, not now. There is also a risk that your data that you may
put in, is something you could trust in the future, will be hacked. Do you see that? Do you
think this is any risk? Well, you can be like that, but you can also be like that. We've had one
customer whose Facebook account has been hacked. And that was in an AI-initiative,
probably some sort of a key or a keychain or AI. But they were hacked, and they had nothing
to do with AI. Things don't happen anyway. But of course there are big threats from AI in
general for humanity and for me personally as well. But nothing that I see right now. We will
see what happens in 10 years.

Question:

Does incorporating AI in your marketing make it simpler to achieve your goals?

Answer:

Yes, I think so. Again, faster, more effective, better. Yes.

Question:

What choices are you now making that AI could automate for you by relying on
preserved data? What is preserved data?
Answer:

So, you have saved everything you already had! Yes, and I think we will have a lot more
data soon. Because there is still, in a way, it has been a long time since you have collected
data on everything you do. Both for your customers, in campaigns that you work with, but
also business data that you work with. And I'm thinking about data that can be anything from
how many minutes we spend in each program, which customers we spend the most time on,
which we can make the most money on. All kinds of data is saved, and no one can really sort
this or use all this data. Sometimes we can do it in certain ways, but not when we're talking
about big data in general. So, that I think is the next thing, that you can actually use all this
data that you save in a good and easy way. We were owned until April 1st by a company
called Spectrum One, who work a lot with data handling, structural data, and they have a
program called Enriched CRM with new and better data, for example. I think that all of these
types of functions will be easier and better to use in the future. And then they have this data,
but you don't really know what to do with it.
It's a pretty big project to structure this and get an insight from the data and work with it in a
good way. And I think that it will be easier to do in the future. and I have a lot of experience
in this field. I have a lot of experience in this field. I have a lot of experience in this field.
things that are close to his. Okay, thanks. Was that all? Yes!

Appendix 8

Question:

Alright, What do you do? a little bit of your background? Your position in the
company?

Answer:

I am the CEO of Eniro Group. It's one of the biggest media agencies in the Nordics, if not the
biggest. We are probably the biggest within our niche, which is SMBs, small and
medium-sized businesses. We're definitely the biggest. We have a couple of marketplaces. In
Denmark we call the crack.dk and in Norway give a seeded and new and in Finland Oh 100
100 dot fi.

Question:
Okay and how long have you been in the industry?

Answer:

Well in the industry I would say I've been for a couple of years more or less I used to work
for another company more specialized within analytics and data and insights, but truly being
operational in a company working directly with marketing, digital marketing, 11 months.
Question:
how much AI are you using currently in your company?

Answer:

Far too little.

Question:

On that note, is there any tools right now that you would like to adopt into your
company to make it easier for you to work with?

Answer:

We have adopted several tools. All right. So, with that, I mean, having that said, that we
work too little with AI doesn't mean that we're not working with AI, but I'm saying if you
would ask me in six months time from now How much is the increase from the AI tools six
months ago? It will probably be Tenfold so it means no working too little right? Yeah, most
of that. It will be another tenfold So I'm expecting it to grow To grow in relevance and
importance for our for conducting our business.
<what we do and we implement an AI tool it's not just taking me taking it in and go to a
couple of my guys working and tell them yeah you start working with AI I can't do that
because I'm too big of a company I need to have implementations in order to make everyone
working with it.
0:09:16
How do I do that? Just give them a tool? No, that cannot be done. So I need to have a
structured process for the implementation, which is required when you're a little bigger
company. Yes. Where were we? We're on it. Yeah,

Question:

could you just give us some examples what kind of tools are you using right now which
is based on AI?
Answer:
Do you want the names of the tools?. I don't know if that is okay. But maybe what they do for
you. that we can talk about.
So, yesterday we actually, well we have an AI tool, creating automatic logos. You start your
own company, you want to be a plumber or a car rental company, whatever, you want a logo,
right? We'll create that logo for you, with our AI tool, in the platform.
That is one of it. The other one is websites. You need a website if you start a company. Well,
many of them starting a company wants a website, not all of them. And that's also decreasing
statistics. Less and less companies actually have websites. Because they go directly to
Facebook, to Instagram, to TikTok, and create their web shop in those platforms, social
media channels. I read an article saying that new companies, 65% of all new companies
actually create website and that is decreasing trend, which is really interesting.
I mean, and we have a big business doing websites. How will that affect us, right? What
could we do? Should we try to get into creating the Facebook pages and social media pages
and e-commerce and so on. So that is the tool for creating logos via AI. The other one is
websites. We do create websites and we have drafts and we build upon those drafts to our
new customers in order to build the websites. What we have done now is implementing a tool
that actually does it for us. What that tool does, we have through an API, so through an API
for pictures and content and also access to the internet, of course, this tool creates a website
for you. What company, what's your dream? What company do you want to start?

