17/12/23, 12:32 Gartner Reprint

Licensed for Distribution

Top Trends in Data and Analytics, 2023

28 March 2023 - ID G00776758 - 38 min read

By Gareth Herschel, Ramke Ramakrishnan, and 26 more

Top trends in data and analytics technology and practices can help anticipate change and
transform uncertainty into opportunity. Data and analytics leaders must factor these trends
into key investments that drive new growth, efficiency, resilience and innovation.

Overview
Opportunities
“Thinking Like a Business” allows the D&A function to focus on adopting a proactive approach
to delivering value while also accepting the responsibility of “owning” the implications of its
activity.

Moving “From Platforms to Ecosystems” thinking allows the entire organization to be both
more agile in delivering immediate capabilities and to ensure a consistent and holistic view of
both the technical architecture and the organizational context of its activity.

“Don’t Forget the Humans” is a necessary rallying cry for organizations that may have become
too “data-centric or technology enamored” and lost touch with the engagement of their
employees and their broader responsibilities to the organizations and society.

Recommendations
Data and analytics leaders must:

Define their role as leader of a business center delivering key insights to decision makers
across the organization. Act more like an internal consulting company with its own P&L than a
business support function that represents a cost center to the rest of the business.

Plan their D&A strategy in the context of ecosystem thinking where each element impacts and
is impacted by the whole. This means investments in D&A teams must work with their
Enterprise Architect colleagues to ensure D&A capabilities are factored into a consistent broad
technical architecture as well the balance of risk and reward D&A investments can bring.

Plan on engaging with stakeholders across the organization to understand the best approach to
drive D&A adoption (which will not always just mean more and better analysis and insights but
also aligns with human psychology and values).
https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 1/24
17/12/23, 12:32 Gartner Reprint

What You Need to Know

The mission of data and analytics is clarifying from one of managing data resources and creating
analytic insight to delivering value. This need to deliver provable value to the organization, and to
do so at scale, is driving a variety of trends across the data and analytic function (see Figure 1).

Figure 1: Top Trends in D&A for 2023

These trends can be broadly grouped into three themes:

1. “Think like a business” is a necessary requirement for focusing on the value delivered to the
organization rather than the production of insight.

2. “From platforms to ecosystems” is the adoption of a more agile and all-encompassing

approach to thinking about how data and analytics integrates into the rest of the organization.

3. “Don’t forget the humans” is the recognition that, although organizations are increasingly
automated, data and analytics (especially AI) is better aligned with people rather than used as a
substitute.

Theme 1: “Think Like a Business”

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 2/24
17/12/23, 12:32 Gartner Reprint

The data and analytics strategy of a think-like-a-business leader identifies how the data and
analytics team will deliver maximum value to the organization in a “competitive landscape”
against disconnected D&A initiatives across the organization, external substitutes or a “no data”
mindset.

This shift in mindset has four key implications:

Be proactive, not reactive: Leading CDAOs are proactive about driving awareness,
understanding and usage of data and analytics across the organization. They are not “order
takers” waiting for requests for a new dashboard or predictive model.

Devise an approach for selling data and analytics to the rest of the organization: Framing the
value of data and analytics in the language of the “customer” as a “product” that will drive
growth and innovation, manage risk and reduce costs to the rest of the organization (their
mission-critical priorities) is essential.

Balancing cost of delivery with the value delivered: D&A leaders have a variety of mechanisms
to deliver value (for example, developing skills in-house versus working through consulting
partners), and part of thinking like a business is to understand the trade-offs implied by these
different approaches. D&A leaders must understand the cost implications of different
technological approaches, not just to minimize costs, but to enable a discussion about which
approach delivers maximum ROI.

Own responsibility for data and analytics assets and outcomes: Thinking like a business
means acting as part of a value chain or ecosystem and also accepting responsibility for the
consequences of your own actions. Not only do D&A leaders need to ensure the adoption and
usage of their insights, but they also need to act to avoid improper use of the assets they are
creating.

Trends reflecting this theme:

Value optimization drives a focus on business impact.

Managing your AI risk drives thinking about the risk and reward of adopting new technologies.

Data sharing is essential reflects the need for a full understanding of the environment.

D&A sustainability acknowledges the double-edged sword that analytics represents.

Observability drives the focus on understanding the cost drivers of analytic activity.

Theme 2: “From Platforms to Ecosystems”

Increasingly strategic use of data and analytics and business need for agility are driving a shift
from siloed and disconnected components in data and analytics platforms to integrated D&A
ecosystems. This is enabled by market convergence across data and analytics capabilities and

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 3/24
17/12/23, 12:32 Gartner Reprint

improved services composability. But a D&A ecosystem is not just the network of technical
platforms. As a socio-technical system, it should also encompass considerations such as:

When technology is a substitute for human workers and when it is a complementary

enhancement

The role that humans and AI play in the broader social and environment ecosystem

This shift in focus has three key implications:

1. Increasingly comprehensive and integrated end-to-end architectures make adding new

capabilities more seamless and less costly: D&A capabilities that are less costly and that have
a less complex architecture require little or no integration of products and cloud services. This
lowers the barriers to adoption and increases the viability of marginal or experimental use
cases, thus increasing the potential for pervasive use of D&A across the organization.

2. Better understanding and sharing of resources and capabilities enables greater value to be
created: The opportunity to reuse capabilities across the organization not only offers more
return on D&A investment, but also allows deployment of common functionality across multiple
different use cases (such as embedding common reporting functionality into multiple different
operational systems).

