Conversational AI Cheat Sheet: Choosing The Right Solution For Your Business
Conversational AI Cheat Sheet: Choosing The Right Solution For Your Business
1
Tractica (2019, Q4). Customer Service Virtual Digital Assistants; IDC (2019, Q4). Worldwide AI System Spend for Automated Customer Service Agents.
2
Conversational AI:
What is it and how does it work? CONVERSATIONAL AI APPLICATIONS
Before diving into specific Conversational AI, it is important INCORPORATE CONTEXT, PERSONALIZATION,
to understand exactly what Conversational AI is and how AND RELEVANCE WITHIN HUMAN TO
it works. COMPUTER INTERACTION. CONVERSATIONAL
DESIGN IS A KEY PART OF DEVELOPING
Conversational AI is the set of technologies behind CONVERSATIONAL AI APPLICATIONS THAT
automated messaging and speech-enabled applications FLOW AND SOUND NATURAL.
that offer human-like interactions between computers
and humans. It provides human-like communications
by recognizing speech and text, understanding intent,
deciphering different languages, and responding in a way
that mimics human conversation.
3
4
How does it work?
Conversational AI uses various technologies such as Automatic Speech
Recognition (ASR), Natural Language Processing (NLP), Dialog management,
and Machine Learning to communicate with humans at a human level.
5
CONVERSATIONAL AI AT A GLANCE
ASR Natural Language Generation
Listening Offering response
Application receives information input from human—either Dialog management orchestrates responses, and converts
written text or spoken phrases. If input is spoken, ASR, also them into human understandable format using Natural
known as voice recognition, makes sense of spoken words Language Generation (NLG), which is the other part of
and translates them into machine readable format—or text. NLP. Application either delivers the response in text, or
uses speech synthesis, the artificial production of human
Natural Language Understanding speech, or text to speech to deliver the response over
Comprehending a voice modality.
Application deciphers what text means by using Natural
Language Understanding (NLU)—one part of Natural Machine Learning
Language Processing (NLP)—to understand intent behind text. Learning
The application accepts corrections and learns from
Dialog Management the experience to deliver a better response in future
Forming response interactions. This is called machine or reinforced learning.
Application forms the response based on its understanding
of text’s intent using Dialog Management.
6
7
CONVERSATIONAL AI SOLUTIONS: FROM SIMPLE TO ADVANCED
Applied Conversational AI solutions range in abilities and therefore offer different experiences. Thinking about what your
brand wants to deliver is key in choosing the right application.
Chatbot
The simplest example of a Conversational AI application. FAQ bots, or what most of us know as chatbots, are basic answer
and response machines that have popped up on the websites and mobile apps of many brands as a way to automate simple
interactions between a brand and its consumers. Most chatbots do not use NLP or dialog management to offer dynamic
responses or machine learning to improve over time.
8
Chatbot | CX Chatbot | Important considerations
When interacting with chatbots, consumers must type Chatbots are limited to text channels, which are rarely
the exact keywords required to receive the appropriate the channel choice of consumers or operational spend
response. The conversation flow is controlled by the bot, for companies. Therefore the impact on your customer
and the flow breaks or gets stuck in a loop when the experience and financials could be minimal. This also limits
bot is unable to understand the customer’s intention. your ability to scale across other channels and therefore
Chatbots often fail to maintain context over the course can’t grow with you as you continue to expand your CX
of a conversation and cannot understand customer intent initiatives as they change and grow with your business.
in many situations. Consumers often feel like they are
going around and around in circles to get questions
answered or information needed. This can lead to wasted
time on the consumers part and a frustrating experience
with your brand.
9
Can I order 4 apples please.
10
Virtual Customer Assistants
Virtual customer assistants, often referred to as Intelligent Virtual Assistants, are more advanced Conversational AI systems
that serve a specific purpose and therefore are more specialized in dialog management. These applications are able to carry
context from one interaction to the next which enhances the user experience.
11
Hello and welcome to Franklin Fruit.
How can I help you today?
Virtual Customer Assistants | Important Considerations
Virtual customer assistants can be an excellent way to offer Hello—yes. I needed to update my order.
12
WHICH ONE IS RIGHT FOR YOUR BUSINESS?
As you can see, there are two main types of Conversational AI applications that vary in terms of complexity and experience.
The critical difference can come down to the vendor you choose. Just like applications, vendors differ in their type and
area of expertise, scalability, and proven results. Make sure your decision-making process includes a vendor vetting process
where you ask detailed questions about experience, design, and referenceable results to ensure your application works
exactly the way you want your brand to be represented.
13
About Interactions
Interactions provides Intelligent Virtual Assistants that seamlessly assimilate Conversational AI and human understanding
to enable businesses to engage with their customers in highly productive and satisfying conversations. With flexible products
and solutions designed to meet the growing demand for unified, omnichannel customer care, Interactions is delivering
unprecedented improvements in the customer experience and significant cost savings for some of the largest brands in the
world. Founded in 2004, Interactions is headquartered in Franklin, Massachusetts with additional offices worldwide.
866.637.9049
interactions.com
14
interactions.com