Question: For us a marketing

agency. Answer:
So you're creating a marketing agency. So we have a customer on the phone and he wants for
his digital marketing agency, his name is this, and he wants also to add these and these and
these services. Okay, bam. Before, we have statistics on it, before it took us about three days
to create that website in order to make it look good, feel good, you know, the look and feel of
it, but also the optimization for the search engines, the SEO and the really good content. It
took us about three days because everything that we did, I mean, putting, dragging, dropping
a picture there and writing some content there and some customer case studies there, it took
us about three working days. Now we did it with this AI tool. How long did it take?
30 seconds. Saving a lot of time there. It's saving a lot of time, it's saving a lot of money, but
it's also enhancing the quality. We don't have any errors in the spelling or in the grammar
because it knows the language is 100 percent, right! Yes. It doesn't misspell. What we do
have now is actually a challenge. How do I do it in a What am I doing with my people doing
the websites?
How can I use them for other tasks to ensure the quality or the compliance or that the data
that the AI writes actually is accurate? So that's the challenge. So the websites is another one.
The third one where we use the AI tools is to create blogs and newsletters so basically
contents for an ad I'm putting an ad on Facebook I want to have some text about it BAM
what tonality what kind of nuance what is my target audience are they young or they old or
they're women are they men what kind of language or should it be happy or style or
professional and we're using AI to create those those contents either to send out a new
newsletter to put on your blog or make a Facebook ad For your website. That is the third
Tool that we're working with What else I myself using an AI tool for my emails.
So every email that I get, I have an AI tool installed that actually suggests answers. So it has
read all my content in my email. It knows how I speak or how I write, what kind of language
I use, and how I change that, depending on who is the receiver of the email. If I send it to a
partner or an investor, the language is in a certain way, right? But if my colleague sends me
an email, it's maybe much shorter, not so nice, but only done in a very hasty way, and it does
That helped me a lot Yes, so moving on with another question We haven't implemented yet
in the company, but we are implementing it, me and the VP, the Vice President for Marketing
Partner.
We have a meeting on Google Meets, right, me and him. So that tool, you press, you play it
when you have the meeting on Google Meets, so it records it. After the meeting has ended, it
also takes notes from your meeting. So I don't think I have a pen and paper aside to write
what we agreed upon, what we said. It summarizes the meeting for me, taking notes and
summarizing and also pointing out decisions that we took. Okay, Mika, we do it this way,
and you're going to do this until that date and we're going to come back and we're going to
have a meeting on that day. Is that real one? Yes, he says. And he tries that. And those notes
are then sent out to the ones in the meeting. Which was really great.

Question:
We would need that right now. Now we know. Okay, so our next question is, are you
aware of any competitors or similar enterprises that have successfully implemented AI?
And if so, do you see them having an advantage in the market right now?

Answer:

I'm not going to specify any competitors now and talk about them and how they are ahead or
behind me. So talking about general companies in my industry I Have a feeling they're not
taking it seriously. I had a friend, his name is Walter, he had a company, a communication
agency, okay, so he did the strategies and the communication for companies. He had, not that
big company, it had a revenue of about 80 million SEK, there were 40 people in that
company. So it had 40 employees and he was the founder and the CEO of course he sold it
three years ago he sold the company became well independent from anything so now he's got
a real nice house and a really nice car but what he also had was a non-competing
commitment to that company he wasn't allowed to work within communication for three
years, but he sold it three years ago. In February, it was three years ago. He started a new
company now with 40 new colleagues. You know what the names of the colleagues are? No.
Not me either, because they're all robots. So he created a new communication agency with
only himself and 40 robots. Of course, he could have hired 80 robots, but he wanted to make
it like his old company, but all replaced by robots. That's amazing. he had worked for that
company for 15 years approximately and built up a revenue of 80 million, 80 million. He's
worked with this company since February. It's two months, right? He went up 10 million in
revenue.
10 million. Because he's got 40 people doing this job for him. At almost no cost at all. You
know, they don't need vacation, they don't need good coffee, they don't need big fabulous
office spaces. They work all day. So he creates these strategies and pictures and
communication roadmaps for the big customers of it. 10 million in two months.
I am stock listed. I can't tell you about my tools, I can't name them for you, because I have
that information, that secret, but for him it's a private company. So he can go more in details
about the tool and the names of the tools

Question:

How did you go about selecting the right tools and technologies for your business needs?
And what were some of the main factors you were concerned about?

Answer:

Okay, let's start with the second part of the question, what I'm concerned about right and that
is GDPR And then you and you AI act have you heard about the new AI Act from the EU? A
new AI Act coming out you should research it and write about it in your essay It's coming
and it is being implemented This year what it talks about is that the information the data that
your AI is using to generate new information. Who does it belong to? So, chat GPT, it takes
data and information from all the internet, right?
But it also uses you, it used to use you to learn more. What they have done now, chat GPT
and open AI, is eliminating its possibility to learn from you. So it cannot take your
information anymore. It cannot develop by you using it because they weren't allowed to. And
the GDPR, who owns the information and what information can we actually talk about in the
content that comes out? Those were, of course, concerns.
But also, of course, we have probably heard it I mean the bias of the AI how biased is it and
what problems could be giving us and how should we control and have internal controllers to
actually call it secure the quality of the outcome before publishing it or giving it to our
customers I mean you've probably heard about what Chat GPT used to say about black
people and woman and Jews and yeah and because that was what it was reading right there
was a there was a Twitter account, I think it was two or three years ago, a Twitter account
that was all AI, it was a robot and it tweeted and its main objective was to get as many
followers as possible. I don't know if you've read about it or heard about it. But what it
actually posted on Twitter, it was horrific things, but it got because it was coded to get as
many follows as possible. So that's what it did. So it did a job basically, then they had to shut
it down. So that's of course a concern. I also read an article about other concerns and it said
that 50% they asked a lot of developers working with coding AI, 50% of those developers
answered that it was 10% risk that the AI would make the human being go extinct.
50% of them said that the chance of AI extincting human being is 10%. Okay. Do you see
what I mean? Imagine you boarded an airplane. And then they told you, well you know, you
want the service guys, the mechanics who maintain and have built this airplane, 50% of them,
half of them are saying that there is a 10% chance that you all will die if you ride this
airplane. Would you do it? No. Just 10%? And it's only half of them. Would you do it? No.
Of course not. Nobody would. But now here we've got the developers actually developing
this tool. 50% of them say that there is a 10% risk or chance that humans will go extinct
because of AI. Nobody's listening.