3. Integration of D&A into the organization and societal ecosystem has a broad and diverse
impact: Data and analytic governance needs to embrace discussions about how to reconcile a
shortage of human skills with:

Increasing technological capabilities

Issues of data sovereignty

The acknowledgment of data privacy and security expectations

Considerations of the environmental impact associated with the sourcing of technology and
power

Trends reflecting this theme:

Data sharing is essential is a necessary part of understanding the organizational ecosystem.

D&A sustainability connects the organization’s behavior to its impact on the broader ecosystem.

Observability traces the flow of data and insight throughout the ecosystem.

Practical data fabric enables the construction of the ecosystem.

Emerging AI will transform the ecosystem, especially by integrating humans in new ways.

Converged and composable ecosystems are the new technological bedrock to systems delivery.
https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 4/24
17/12/23, 12:32 Gartner Reprint

Theme 3: “Don’t Forget the Humans”

Concepts such as robotic process automation (RPA), techniques such as AI, and use cases such
as driverless cars and algorithmic trading are all designed with an underlying expectation that
things would work much better (or at least more cheaply) without humans. However, a number of
factors are driving a more balanced view of D&A that accepts that, although AI tends to
outperform humans (in some narrow use cases), AI plus a human tends to deliver better results
than AI alone. In certain use cases, it is even undesirable to actually remove humans from the
process.

This focus on retaining the human dimension to D&A has three key implications:

Driving adoption of D&A by working with business users to ensure accessibility, trust and
relevance to business users: Giving workers the ability to control their own environment has
repeatedly proven to increase performance and satisfaction. Giving data and analytics users
the power to influence or even control the way analysis is created and used increases the
likelihood of its use in their decision making.

Recognizing human involvement across the range of decision making, from decision support
to decision automation: Even the most automated decision-making process needs human
involvement in its initial design and formation and in its subsequent monitoring and evaluation.

Ensuring D&A deployments consider multiple aspects of risk: The risk of rogue AIs, humanity
disenfranchised by ubiquitous technology or environmental collapse are science fiction tropes
for a reason: All are catastrophically plausible. The need for D&A leaders to guard against the
worst possible implications of trends in D&A while ensuring the best potential outcomes
requires careful attention and diligent governance.

Trends reflecting this theme:

Managing your AI risk is the recognition that AI delivers unpredictable impacts that require
careful governance.

D&A sustainability traces the impact of D&A on the human environment and quality of life.

Emergent AI transforms the relationship between humans and organizational systems.

Consumers become creators reflects the changing role of business users in modern D&A from
being passive consumers of insights to actively engaged insight creators.

Humans remain the decision makers acknowledges the primacy of humans in the decision-
making system.
Value Optimization
Analysis by Rita Sallam, Alan Duncan, David Pidsley, Andrew White

Description:

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 5/24
17/12/23, 12:32 Gartner Reprint

Value optimization from an organization’s data, analytics and AI portfolio requires an integrated
set of value engineering activities and competencies. First, building a value story identifies and
articulates stakeholder impact on processes, activities and decisions from D&A initiatives in
business terms. This is needed to get buy-in from business peers and establish how results will be
measured. This also provides the foundational value assumptions needed to prioritize
investments as part of portfolio optimization and D&A strategy and operating model development.
Value stream analysis and design thinking are techniques that can help identify potential high-
value, cross-enterprise options and decisions. Measuring outcomes using outcome-driven metrics
that link technical outcomes to business outcomes and adjusting the portfolio as needed ensures
the organization is realizing the expected value.

Why Trending:

Ongoing and constant business and technology disruption is putting increased pressure on all
business and technology leaders to demonstrate the value of their initiatives.

Most D&A leaders struggle to articulate the value they are creating for the organization in
business terms. A belief case (as opposed to a business case) is no longer enough. Only 35%
of respondents were effectively meeting an objective of showing demonstrable and verifiable
value to data and analytics stakeholders. 1

D&A is increasingly distributed in business lines with significant benefits to business domains.
However, alignment and collaboration with D&A teams is needed to ensure D&A-enabled
change, organizational agility to respond to change and connected decision making.

Focusing solely on ROI to guide investment decisions leads to overweighting tactical versus
disruptive benefits and misallocation of resources.

Implications:

D&A leaders must show a clear link between D&A initiatives and the business outcome
improvements to processes, activities and decisions tied to the organization’s mission-critical
priorities. Otherwise, data and analytics will be viewed as a cost center ripe for reduction when
resources are constrained rather than as a strategic growth and innovation driver.

Discovering and aligning to value streams to deliver D&A initiatives maximizes desired business
outcomes while efficiently utilizing scarce resources and reducing excess system waste.

Investments in data and analytics initiatives can yield short-term financial returns in revenue
growth, productivity, and cost and risk reduction, but their more significant impact is often on
leading indicators of financial outcomes and future value. These financial-influencing benefits
(including an organization’s brand value; know-how; innovation; organizational agility;
achievement of mission; leverage to other projects; and employee and customer retention,
acquisition and loyalty) are often excluded from value assessments. A comprehensive
approach to value identification, communication and option selection is needed to optimize
value realization.
https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 6/24
17/12/23, 12:32 Gartner Reprint

Most D&A organizations build business cases to get funding and then deliver the technology,
but never measure the actual business impact and results. Actually measuring business
outcomes is needed to understand the true value that data and analytics delivers.

