2020

TDWI Analytics Maturity Model

Assessment Guide

By Fern Halper
Research Sponsors

Research Sponsors
TDWI RESEARCH

2020

TDWI Analytics Table of Contents

Foreword from the Author . . . . . . . . . . . . . . . . . . . . . . 3
Maturity Model Value of the Model . . . . . . . . . . . . . . . . . . . . . . . . . 3
Assessment Guide Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Trends in Analytics . . . . . . . . . . . . . . . . . . . . . . . . . 5
TDWI Analytics Maturity Model: The Context for
Benchmark Scores . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Model Dimensions . . . . . . . . . . . . . . . . . . . . . . . . . 7
Stages of Maturity . . . . . . . . . . . . . . . . . . . . . . . . . 8
Overview of the Stages . . . . . . . . . . . . . . . . . . . . . . . . 9
Stage 1: Nascent . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Stage 2: Early . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Stage 3: Established . . . . . . . . . . . . . . . . . . . . . . . . 11
The Chasm . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Stage 4: Mature . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Stage 5: Advanced/Visionary . . . . . . . . . . . . . . . . . . . 14
Evaluating Benchmark Scores . . . . . . . . . . . . . . . . . . . 16
Scoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Research Sponsors . . . . . . . . . . . . . . . . . . . . . . . . . 18

© 2020 by TDWI, a division of 1105 Media, Inc. All rights reserved.

Reproductions in whole or in part are prohibited except by written permission.
Email requests or feedback to [email protected]. Product and company names
mentioned herein may be trademarks and/or registered trademarks of their
respective companies.

This report is based on independent research and represents TDWI’s findings;

reader experience may differ. The information contained in this report was
obtained from sources believed to be reliable at the time of publication.
Features and specifications can and do change frequently; readers are
encouraged to visit vendor websites for updated information. TDWI shall not be
liable for any omissions or errors in the information in this report.

tdwi.org  1
TDWI Analytics Maturity Model Assessment Guide

About the Author

FERN HALPER, PH.D., is VP and senior director of TDWI Research for advanced analytics, focusing
on predictive analytics, social media analysis, text analytics, cloud computing, and other “big data”
analytics approaches. She has more than 20 years of experience in data and business analysis, and
has published numerous articles on data mining and information technology. Halper is co-author
of “Dummies” books on cloud computing, hybrid cloud, service-oriented architecture, and service
management, and Big Data for Dummies. She has been a partner at industry analyst firm Hurwitz &
Associates and a lead analyst for Bell Labs. Her Ph.D. is from Texas A&M University. You can reach
her at [email protected], @fhalper on Twitter, and on LinkedIn at
https://www.linkedin.com/in/fbhalper/.

About TDWI Research

TDWI Research provides research and advice for data professionals worldwide. TDWI Research
focuses exclusively on data management and analytics issues and teams up with industry thought
leaders and practitioners to deliver both broad and deep understanding of the business and technical
challenges surrounding the deployment and use of data management and analytics solutions. TDWI
Research offers in-depth research reports, commentary, inquiry services, and topical conferences as
well as strategic planning services to user and vendor organizations.

Sponsor
Amazon Web Services, Matillion, Snowflake, and Tableau sponsored the research for this TDWI
Analytics Maturity Model Assessment Guide and its accompanying Interactive Assessment Tool.

2  TDWI RESEARCH
 Foreword

Foreword from the Author

Today, many organizations realize that analytics can provide an important competitive advantage.
These organizations want to move forward to evolve their analytics strategies beyond spreadsheets
and dashboards; many seek to build self-service capabilities to democratize analytics. Still others
are moving to utilize machine learning and other advanced forms of analytics such as natural
language processing (NLP). TDWI expects interest will continue to build for analytics technologies
and methods, particularly as organizations include less traditional users as part of their growing
analytics culture.
Of course, analytics maturity is not just about analytics techniques and software. It includes the
data management to support disparate kinds of data from various internal and external sources.
For example, is your infrastructure scalable to support increasing amounts of diverse and disparate
data? Maturity also involves the cultural and organizational processes that enable companies to
become more data-driven. This includes development techniques as well as the current processes
that allow a wide range of people throughout the organization to manage, govern, and utilize the
data and analysis.
TDWI created the Analytics Maturity Model and Assessment in response to requests from
organizations to understand how their analytics deployments compare to those of their peers in order
to provide best-in-class insight and support. The assessment measures the maturity of an analytics
program in an objective way across five dimensions that are key to deriving value from analytics.

Value of the Model

The Analytics Maturity Model can help guide business and IT professionals on their analytics
journey. It provides a framework for companies to understand where they’ve been, where they
are, and where they still need to go with their analytics deployments. The model can also provide
guidance for companies at the beginning of their analytics process by helping them understand best
practices used by companies that are more mature in their deployments.
A great feature of the Analytics Maturity Model is the interactive benchmark assessment. At the end
of the survey, you will be able to measure the maturity of your deployment in an objective way and
compare your maturity level to that of your peers. This TDWI Assessment Guide is designed to help
you understand the phases of maturity in analytics as well as help you interpret your benchmarking
scores. I trust you will find it useful.