Question:

How did you go about selecting the right AI tools and technologies for your

business? Answer:

Yeah so we had a list and we had investigated, well, via Google, of course, what we did first
is see what could actually be helped by AI, and we started this project last summer. So we
had the big list first, what kind of places in the organization could actually be enhanced or
efficient, more efficient by an AI tool.
After identifying those areas, we looked for companies out there that were actually working
with this. So we had 10 companies that we believe these should be in the project of our pilot.
We talked to all of them, but only four got actually through the eye of the needle and one of
them then after that fell out as well and then we took three of them and implemented all of
them. Then we did pilots with all of them to see are they really as good as they are,

Question:

Approximately, how long it took for you like to actually implement those in your
organization?

Answer:

Three months. Three months.

Question:

How has the integration of AI impacted your marketing team workflow?

Answer:

Can you elaborate a little bit on that? Sure, basically like... Like the progress of basically, I
guess you kind of answered this from the very beginning because you said like for you it took
you from making a website in three days to 30 seconds. So I guess we have an answer for
that one.

Question:

How have your customers' response to the use of AI in your digital marketing efforts?

Answer:

Very positively.

Question:

Did you have any customer with concern because you guys are using AI? Or like, it felt
threatened?

Answer:

A few questions, and I would actually expect some of them to reach out to us and say
something negative about it or being concerned about it. But no, none so far.

Follow up question:

Alright. Can you provide any example of successful AI-based digital marketing
campaign that your company has executed so far?
Answer:

Yes. Yes, I can. Without talking about the actual clients that we have done it for. Our clients
are SMBs, right? They're from 1 million revenue to 100 million revenue. That's the target
audience of ours. We have about 50,000 customers. That's a lot of customers. That's a lot of
people. so what we have done is it was a logistic company. They market shares in their area
and wanted to try out well new kind of content in order to push visitors to their website
which we did together with them and we increased it by 570 percent. 570 percent not before
doing a campaign but compared to other campaigns that they have done. So compared to the
other campaign that they have done manually 570% increased visitors. So the content and the
pictures and the colors were all details decided by AI.

Question:

With this success are you now planning to do kind of more campaign like this just based
on AI?

Answer:

My objective is that we won't be doing anything manually when we are done with
implementation. Every campaign will be AI driven
Question:

How do you ensure that your AI based marketing efforts are ethical and respectful of
customer privacy?

Answer:

That's a great question and that is also what we talked about before, right? The GDPR and the
privacy and the ethics. So what we do now, we never ever release anything before securing
the quality of it manually. But we also have a very strict and rigorous contract with those
providing us with the AI tools. Because we don't have any of our own. we haven't developed
any AI, right? We're licensing those. We are having very rigorous and structured contracts in
order to, well, be, to secure our own backs and our customers' backs so that we have it all in
the contract, but how we limit the outcome of the tools.

Question:

Okay, so now we're going to go a little bit more to CRM related questions. Do you use
CRM in your company right now?

Answer:
Yes.

Question:
Are you currently using also like AI, which is basically inserting your CRM system?

Answer:
No. Not yet started because we are in a project where our CRM right now maybe isn't the
CRM we are having in the future.

Follow up question:

Okay, so you are planning kind of like to implement it but like maybe on a long term or
like in a near future?

Answer:

We are definitely planning to implement it, but we do have a CRM system right now. Yeah.
It might not be the one that we will be using in the future. Okay, well. Don't tell that to our
vendor.

Question:

When it comes to your perspective, have you seen any changes after implementing AI?

Answer:
 Well, I think the AI tools in most areas will give us much, better margins and better quality.
And the margins will be something good for us, but it will also be better for our customers,
since we can make our process much more efficient, our prices will go down, right? We
won't have to charge our customers as much as we're doing now, but still have a better
margin. term, lead to more customers wanting to work with us. I cannot talk about figures,
even percentages in this. You know, we are stock listed, I need to be limited by what I can
do.

Question:

But you can just confirm that it’s gives you positive result after implementing AI?

Answer:
Yes, it does.

Question:

so now we're going to move on a to the digital marketing question. So, what are the
challenges in your current digital marketing efforts that you believe that AI will help
you solve?