Actions:

Optimize value by building value stories that establish clear links between D&A initiatives, the
organization’s mission-critical priorities, and the stakeholder business outcomes and value they
create. The value story should include outcome drive metrics to link technical outcomes to
business outcomes and should identify when adjustment is needed.

Identify opportunities to optimize how the organization makes decisions by using Gartner’s GDI
model. 2 Identifying how data and analytics improves specific processes, activities and
decisions is a critical piece of identifying stakeholder business outcomes.

Make the best possible resource allocation decisions by creating a consistent approach to
scoring investment options across financial and financial-influencing value and risk. Mature
organizations establish a value council consisting of D&A and business leaders and measure
results and embed value engineers into product fusion teams.

Maximize the business value of your D&A investment by exploiting value stream analysis to
connect related decisions to maximize decision impact and minimize unintended
consequences.

Realize expected business outcomes and value by building a value stream management
competency that combines value stream identification, embedded DevOps and lean thinking
with data and analytics delivery processes.

Drive decentralized D&A users to work together to speed up iterative development of D&A
solutions by establishing cross-functional fusion teams supporting your value streams.
Enhance these cross-function, cross-discipline teams with lean and agile practices.

Further Reading:

How to Optimize Enterprise Value From Data and Analytics

Optimize Decision Making and Business Value Creation by Aligning D&A With Value Streams

Change Management and Communication Unlock Data-Driven Business Value

Creating a Modern, Actionable Data and Analytics Strategy That Delivers Business Outcomes

Managing Your AI Risk

Analysis by Pieter den Hamer, Svetlana Sicular, Sumit Agarwal, Avivah Litan

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 7/24
17/12/23, 12:32 Gartner Reprint

Description:

With the growing use of AI, new risks are emerging that require mitigation. These risks relate to
bias and discrimination, trust and transparency, privacy and ethics, safety, security, finance, and
corporate reputation. More risks and more laws and policies to regulate the use of AI have led
organizations to start adopting guidelines, practices, roles and tooling for AI governance, including
“responsible AI.”

Why Trending:

Although many people are excited about AI’s potential, there are also deep concerns about its
growing power and proper usage. Many examples in public media about unethical or
malfunctioning AI have greatly raised the awareness among consumers, citizens, policymakers,
politicians and companies about the risks of AI.

Organizations are increasingly aware that scaling AI from local use and prototyping to
companywide production systems means that AI is ever more business critical. Thus, scaling
the use of AI introduces significant risks that require monitoring and mitigation.

Governments and corporations are introducing more policies, guidelines and legislation to
regulate the use of AI, balancing its innovative power with the need to protect privacy and avoid
unfairness in more automated and opaque AI-powered decision making.

Implications:

A growing number of industry organizations, open-source initiatives, commercial tech firms and
others offer guidelines, tools and techniques for AI governance and responsible AI that
companies can use to start supporting their own AI risk management initiatives.

Risks include but are not limited to ethical risks such as bias, discrimination or privacy
violations. Attention toward risks (e.g., the security of AI itself, the risk of model theft, the
poisoning of training data or the circumvention of fraud detection) is also growing. Other risks
include financial damage or customer churn in the use of chatbots or automated decision
making.

Managing AI risks is not only about being compliant with regulations. Effective AI governance
and responsible AI practices are also of key importance to building trust among stakeholders
and to catalyzing the use and adoption of AI.

Actions:

Use the Gartner TRISM framework (see Further Reading) as a structure to implement and
execute risk management for AI.

To balance the managing of risks with the required flexibility for experimentation and innovation
with AI, follow an adaptive approach to AI governance.
https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 8/24
17/12/23, 12:32 Gartner Reprint

AI governance requires a combined approach of guidelines, processes, roles and techniques.

Adopt these as part of the broader emerging discipline of AI engineering while leveraging the
growing support for fairness, model monitoring, security and other aspects in AI platforms.

Many AI risks, in particular ethical risks, pose challenges and dilemmas that cannot be fully
regulated through guidelines. Instead, they require awareness of practitioners and involvement
of stakeholders, preferably through an AI governance board.

Further Reading:

A Comprehensive Guide to Responsible AI

Top Strategic Technology Trends for 2023: AI Trust, Risk and Security Management

How to Manage the Risks of Decision Automation

Observability
Analysis by Ramke Ramakrishnan, Melody Chien, Ankush Jain, Jason Medd

Description:

Observability is a characteristic that allows the system’s behavior to be understood based on its
outputs and enables questions about their behavior to be answered. Observability in the D&A
system allows the practitioners to monitor and comprehend its behavior across multiple
components within the ecosystem by observing external outputs such as its activities,
measurements, requests and dependencies. It enables D&A personnel to control the system’s
usage, management and inference to make timely, cost-effective business decisions using reliable
and accurate data.

Observability can be applied in a wide range of areas within D&A, including but not limited to data
quality, data drift, data pipelines, data orchestration, management operations, analytics, ML model
deployment, systems performance, infrastructure with AI-enabled optimizations and
recommendations.

Why Trending:

With modern D&A systems increasing in complexity and distribution while generating a large
volume of data, a significant effort is required to parse, understand and facilitate changes so
that they become more adaptable to deliver new capabilities.

D&A platforms that are distributed combine multiple architectural patterns and services tiers.
These include data storage, data fabric, metadata management, data integration, data analytics
and DSML deployments, and they will require a comprehensive and integrative approach to
observe, inform and deliver recommendations with better governance to achieve business
goals.