Answer:

Well, the efficiency is one of them. Our backlog will be diminished, basically, with the AI
tools, but it will also help us to create new revenue stream but it will also enhance the quality
of the outcome. So no grammar misspelling or wrong grammar or a language that we can, I
mean if we can, our AI, what language we are speaking, I'm not talking about Swedish, but
I'm talking about what kind of language we use for different customers, then we will also
have a product that's much more, it will be the same every time, where, where of now, we
have Marika and we have Joel doing the same job, but if it differs two weeks between these
jobs, they will look differently.
If he does it in the morning or if he does it in the evening, if she does it when she's in the
office or working at home, it differs, right, the outcome. Having the AI tool will create a
homogenous outcome. Yes.

Question:

How could AI be used in your marketing, especially help your organization adapt to the
future, like to the future market?

Answer:

either adapting or die I would say and as I also repeated before it my advantage right now is
that it seems they're not taking it seriously and I am surprised that this is their mindset and
also decrease my cost base, right? If I can decrease my cost base compared to my
competitors, I can also decrease my prices and have much more competitive pricing.

Question:

Exactly. Will incorporating AI in your marketing make it simpler to achieve your goals?
Answer:

since I came in 11 months ago, we started directly with it, I mean during the summer. First of
all, just a little back, when I came in as the CEO, I had seven persons in my management
team. I fired them all. They all had to go. The first day actually. So from the first day they
had all to go. I don't know, there was a report in the newspaper, I don't know if you read it.
So they all had to go and why they all had to go, well, they didn't have the mindset for the
future of this company. It was still working in Enelo as Enelo was three years ago. So during
that summer I was all alone. I was sitting in this office all alone working and talking and
discussing with myself. After the summer I had a plan of implementing it because I had a 90
days plan when taking the position of the CEO. I had a 90 days plan what I wanted to achieve
and one of them was having a plan for AI.
So directly after the summer of me recruiting my new management team, I had the plan for
them to actually execute on. Your question was, will it change or help me to achieve my
goals? What I am rather saying is that AI made me have new goals, much higher goals. So, I
don't have goals that are based on not having AI. My goals are based on actually having AI
implemented. Do you see the difference? So, if you have, if you're manufacturing trucks,
Volvo, okay? They're manufacturing trucks, that's their business. And then AI comes and
then you would ask them the same question will AI help you reach your goals? Their goals
weren't identified or quantified after AI it was quantified before AI and then AI came how
can it now help? Well I don't have any goals before AI. My goals were created with the
knowledge of AI helping me to reach them. Without AI, those goals are worthless. So they
are based upon me having AI in the matter. So basically you thought into the future already
before even you started your company? and for me it was so clear what was going on, what
was happening. When ChatGPT came, that was when people started talking about AI, right?
We knew AI before ChatGPT, with other tools. We're not even using ChatGPT. That I can
tell you. We're not using it. We have other tools. ChatGPT is a great tool, but not for us at
this moment. I'm using it myself sometimes to write ChatGPT, to write letters to the
organization. I send out monthly letters. And it's a writing. But we're not using it in the
business, conducting the business. It's like more for personal usage, you kind of take charge
of the PNL,

Question:

So our final question would be what choices are you now making that AI could
automate for you by relying on preserved data?

Answer:

Say that again. What choices are you now making that AI could automate for you by relying
on preserved data. So a choice I'm making based on the AI? No, what choices are you
making right now that could be used for AI instead. So what services or what actions am I
doing that could be replaced by AI? Yeah, we have a list. We do have a list. I'm thinking
about how much of it... This will be affecting my organization heavily. Even if you're not
writing an article in the newspaper, you're actually writing an essay that will be published
someday, right? Sure. And you having that information now when it's not public would mean
that I am breaching my limitations. I choose not to answer that but I could tell you that there
are a lot of areas that is still not affected by the AI but that will be affected and that will
affect my organization that will affect people it will affect people's lives.
Appendix 9
Question:

Could you state your name? What do you do? a little bit of your background? Your
position in the company?

Answer:

For the past three years at developing machine learning, actually applications, deep learning
algorithms for infrared sensors and real-time surveillance systems. That has been for the past
three years. And then I founded the company, so I'm the project manager and the CEO of the
company for the past, since 2020.

Question:

Okay, wonderful. And what comes to your mind when you hear the word AI? or
Artificial Intelligence?

Answer:

Oh wow, that's a very broad question. Like the first thing is automation I guess. But AI is a
multiple different stuff. So you could look at it as creating your own brain. I started this by
looking, I have one application So I can literally say I am the one that can detect Gerard
stealing the best. I have this application in my brain. So I recreated it as a virtual machine.
This is the first thing I see as an AI.

Question:

when did you become interested in AI?

Answer:

I've always actually been interested in machine learning ever since like the first like years of
that basically mid-decade you know in 2015-16 when it started to when we developed faster
algorithms that could process data faster that's literally the birth of AI. AI was actually
created in 1952 so it goes way beyond you know.

Question:

in your company currently, how much AI are you using as a tool?

Answer:

It's just AI, like do you mean in the sense of the product or do you mean in the sense as
using it as a tool for business means?

Question:
 I mean for instance for your product how much AI are you using in your product to
like develop your products or use your products?