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 9/24
17/12/23, 12:32 Gartner Reprint

Enterprises adapting to agile development practices and accelerated data-driven

transformation using xOps are seeing an increasing demand to gain comprehensive visibility
into the entire D&A life cycle. This includes processing, orchestration and execution of data, ML
models and pipelines supported by machine learning and predictions to improve D&A system
performance, uptime, trustworthiness and total cost of ownership.

Traditional methods for managing data pipelines are no longer sufficient in preventing
downtime. Disparate interfaces for monitoring events, inability to correlate multiple faults in a
pipeline to rapidly identify root causes, and statically defined rules to monitor data cannot
accommodate the demands of modern data pipelines to ensure reliability and consistency.

Implications:

More organizations are seeing an increasing need to monitor, track and record all activities,
including actions and movements. The hope is to help with the impact and root cause analysis
for issues and offer recommended changes for better management and deployment of their
D&A applications.

Agile development practices and accelerated data transformation using dataOps,

microservices and serverless functions contribute to an increasing demand for accelerated
delivery, greater details and visibility to mobilize data-driven decisions.

Every organization looks at observability differently depending on what it wants to track and
trace, as well as on resources and capabilities, with the objective to identify issues and address
them proactively before the failure happens.

Effective cost management on cloud analytics deployments using FinOps requires the
intelligence of data observability in the capacity planning processes in order to track resource
utilization to align with business growth targets and to improve performance.

Actions:

Develop practices to track every aspect of your D&A systems and its components. Start with
measuring data usage and quality to demonstrate the business value of data observability
supported by machine learning and predictions.

Establish a robust data collection process that includes activities, measurements, requests and
dependencies across multiple systems, services and components to develop better insights.

Understand your D&A system’s current state based on the movement, access and utilization
data it generates. In this way, you can proactively learn, understand and visualize through
applied intelligence to determine predicted outcomes and be proactive in operating D&A
systems efficiently and at reduced costs.

Manage your D&A systems to not only observe and improve infrastructure, data collection and
AIOps processes but also gain insight into how the consumers are interfacing and interacting
https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 10/24
17/12/23, 12:32 Gartner Reprint

to ensure that they are getting the best experience.

Further Reading:

Quick Answer: What Is Data Observability?

Innovation Insight for Observability

Monitoring and Observability for Modern Infrastructure and Applications

Data Sharing Is Essential

Analysis by Thornton Craig, Lydia Clougherty Jones, Lydia Ferguson, Andrew White

Description:

Data sharing is a business-facing KPI of providing enterprise value. It is an essential business

capability, enabling access to the right data at the right time to derive the right insight. It includes
sharing data both internally (between or among departments or across subsidiaries) and
externally (between or among parties outside the ownership and control of your organization).
Data sharing initiatives can also create the foundation to support data as a product (data
products) and to share data across marketplaces or data exchanges. Mandatory data sharing can
be driven by legislative requirements, and voluntary data sharing aims to increase competitive
advantage across industry consortia.

Why Trending:

Organizations increasingly need more types and combinations of data to feed their use of data
and analytics to drive business value. This data is often found in silos within the organization
and, increasingly, external to the organization.

D&A leaders responsible for building a data-driven enterprise increasingly depend on data
sharing, yet they struggle to identify, locate and enable reuse of the data that is best-aligned to
the business case.

Sovereign data strategies and policies highlight data sharing as a key priority for increasing
government efficiency and generating value in the public and private sector.

Implications:

D&A leaders know that data sharing is a key digital transformation capability, but they lack the
know-how to share data at scale and with organizational confidence. Data governance
programs must work with data sharing initiatives to ensure the data ecosystem is mastered
and trusted.

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 11/24
17/12/23, 12:32 Gartner Reprint

Organizations can create “data as a product,” where data and analytics assets are prepared as a
deliverable or shared product.

Data sharing collaborations, including those external to your organization, increase data sharing
value by adding reusable, previously created data assets. Automation and machine-readable
metadata provide automatic discovery of datasets and services, lowering the barriers for their
discoverability and resharing.

Actions:

Consider adopting data fabric design to enable a single architecture for data sharing across
heterogeneous internal and external data sources.

Establish trust in internal and external data, metadata, data sources, data sharing technologies,
and downstream reuse and resharing ecosystems through automated mechanisms and
metrics from active metadata insights, augmented data catalogs and automated data quality
metrics. Apply situational trust, eschewing perfect trust, which, ironically, can increase risk.

Align data sharing with accelerating stakeholder value by branding data reusability and
resharing as a positive for business value, including ESG efforts. Identify data reusability
opportunities in self-service analytics and incorporate data reuse and resharing into domain-
and enterprise-level D&A strategies.

Further Reading:

How CDAOs Need to Prioritize Data Sharing Investments for Digital Business Success

Why Situational Trust Is Key to Data Sharing for Business Value

Case Study: Citizen Digital Identity Enables Personal Financial Data Sharing (Singapore)

Quick Answer: How Executive Leaders Must Prepare for the New EU Data Act

D&A Sustainability
Analysis by Pieter den Hamer, Gabriele Rigon, Erick Brethenoux

Description:

Awareness in the market is growing rapidly about the relation of D&A and AI with sustainability, in
particular when it comes to energy consumption, greenhouse gas emissions and climate change.
On the one hand, with their growing pervasiveness and complexity, D&A and AI are increasingly
hungry for compute resources and — as a consequence — electricity and energy. On the other
hand, the potential of D&A and AI in making many other activities more sustainable and helping
businesses and wider society meet ESG goals outweighs their own environmental footprint by far.

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 12/24
17/12/23, 12:32 Gartner Reprint

Although this potential is yet to be fully realized, it should not be overshadowed by concerns about
D&A and AI’s own energy consumption.

Why Trending:

Using AI, including D&A, to improve sustainability across organizations and society at large has
3,4
the potential to reduce global carbon dioxide emissions by 5% to 10%, provided AI is applied
much more pervasively and effectively than today.

Analytical and AI models are getting bigger and more complex, requiring ever more data and
data processing. This is a general trend that is most obvious in deep learning, including large
transformer or foundation models, such as the current GPT-3 or DALL-E 2 models.

If current D&A and AI practices remain unchanged, and with their ongoing adoption, the energy
needed for machine learning alone may grow to 3.5% of global electricity consumption by
2030. 5 This is more electricity than is used by the average European country. And because the
electricity used for ML is not completely renewable, ML’s growing energy consumption may still
offset carbon reductions elsewhere.

Implications:

Organizations are increasingly and proactively applying D&A and AI to monitor and benchmark
ESG goals and — more importantly — to improve sustainability. However, the use of D&A and AI
is not limited to or only driven by sustainability goals. Often they are primarily applied to
improve efficiency and reduce costs, which may indirectly lead to reduced energy usage and
greenhouse gas emissions.

D&A and AI practitioners are beginning to become more aware of their growing energy
footprint. Practices are emerging that propagate the use of specialized hardware, such as GPU
processors, instead of generic hardware, which significantly reduces energy consumption for
machine learning. Other practices include energy-efficient coding, transfer learning, enabling
the reuse of so-called pretrained models, and active learning, which enables more efficient
machine learning and reduces the amount of required training data. Still other approaches
include small data techniques, federated learning and composite AI, the latter leveraging more
human expertise rather than relying on a data-driven approach only.

Actions:

Prioritize AI initiatives that aim to optimize operations in general and — directly or indirectly —
boost sustainability to meet ESG goals. This applies to building management, logistics and
transportation, manufacturing and many other domains and functions in both public and private
sectors. In addition, foster the application of AI in “AI for good” initiatives, using the United
Nations’ sustainable development goals as a framework for prioritizing and monitoring
initiatives.

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 13/24
17/12/23, 12:32 Gartner Reprint

Raise awareness in D&A and AI teams about their own growing energy and environmental
footprint by fostering the use of metrics and observability to monitor energy consumption by
machine learning.

Adopt emerging practices to improve the energy efficiency of D&A and AI. These include the
prioritized use of (cloud) data centers that are powered by renewable energy, the use of more
energy efficient infrastructure, and the use of transfer learning, active learning, federated
learning, composite AI and other approaches. Their benefits include less need for (training)
data and more efficient machine learning, lower requirements for time, compute resources and
energy.

Further Reading:

Infographic: AI Use-Case Prism for Sustainability and ESG

Practical Data Fabric

Analysis by Thornton Craig, Robert Thanaraj, Ehtisham Zaidi, Mayank Talwar, Mark Beyer

Description:

Data fabric is a data management design pattern leveraging all types of metadata to observe,
analyze and recommend data pipeline actions. Organizations use data fabrics to capture data
assets, infer new relationships in datasets and automate actions on data. Practical data fabrics
leverage individual data domains or specific types of metadata to enable business use cases like
data sharing while creating a foundation for comprehensive, automated, future data pipeline
designs.

Why Trending:

Data management remains a challenge in complex modern environments, and data fabrics
offer solutions to manage and enhance data pipelines, especially when focusing on targeted
and practical use cases tied to business requirements.

Cost optimization is driven by data fabrics built on the foundations of balancing collect-and-
connect data management strategies. Existing storage layers and data systems can participate
by providing metadata to the fabric. By analyzing the metadata across all sources, the data
fabric provides insights on effective data design, delivery and utilization, thereby reducing
costs.

Data and analytics ecosystem capabilities like data lineage, data observability and data
catalogs are enabled by metadata-driven data fabrics. Master data management (MDM) and
data governance programs also integrate with and are enhanced by data fabrics in discovering
and managing critical data assets.

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 14/24
17/12/23, 12:32 Gartner Reprint

Implications:

Metadata analysis can expose hidden insights into business demand, metadata sharing can
speed up integration and decision making, and metadata can reinvent governance and reduce
risk. The existing data management systems, analytical platforms and systems of record are
mere participating systems in the data fabric design — they feed metadata to the data fabric.

The data fabric does not impose rip and replace of existing data management infrastructure. In
fact, it recommends connecting existing infrastructure (lakes, EDWs, marts, operational
systems) as “participating systems” that provide the metadata needed for analysis and active
metadata-based recommendations that provide optimal data pipeline design and delivery
recommendations to engineering teams.

By assembling and enriching the semantics of the underlying data, and by applying continuous
analytics over metadata, the data fabric generates alerts and recommendations that can be
actioned by both humans and systems.

Metadata analysis in an active fabric improves data pipeline management and automation
(DataOps) to identify and correct pipeline changes or data drift.

Actions:

Target practical data fabric opportunities by focusing on business use cases or specific types
of metadata captured. Look for use cases linked with other capabilities like identifying data,
governing data or master data.

Begin by monitoring how data is used, and leverage discovery tools to look for new and
unexpected uses of data. This might imply new opportunity or an emerging risk that warrants
some attention.

Initiate a pilot effort to launch a data fabric by identifying an intersection of data, consumption
use cases, users and systems performing the data management, and the affected business
units.