Answer:

So okay so like the product itself is of course an artificial intelligence well when we build
artificial intelligence there is no artificial intelligence for building an artificial intelligence It
is possible to create that type of automation, but not in our circumstance So we don't use that
much AI to build an AI We use here from normal programming of course we it's all See we
work with C and Python of course and we combine these two languages to build a core
setting for our AI. That's the only tools that we use. Then if you talk about like notebooks, it
could be Visual Studio Code, it could be Jupyter notebooks and all of those type of stuff.

So using the tool AI in business, we have never done that. We literally use our own business
skills in that sense to approach customers, to conduct business meetings, reports and so on.
Because right now, it's actually a very good question when you think about it. In the future,
maybe we'll automate it. But usually we're not using any tools. Many people are using
chatGBT, we're not using it. So yeah.

Follow up question:

For instance, for your own marketing, for your, let's say, for your company and such,
are you using your own AI to kind of reach to that extent where you can actually benefit
from the marketing?

Answer:

We have thought about it, but the thing is our company is very different when it comes to...
You know, an AI company doesn't do digital marketing with AI sometimes, because we're a
security company we're not targeting normal individuals we're not targeting civilians we're
not targeting social media in general we are literally working with the government or the
police and they usually they we don't need AI to market to them in that sense so basically i
kind of already know which is my niche so you don't really need like to market yourself more
than that because you're working just with the government basically. so we have like it's
called direct sales. I created this tactic actually here in Lund University, I call it direct
aggressive sale which is just rushing on them, knocking on the door and selling straight.

Question:

Do you have any competitors which have similar products that you think are doing a
much better job because they are using much more AI, let's say in their business wise or
in their marketing and such?

Answer:

Yeah definitely I think there are two companies there are like one German and one French
that is actually a very good competitor we are quite different than them but they have
developed in their niche they have developed really amazing tools and applications which
they have used but looking in our niche because we are more for like real-time surveillance
but as well gun detection and we can literally see you carrying a gun under your clothes by
using infrared sensors micro millimeter waves and all of these we can see literally what
you're hiding they don't have these type of applications and the way that they are marketing I
think they are marketing without AI as well so AI companies are quite different because we
create the AI that then retail is used for their own marketing. Yeah, that is basically the
difference.

Question:

how did you go about selecting the right AI tools and technology for your business
needs? Or your product?

Answer:

Yeah, so it comes, if I go out of my sphere, because I have a lot of information about my

partners and industries I could share with you rather than us since we are a security company,
and we don't use that much AI within our digital automation. But usually how people are
using, there are multiple different models of AI. You have models that are called supervised
learning models in which you're teaching the AI, and then you have unsupervised learning
models where the AI teaches itself based on parameters given. So you have the first set that is
supervised. They're using that one, for instance, in bankings, which they call, if you see this
number, this number means bad. Or if you see this number, this number means good. And the
banks, for instance, Morgan Stanley, have been using AI since 2010, actually, in developing,
reading their documents. Before, it took them around 32 years to finish a decade, like, a
decade documenting And what they do is that they revise it after 10 years and so on. With AI
it took them 2.3 seconds. Yeah, so it just took all of this data and calculated. So this is how
you could use it in the banking system by checking documents and all of this type of stuff.
Me, for us, we're using it to whenever you do something wrong in society the police or ICA
wherever it's deployed will get a notification within a nanosecond yeah so there are multiple
different applications so it's a very broad question.

Question:

How do you go by going forward from or differentiating yourself from the competitors?

Answer:

Very good question actually. It's a very, this is what's making AI now go crazy because
people are trying new stuff. So how we differentiate us is that we are tweaking the models
with more parameters. Fortunately we are invested like supported by Lund University so we
get a lot of funding from there. We use that funding to create more data because AI, the
programming is nothing. The data is extremely important, what you feed it because people
learn from that. So what we're differentiating us with is basically how we are collecting the
data, how much data we have, what specific the data is. For instance, I will give you a very
good example. We had data collection for infrared cameras where we can see the blood flow
of your face, how the blood is moving. The only data sets that existed in the world was
Chinese database basically.
The problem is that will make the AI biased because if the AI doesn't see a non-Chinese
person, it will say that's not even a human. So it won't analyze it. But we created the first
European and Scandinavian data set for motion detection and recognition. And this is what
makes you very differentiated from other AIs. And this is what sells. This is what people
want. They want that data. So a very good question. I hope I answered it.
Question:

How do you currently manage customer data and analytics and what tools and
technologies do you use to support your CRM efforts, if you have any CRMs?

Answer:

Yeah, like so we have a different type of It's quite different for us in that sense So we we are
we're having a lot of that type of data coming in but that type that type of data is maybe more
focused on privacy control I suppose So we we have a lot of GDG PR problems and
customer data coming in because our data is not just Shafi entered the ica and bought this
Snickers and then it saved him my data because he purchased something. It's more either of
two products. Shafi entered the building, I saw his blood movement, I saw his biometric data,
I know exactly his subconscious feelings, I know exactly what he's thinking about and where
he's looking. That type of data is very very crazy. So how we work with that data is that we
try to work with GDPR. We can't identify individuals. We try to create AI applications.