Don’t fall into the trap of looking for “data fabric tools or platforms.” Data fabric is a design that
should be enabled by architecture, best practices and multiple technologies. As of now, it is not
supported by one single tool or platform.

Further Reading:

Quick Answer: 5 Ways Data Fabric Design Supports Cost Optimization in Data Management

Quick Answer: What Is Data Fabric Design?

From Logical Data Warehouse to Data Fabric

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 15/24
17/12/23, 12:32 Gartner Reprint

Emergent AI
Analysis by Pieter den Hamer, Erick Brethenoux, Sumit Agarwal

Description:

AI has only just begun. Many trends and innovations continue to impact the way AI evolves and
how organizations can apply AI in practice. This trend relates to a next wave of AI, enabled by
emerging topics such as transformer models, decision intelligence, graph technology, simulations
and AI, composite AI, synthetic data, and AI engineering. Their common denominator is to bring AI
to the next level by creating more business impact and overcoming the limitations of current AI.
As a consequence, AI continues to improve significantly in terms of scalability, versatility and
adaptability.

Why Trending:

Many organizations find it challenging to scale and operationalize their AI initiatives. On

average, the time from prototype to production amounts to seven months. “AI engineering” is an
emerging discipline that helps to shorten time to value while also managing risks and enabling
responsible AI.

Transformer models such as GPT-3 (and ChatGPT), Dall E-2 and BERT are able to generate
texts, images and other content, making generative AI a reality, accessible to many
organizations and individuals. Moreover, transformer models and their human-like outputs are
greatly fueling further interest and investments in AI.

The currently dominant, fully data-driven deep learning approach to AI has its limitations.
Simulation and synthetic data help to overcome challenges related to data and privacy. More
profoundly, composite AI — including knowledge graphs and causal graphs — brings together
multiple AI techniques, creating a synergy between human-driven and data-driven learning and
intelligence.

Decision intelligence moves AI from insight-provider to levels of much greater impact, including
augmentation, automation and — ultimately — autonomous business.

Implications:

The next wave of AI will enable organizations to apply AI in situations where it is not feasible
today, often because of data or complexity challenges. This will broaden the envelope for AI
usage, making AI ever more pervasive and valuable.

AI will become less of an uncertain experiment and more of a robust engineering discipline,
lowering its adoption threshold and shortening time to value. Moreover, AI will be scalable
across silos, creating synergies between actors and units at enterprise and business
ecosystem levels.

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 16/24
17/12/23, 12:32 Gartner Reprint

AI — particularly generative AI combined with decision intelligence — will greatly impact how
knowledge workers are writing, how designers are creating and how managers are making
decisions. Many professions, roles and processes will change, even to the extent that
organizations need to go back to the drawing board when it comes to their structure, business
models, corporate identity and culture.

Actions:

(Re)assess opportunities for using AI within and across business areas, keeping in mind that AI
can now work with less data and deal with greater complexity than before.

Start adopting AI engineering practices, including a stronger collaboration between data

engineers (DataOps), AI experts (MLops/ModelOps) and software developers (DevOps), as well
as AI governance and responsible AI.

As AI “joins the team,” and impacts business roles, activities and processes, make sure that
change management and HR aspects are fully addressed in AI initiatives.

Further Reading:

Building a Digital Future: Emergent AI Trends

Quick Answer: What Are the Short-Term and Midterm Implications of ChatGPT for Data and
Analytics?

Converged and Composable Ecosystems

Analysis by Ramke Ramakrishnan, Julian Sun, Adam Ronthal, Donald Feinberg, Carlie Idoine

Description:

Converged D&A ecosystems are about designing and deploying the D&A platform to operate and
function cohesively through seamless integrations, governance and technical interoperability.
Assembling and deploying configurable applications and services delivers ecosystem
composability.

Further, this allows the D&A systems to be more modular, adaptable and flexible to scale
dynamically and more streamlined to continuously meet the growing and changing business
needs.

Why Trending:

Modern D&A ecosystems require modular and flexible deployment practices that can provide
simplicity and agility for the individual components to configure dynamically to scale and better
respond to new business demands and varying requirements.

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 17/24
17/12/23, 12:32 Gartner Reprint

Simplifying and cohesively eliminating multiple components through a centralized set of

services can be game changing and transformational for business users. Also, the challenges
of managing and governing D&A solutions that are increasingly distributed on multiple clouds
and even on-premises remnants have introduced a transformational and holistic approach to
their deployments.

Increasing use of low-code and no-code capabilities give business users the ability to create
more customized and packaged business analytics capabilities with business-centric
modularity and embed into applications via APIs to serve their business objectives.

Modular D&A ecosystems can drive innovation and support dynamic business systems and
applications. They can do so by breaking down the traditional, large and complex architecture
into composable elements to drive standardization and specific capabilities, including
autonomy, to make changes without interrupting other components with flexible governance.

Implications:

The degree of composability and convergence of technology components will become a major
differentiation for organizations when assessing analytics or deploying analytics applications.
This is because they will enable the business to proactively build, deploy and adapt to change
quicker than ever before.

D&A ecosystems should have building blocks to compose and recompose with convergence for
fast delivery of analytics applications in a composable enterprise. These building blocks should
help organizations to build their application with business-centric modularity, enabling a top-
down design with value streams of the organizations.

Increasing complexity, cost and effort of managing complex data and analytics platforms
propel a need to simplify the architecture and components and move toward a revived
approach for a cohesive D&A ecosystem through consolidation and modular deployment.