Maybe now my eyes are seeing you and the data collection is behind me. So when the data is
coming here in my eyes, I delete the footage and I only send, I saw Shafi had a red t-shirt,
but I can't send you the video Yeah, this is how majority of AI Companies are working today
with privacy control and CRM so on they try to do it as much video footage as much
knowledge about the individual and only extract the numbers Because you can get sued that's
cost a lot of money of keeping those data safe, like we have created a whole infrastructure,
first of all cyber security, because any AI is at risk of being hacked, and if you can hack an
AI, you're literally hacking a brain and you can control it. So, this AI contains that data, so
people will want to stop, they want the data to take it from you and use it for their own
purpose which FBI and CIA and all of these majority big security intelligence companies do
we institutions and secondly if you have for instance people want to control your AI that's
also a big security issue so we built and sorry, I might go a little bit technical here but yeah
we built us a crazy cyber security infrastructure which is very needed for keeping these type
of data sets safe and for keeping the AI safe.

Usually people put it in, do you know, this is very important if you actually talk about data,
it's called Amazon Web Services. AWS and then you have Azure from Microsoft and then
you have Google have theirs and there is multiple companies working there. These people
provide service that are heavily like protected so you can't just hack AWS. You need to be a
crazy hacker to do it. But you're still giving up your data to Jeff Bezos. He can still access it
at any time. Yes, I don't know, but I was asked, what impact do you think AI could have on
your company's ability to personalize customer experience and improve overall engagement
and satisfaction, I guess this is for the clients you have, maybe the government or security.
security to use AI in it We use AI to for increasing customer satisfaction. We are doing it in a
third-party way like we are Increasing the customer satisfaction at the e-cow which is our
customer and we're providing the security system for them so it goes that far and and
Basically, I would just say this is quite straightforward. We do it by enabling first of all
enabling security for their customers, which will increase their customer satisfaction and it
will increase further on eco-satisfaction because their customers are satisfied, being satisfied.
So it's very, it's a lot of business questions you have. Usually AI companies do not work with
deploying AI into the business scene, since we are creating the AI that is deployed to the
business scene.
Question:

So speaking of your company, let's say, are you still relying on human labor or do you
think that you can also replace your workers or let's say your programmers with AIs
that can actually generate AIs?

Answer:

Well, it's quite impossible to generate AI from an AI. You can generate the code, but you
can't generate the data. The data is very difficult to generate. You need to collect it by
yourself. So these people are there, so we don't use AI there. We use maybe mid-journey
sometimes to create content and so on, which is a good content creation AI. And then you
could use it sometimes, we use chatGPT sometimes, but for some questions about errors, we
could ask it. But usually we do not use AI to create AI. As we use AI, it's very difficult to use
AI to create AI.We employ people from India a lot now to sit in front of a computer for
approximately 20 hours to see what is a gun and what is not a gun. And they have just to say,
yes, it's a gun.

Question

Do you see AI as a part of your ROI or an unnecessary and disadvantageous tool right
now?

Answer:

No. Overall is definitely a return of investments for the company. In Sweden, I have to pay
my employees, I have to pay them an insane amount of taxes just for me so they tax them. Do
you know with an AI it works for free 24/7? That's what my customer was saying: employing
a police officer that is looking at the screen. Do you know how much it costs for a police
officer to look at the screen for 8 hours? You know how much that cost? 60,000 kronora
month, you know how much an AI costs? Zero. Zero and does AI go on a sick leave? No.
Does AI go on a baby leave? No it does not go. So of course it's the biggest return of
investments and that's why it will kill 60% of the jobs that exist in the market because they
are too costly.

Appendix 10
Question:
So Tell us about yourself, your current occupation and your field of work?

Answer: I guess I work as what we call an advisor within the area of customer experience
and the solution that we sell. So I'm talking to customers and explaining to them how we can,
with our solutions, help them to get a better customer experience for their end customers. So
I've been working at SAS for 13 years, and worked in the industry with SAS and marketing
for 25 years. Yeah, my name is … and I work as a solutions architect with SAS. So I'm
basically helping our customers design and implement our solutions, and taking the technical
considerations more into mind, basically making sure that everything works and runs as it's
supposed to.

Question:
What comes to mind when you think about AI?

Answer:
Ooh, I can start. First of all, I mean, AI is for us as an analytics company, not a new thing,
right? We have been doing the Google Analytics for, I mean, it was founded more or less 50
years ago, with the idea of making it easier to make use of advanced analytics. And I guess
AI is essentially something that we have been doing for at least the last 20 years or
something, although it hasn't been mentioned as AI. Nowadays, I mean, it's, from my
perspective, which is focused more on marketing, that is in the neighborhood for automation,
to automate processes, automate decisions.

So that is one way of looking at it. Then obviously, recently, a lot of focus has been on
ChatGPT, and those kind of things. And yes, they can also be part of this. And then we can
come back to that later on. But for me it's this about automation and scaling the decisions,
right? Yeah, I would agree in parts of that. I mean, the things I think about when it comes to
AI is mostly that it's a lot bigger than people think in that sense. I mean, it seems that
probably AI could be most things. Right now we're talking about just ChatGPT and basically
generative AI, as … is speaking about. But for me, AI is also a lot about removing feelings
and basically working data-driven in the sense, basically, what's the data telling us? Can we
work predictive? For instance, can we use math and history in a sense that removes the
human decision from a lot of questions that shouldn't have feelings in them.
Question:
How much is your company using AI, and for what?