Actions:

Deliver a flexible and scalable D&A ecosystem that allows easier integration of different
technology components to adapt faster to changes and accelerate business growth with
increased efficiency

Operationalize your D&A platforms to eliminate complexity by providing simple and unified
access and management, improve efficiency and agility through interoperability, and facilitate
quicker deployment of new services to customers.

Enforce your technology architecture or adopt your vendor-based solution architecture to be

modular. Evolve your application into creating building blocks that can deliver capabilities to
drive specific business outcomes and are well-connected by APIs and services to support
broader enterprise needs.

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 18/24
17/12/23, 12:32 Gartner Reprint

Leverage your existing inventory of capabilities, processes and technologies to develop a

baseline, and enhance that baseline to to be more modular, reusable and provide greater agility
in support of new business needs at reduced costs

Further Reading:

Case Study: Composable Platform Strategy to Drive Business Agility (Nike)

Predicts 2022: Composable Applications Accelerate Digital Business

Use Gartner’s Reference Model to Deliver Intelligent Composable Business Applications

Consumers Become Creators

Analysis by Rita Sallam, David Pidsley, Georgia O’Callaghan, Julian Sun

Description:

To get more people to use analytics for more decisions, organizations must widen their focus
from enabling analysts, who are a small slice of the user base. Instead, they should give content
consumers across the organization the capabilities and skills they need to become content
creators. Achieving this elusive last mile of analytics impact will require expanded adoption of
capabilities such as automated insight generation, conversational experiences, dynamic data
stories, along with embedding insights in application workflows. It will also require an intense and
ongoing focus on people and how to facilitate data-centric behavioral change.

Why Trending:

Most user experience design has been focused on analysts, which make up 10% to 30% of the
user base in most organizations. However, mean adoption of analytics and business
intelligence still hovers around 30%. 6 This is because analyst-centric user experiences are too
difficult for less skilled business or operations people needing to ask their own questions or to
know which insights are the most important and actionable.

Most analytics consumers access insights in predefined dashboards that often overwhelm
users with data, if they can even find the dashboard with the data they need. Consumers must
then manually explore what happened, why and what to do about it, but they often don’t have
the skills to do this. Moreover, this manual approach is time consuming and error prone, and
this type of exploration can lead to incorrect conclusions and flawed decisions and actions.

There is a convergence of capabilities in the market between analytics and BI platforms and
data science, machine learning capabilities and natural language processing that is creating
new user experiences. Recent advances in generative AI and foundation models, such as
ChatGPT could have significant longer-term impacts on how analytics content is developed and
consumed.

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 19/24
17/12/23, 12:32 Gartner Reprint

The rise of data and analytics cloud ecosystems is enabling more seamless embeddability of
the insights from advanced analytics into the application workflows used by business and
operational workers that would not traditionally access analytics platforms.

Implications:

For more than 20 years, visual exploration-based dashboards and reports have been the
primary way users monitor and explore data. Now, the percentage of time users spend in
predefined dashboards will be replaced by conversational, dynamic and embedded user
experiences that address specific content consumers’ point-in-time needs.

This will shift advanced analytical power creation to the consumers, giving them content
creation and decisioning powers previously reserved for analysts and citizen data scientists.

Technological immaturity, organizational immaturity, low levels of data literacy, resistance to

change, lack of sufficient investment and incentives in change management, lack of trust in
perceived “black box” approaches, and concerns about the availability of and trustworthiness of
data will inhibit adoption and must be proactively managed.

The typical focus on short-term tools training will not be sufficient to motivate the behavioral
change necessary to fully realize business outcomes via improvements to processes and
decisions.

Actions:

Optimize value from data and analytics investments by putting more of an architectural focus
on the consumer role. Experiment with new augmented, conversational and embedded
experiences for consumers, who are, by far, the largest percentage of the user base.

Inspire users to make the behavioral changes necessary to leverage new capabilities to
optimize processes, activities and decisions by investing in communities, data literacy,
incentives and value storytelling. Make it clear what’s in it for them.

Maximize adoption and value realized from giving consumers new insight creation and decision
making powers by teaching them how to impact their specific business problems, decisions,
opportunities and processes.

Build trust in autogenerated insights and models by back testing and offering users scenario
planning and “what if” analysis to explore scenarios around autogenerated insight and looking
for explainability features in selected platforms.

Further Reading:

Market Guide for Augmented Analytics

Best Practices to Enable Effective Citizen Data Science

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 20/24
17/12/23, 12:32 Gartner Reprint

Humans Remain the Key Decision Makers

Analysis by Gareth Herschel, David Pidsley, Erick Brethenoux, Roy Schulte, Pieter den Hamer

Description:

Some of the most impressive examples of data and analytics such as algorithmic trading or self-
driving cars are highly automated. Developments in AI are opening up further opportunities to
automate decisions, thus replacing human expertise with synthetic experience. But not every
decision can or should be automated. The recognition that, at some level, every decision remains
a human decision is causing a renewed focus on decision support to enhance the decision
making of human experts. It is also causing a reevaluation of the role in decision augmentation
and decision automation to account for ethical concerns.

Why Trending:

AI systems increasingly have the capability to replace human decision makers, but are often
subject to strategic, ethical or logical blind spots that make their use dangerous (in some
cases, literally). The power to deploy advanced analytics “because we can” requires a balancing
human consideration of “but should we?”