Answer:
Well, yes, we are both building it, but we are using it, I guess, in many ways. I mean, one
way of doing it for us is that we are monitoring all our visitors to our sites, on our website,
web properties, and analyzing and trying to optimize the marketing and what we want to
show them digitally. So you can argue that yes, we are analyzing our customers, both in
batch but also real time, in order to see what should we propose to them next based on what
they have done.
Question:
Are you aware of any competitors or similar enterprises that have successfully
implemented AI? If so, do you see them having an advantage over other companies in
the market now with their way of using AI?

Answer:
I mean, there are a lot of competitors out there doing things and creating software and
algorithms that obviously is doing what we are doing.
So yes, there are a lot of them out there. You can see the big ones that are obviously in the
discussion right now about generating AI, but you also have the open source models,
languages that is also doing a lot of things that we are doing. Like R and Python and those
kind of things. So there are a lot of things out there. A lot of the crowd platforms have their
own kind of AI or advanced analytics solutions embedded in their environments. Yes, and
as
for your company, are you planning to adapt even more AI in the upcoming years or like near
future? Well, that I would assume. I mean, obviously I'm not part of, neither of us is part of
the kind of internal process or how we are using our own systems.
We are more meeting our customers, trying to convince them on how to use our solutions.
So, but I would see that in general, the uptake on using AI and the adoption of AI is
increasing in the market. So amongst our customers, yes, we see a greater adoption of AI.
Question:
How did you go about selecting the right AI tools and technologies in your business?
Answer:
Well, we are using our own tools, so that's the easy part. So, I mean, given that we are
building our own AI tools, right, we are, I mean, we sell AI software. So, obviously we are
using our own software for that. So we are using AI in our CRM.
Follow-up Question:
What kind of AI CRM do you use, and for what purpose? If you can elaborate a little
bit more?
Answer:
I mean, I can elaborate. The solution that we use, we use our own marketing solution, our
Custom Intelligence 360. And essentially, what it does it's it's a combination of trying to
combine data channels and content because what marketing is all about obviously is to create
a great experience for the customer which means that you need to be relevant towards your
customers and to be relevant you need to have as good understanding about the customer as
possible. And with that comes, first of all, data. You need to get hold of all the data that you
can get. And also given that data can be spread among different data sources and systems, the
accessibility to the data is important. Data quality is also important. And then we are
applying business rules, first of all, starting there. Then we are in batch taking all the data,
putting them together and analyzing the data in order to find clusters and segments of
customers that performs in a specific product or offer. So that is the foundational piece of it.
Then the software itself also uses AI and machine learning in many ways in order to produce
product recommendations.

We are doing A-B testing, multivariate testing, multi-arm banded testing in order to test
content on the site, which content provides the best kind of performance. We use it for
attribution modeling to algorithmically assess what kind of channel and campaign is helping
customers to convert. it to optimize, we're talking about optimizing the customer journey and
with that using AI in order to find what is the best possible path for this customer going
forward. So there are many ways we are using it and I think the AI use cases will just grow.
And this, like the CRM usage right now. I mean the way we do marketing If we will see
around called is talking about marketing The CRM is many things right? It could even be a a
sales things to that way. We're more or less have a call center support, or we are actually
doing marketing. But when marketing says, I would say, like we have changed it rather,
rather than the landscape of doing marketing has changed, right, digitalization has changed
the way we do marketing today. And then, yes, and with that, it becomes more and more
channels, and then it becomes more and more complex to do this. And yes, from that end, we
need to adapt organization and strategy in order to make it as effective as possible
Question:
How has the integration of AI impacted your marketing team's workflow and processes,
and how have you addressed any potential challenges or concerns?

Answer:
I mean the way we do marketing If we will see around called is talking about marketing The
CRM is many things right? It could even be a a sales things to that way. We're more or less
have a call center support, or we are actually doing marketing. But when marketing says, I
would say, like we have changed it rather, rather than the landscape of doing marketing has
changed, right, digitalization has changed the way we do marketing today. And then, yes, and
with that, it becomes more and more channels, and then it becomes more and more complex
to do this. And yes, from that end, we need to adapt organization and strategy in order to
make it as effective as possible
Follow-up Question:
And what are some of the challenges or difficulties that needs to take into consideration
when it's implementing AI with CRM?
Answer:
I think the implication of this is, I think, it is to understand what has driven the decision.If
you are sending, I mean, in essence, what AI will have to support is to select and help us to
understand what channel and what offer should we give the specific customer. And in order
to understand why we came to that conclusion and have that kind of threat to understand
what happened and what came up to that decision. I think that is the important piece to get
hold of when it comes to chat GPT and those kind of things, which is also on the horizon,
which is also what we can do and use in the marketing piece, then it's all to test and
understand and test the algorithm to see that you are fine with it. Because if you're going to
automate a decision based on the third party algorithm, like GPT, then you need to test it and
verify it before you put it into production. And there are ways to get around that, right? You
could either have maybe the generated AI algorithms to support you in your creative process,
but you are still, as a person, to take the final decision.

So for instance, you can have a generated AI to create an email route, but you are in the end
responsible for concluding the email that will be sent. So that is when we are maybe creating
new ways for the organization and for the processes when using AI

Follow-up Question:
Do you have any ethical concerns when it comes to AI? Like, are you afraid, for
instance, it might gather a lot of information and then it might get hacked or something,
or like, it will become under cyber attack? Do you have any precaution for that type of
situation?
Answer:
Not specifically more than we have for other systems, I would say, in that sense. I mean, we
have worries about AI, I would say, more ethically about it being a black box, usually, so you
can't really verify, for instance, what comes out of chat GPT in that sense, as Thomas was
talking about. But when it comes to security, from a security perspective, I would say AI
systems are systems like other systems. So I wouldn't be more afraid of someone hacking the
AI system rather than another system in that sense. But from an ethical point of view, then,
that is more on also to secure that the data is not biased, So to secure that the data that you
have can really reflect the truth before you are actually trying to have the algorithm learn
from the data, So if you have bad data, obviously you will have a bad end result. So I think
that is more important in that sense, to secure that you will not get biased results,

Question:
What are the key challenges in your current digital marketing efforts that you believe
AI help you solve?

Answer:
I mean, first of all, what we see today is that when we are with the digitalization, we did
traditional campaigns, batch campaigns, where we might have segmented customers, let's say
10,000 customers, and based on the specific selection criteria, we found out, we want to
contact these and offer them, give them an offer. And we were in that old world, we as a
company were deciding when we wanted to communicate with the customer. Today, with the
digitalization, that has changed. So every interaction is now driven by the end consumer and
customers. When they are using the app or the website, they are starting the interaction with
us as a company. Which means that for us to be able to personalize the experience, we need
to automate all of this. So I think automation is the key change that we see has happened. So
we need to automate and with that, in order to, we can automate that by doing analytics and
put the result of the analytics of the scoring models and segmentation and put them into place
to personalize the experience for the customer. We can put in business rules, say, we want to
show them product A, but based on that the customer actually showed interest in product B,
let's push for product B instead. These kind of decisions is also what we need to automate in
the process. And in the end, we can also initiate a real-time decision. So based on specific
behavior, we might want to call a model to run and take a new decision.

So it depends on that and those kind of things could potentially be if we want to make a
credit decision in realtime, for instance, then you would run a model in real time and based
on past and current behavior that will affect that model. So automation is the biggest change,
I would say, that is driving the need for AI and also how we are approaching and working
with.

Question:
How do you envision AI-powered tools improving customer engagement and helping to
recruit new customers?

Answer:
I mean, as I was talking about, automating the decisions and being more relevant
What AI can help us be, If we want to create a great customer experience, we need to secure
that you as a consumer get a relevant customer experience. If you come into the site and they
are trying to push a specific product, but you are, I mean, they are pushing jeans, but you
have been over and over showing interest in shoes. Why shouldn't they push shoes for you,
right? Give me an example. If a company doesn't reply and listen to you as a customer and
understand what you are interested in and actually show that in the communication you will
actually leave that brand and go to another brand instead. So we need to be better and be
relevant and I think that is what we see.

Question:
Will incorporating AI in your marketing make it simpler to achieve your goals? Do you
feel like it's changed?
Answer:
Like AI basically making it easier for you to reach your goals since you are also selling AI
services? I mean, I think... I would actually ask also this from one of our customers rather
because what we are seeing is that we are helping customers to be more efficient and actually
be more to the point because when implementing AI in the marketing process you will
actually be more relevant and that will increase conversions, it will increase sales, it will
increase that the customer comes back, it will increase loyalty. And all of these kind of KPIs,
customer-oriented KPIs, will drive your organization forward. So that is where the value lies.

Question:
have you ever tried to use AI to one of your marketing campaign?

Answer:
Well, I'm not part of that campaign process, but I would say it will still be a mix. You need to
have a strategy behind. If you're going to make a campaign, you need to have a strategy.
What do you want to do? What is the end goal? Maybe what kind of product or product
sentiment are you going to push. So you can't just do AI, you need to have a strategy around
it. And then you're gonna use AI as part of that process, whether that is only for decision,
view or bait. And you can argue, first of all analyze the customer base initially to get an idea
and understanding, and then you can also provide to use AI in the actual real-time case, So in
that sense, yes, we can use it throughout the process, but still you need to have strategy and
processes in place as well.

Question:
What choices or decisions are you now making that AI could automate for you by
relying on preserved data?

Answer:
I mean, I would say that the decisions that we take, From a marketer's standpoint, would be
to define what offers should we do, when should we present them, through what channel, and
for whom. And then obviously automate the recommendation process for our customers. to
be able to make decisions at scale that we are using it for. Because now as well, we can't sit
there and wait for, now customer B is coming into our site. What should we do? We click on
this button, We need to automate all of this. And some of this is based on business services
and some of this is based on AI.

Question:
Do you see AI as a part of ROI, if yes, elaborate why?

Answer:
Yes. In marketing AI is used to automate and take better decisions. Automation and
increased relevancy will improve customer experience which in turn will lead to increased
loyalty, decreased churn, increased NPS, Increased sales (cross- an up-sell), etc.

Question:
When it comes to the financial perspective, have you seen any type of changes gaining
more revenue or profit, after implementing AI into your strategy?

Answer:
A Nordic insurance company used our attribution algorithm in their marketing. The results
include an improved customer experience, increased conversions, and reduced marketing
costs for the organization.