The emphasis on new data and analytics initiatives that support “advanced” use cases misses
an opportunity to use the somewhat reviled “BI and reporting” to achieve significant benefits.
This can be done by working in conjunction with human experts to improve the mass of
strategic or tactical decisions (and even many operational decisions that are unsuitable for
automation).

Implications:

The most sophisticated analysis is not always the best.

Decision support is the most common use case for D&A, but the fact that D&A is not required in
the decision-making process means the burden for adoption falls on the D&A team. Attempts to
force adoption through decision automation or reeducation of users through data literacy
programs risk being a waste of effort at best or self-defeating at worst.

Efforts to drive decision automation without considering the human role in decisions will result
in a data-driven organization without conscience or consistent purpose.

Actions:

Do not limit decision intelligence efforts to those decisions that can be automated. Decision
intelligence is at least as valuable to decision support use cases.

Data literacy programs must include education about the appropriate way to combine data and
analytics with human decision making skills, not attempt to teach data to be a quantitative

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 21/24
17/12/23, 12:32 Gartner Reprint

substitute for human judgment. Data literacy programs that focus on enabling only “data
driven” decisions will not achieve their potential change management potential.

Further Reading:

Innovation Insight for Decision Intelligence Platforms

Use Multistructured Analytics for Complex Business Decisions

Changes Since Last Year
One noticeable change in the trends this year has been a focus on the “why” of D&A, rather than
just the how and what. The trends of value optimization and D&A sustainability must be
fundamental to the way many organizations approach their D&A strategy and relationship with
other cross-enterprise stakeholders.

Another interesting theme this year (referenced above) was the growth of trends around the role
of people in the D&A process, specifically the consumers become creators and humans remain
the decision makers trends.

Several trends identified this year echo those in previous years, including:

Practical data fabric is a reflection of the adoption pattern we have seen around metadata-
driven data fabric.

Data sharing is essential is a more robust statement than always share data. It reflects the
value of a more complete understanding of the environment.

Converged and composable ecosystems has been the result of convergence of two previous
trends: business composed D&A and vendor and regional ecosystems.

Although still a factor, some previous trends are “played out” in the sense that they are a received
fact rather than requiring an adjustment to organization’s strategic planning:

Context enriched analysis

Skills and literacy shortfall

Evidence
1
2022 Chief Data Officer Agenda Survey for 2023. This study was conducted to explore the
business impact of the CDAO role and/or the Office of the CDAO and understand the leadership
traits of the most successful CDAOs that distinguish them from their peers. The research was
conducted online from September through November 2022 among 566 respondents from across
the world. Respondents were required to be the highest-level data and analytics leader in the
organization: chief data officer, chief analytics officer, the most senior leader in IT with data and
analytics responsibilities, or a business executive such as chief digital officer, or other business

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 22/24
17/12/23, 12:32 Gartner Reprint

executive with data and analytics responsibilities. The survey sample was gleaned from a variety
of sources (including LinkedIn), with the greatest number coming from a Gartner-curated list of
over 4,499 CDOs and other high-level data and analytics leaders.

Disclaimer: Results of this study do not represent global findings or the market as a whole but
reflect sentiment of the respondents and companies surveyed.

2
Innovation Insight for Decision Intelligence Platforms

3
Reduce Carbon and Costs With the Power of AI, Boston Consulting Group.

4
Green Intelligence — AI Could Boost Efforts to Fight Climate Change, Economist Impact

5
AI Can Do Great Things — If It Doesn’t Burn the Planet, Wired.

6
R-21523 2021 Gartner Analytics Consumerization-Democratization Survey. This survey sought
to understand the anticipated shift from the current paradigm of modern BI (visual-based
discovery and self-service) to augmented analytics consumers (natural language query, data
storytelling, automated insights and natural language generation). The survey also sought to
understand the importance of ease of use as a major barrier or enabler to this shift. The research
was conducted online from 3 November to 15 November 2021 among 62 Gartner Research Circle
members and 59 external participants. Qualified participants were actively involved in analytics
and business intelligence (ABI) within their organizations. ABI professionals from North America
(55), EMEA (53), Latin America (8) and Asia/Pacific (5) responded to the survey. Disclaimer:
Results of this survey do not represent global findings or the market as a whole, but reflect the
sentiments of the respondents and companies surveyed.

© 2023 Gartner, Inc. and/or its affiliates. All rights reserved. Gartner is a registered trademark of Gartner, Inc.
and its affiliates. This publication may not be reproduced or distributed in any form without Gartner's prior
written permission. It consists of the opinions of Gartner's research organization, which should not be construed
as statements of fact. While the information contained in this publication has been obtained from sources
believed to be reliable, Gartner disclaims all warranties as to the accuracy, completeness or adequacy of such
information. Although Gartner research may address legal and financial issues, Gartner does not provide legal
or investment advice and its research should not be construed or used as such. Your access and use of this
publication are governed by Gartner’s Usage Policy. Gartner prides itself on its reputation for independence and
https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 23/24
17/12/23, 12:32 Gartner Reprint

objectivity. Its research is produced independently by its research organization without input or influence from
any third party. For further information, see "Guiding Principles on Independence and Objectivity." Gartner
research may not be used as input into or for the training or development of generative artificial intelligence,
machine learning, algorithms, software, or related technologies.

About Careers Newsroom Policies Site Index IT Glossary Gartner Blog Network Contact Send
Feedback

© 2023 Gartner, Inc. and/or its Affiliates. All Rights Reserved.

https://www.gartner.com/doc/reprints?id=1-2D5XJT8I&ct=230407&st=sb 24/24