Fern Halper,
TDWI VP Research, Sr. Director Research for
Advanced Analytics

tdwi.org  3
TDWI Analytics Maturity Model Assessment Guide

Introduction
Increasingly, organizations are looking to gain competitive advantage using data and analytics. In a
recent TDWI survey, for example, 90 percent of respondents agreed that they will need to undergo
some sort of digital transformation to compete in the future. Digital transformations change how
organizations do business, and data and analytics are at the heart of the transformation. Analytics
can provide tangible benefits such as deeper insights, increased productivity, improved customer
service, improved operational efficiencies that drive cost savings, and top-line growth that delivers
higher profits. For innovative companies, analytics can drive new products and services.
TDWI sees organizations analyzing data to understand customer behavior, improve operational
efficiencies, and identify fraud as well as for a host of other use cases. To do this, companies are
using traditional query and reporting tools as well as newer approaches such as machine learning.
Many organizations are also looking to visual data discovery approaches to find insights within their
data. Self-service is a top priority among companies. Additionally, as organizations look to analyze
new data, they realize that to succeed they need a scalable, flexible data architecture to support data
management, data access, data governance, and data security.
Not surprisingly, the demand for analytics continues to increase as its value becomes clear. At TDWI,
we see that those organizations that become more sophisticated with analytics also tend to derive
more value from it. In fact, the more mature the organization is in terms of analytics, the more likely
it is to measure top- or bottom-line impacts. This is part of a virtuous cycle where, as organizations
begin to reap the benefits of analytics, they tend to put more advanced analytics in place. The success
builds on itself.

DEFINITIONS

ANALYTICS. DATA INTEGRATION. DATA PLATFORMS.

When people think about A key to analytics success Data platforms are
analytics, they generally is data integration, and in where data is ingested
consider a range of this assessment we define and persisted to be
techniques. These run the it as the collection of provisioned for a wide
gamut from spreadsheets technologies that capture range of applications and
and dashboards to more and prepare data for operations, including
advanced techniques analysis and operations. relational DBMSs,
such as self-service data Data integration includes columnar DBMSs,
discovery and machine such technologies as analytics DBMSs, in-
learning. For the purpose data preparation, data memory databases,
of the maturity model, quality, and metadata NoSQL, Hadoop (and
we refer to the complete management. other open source
spectrum as analytics. technologies), and
storage subsystems.

4  TDWI RESEARCH
 Introduction

Trends in Analytics
There are several interconnected technological and organizational trends that are relevant to
companies looking to become more mature in their analytics efforts. These important trends include:
SELF-SERVICE ACROSS THE ANALYTICS LIFE CYCLE . An ongoing trend in analytics is the move to make
it available to more people in the organization. In fact, self-service—using solutions that enable many
types of nontechnical users to be productive because they are easier to use, do not require coding,
and do not require IT to set up all data access, queries, visualizations, and preparation—is a top
priority among organizations that we survey.
In this definition, self-service isn’t just about the actual analysis portion of the analytics life cycle;
more organizations want self-service functionality across the analytics life cycle. Self-service provides
business analysts and users with more flexibility and agility to build pipelines, access data, and derive
insights, which frees up the data experts to deal with the complexities of an increasingly modern
environment. In terms of maturity, this often requires a balance between business and IT and
building a partnership to ensure trustworthy data.
ARTIFICIAL INTELLIGENCE (AI) AND OTHER MODERN ANALYTICS. AI is the phrase du jour and market
hype around it is soaring. Two key technologies under the AI umbrella include machine learning
(ML) and natural language processing (NLP). Machine learning, although not new (early algorithms
were developed more than 20 years ago), provides a way for computers to perform tasks without
being explicitly programmed. This includes learning and identifying patterns across vast amounts of
diverse data which enables insights, classification, and prediction. Machine learning is fueling new
use cases such as image classification and diagnostics.
NLP involves analyzing, understanding, and generating responses that enable interfacing with
systems using human rather than computing languages. NLP, which can also utilize machine
learning, is being used in chatbots as well as personal assistants and to understand customer
sentiment and intentions. Many organizations are on the cusp of utilizing AI technologies.
AUGMENTED INTELLIGENCE. Tied to the previous trends is the move to provide easy-to-use interfaces
in analytics. A recent trend is to embed advanced analytics such as machine learning or NLP into
the software itself to automate analysis. For instance, some tools provide automated insights that are
available when the user logs in to the system. Some analytics vendors offer tools that help business
analysts (and even many business users) construct machine learning models. Here, all the user needs
to do is to specify the outcome or target variable of interest along with the attributes believed to be
predictive. The software picks the best model. Other tools are more automated.
Many view these tools as a way to put more complex analytics, such as machine learning, in the
hands of business analysts. This is often referred to as “augmented intelligence” or “smart tooling”
and is one way organizations are planning to upgrade the skills of their business analysts to do (at
least some of) the model building. These tools can also help data scientists be more productive. At
TDWI, we see more organizations interested in using these tools, but a mature organization will put
the right processes in place to make sure these tools are used properly.
NEW ROLES SUCH AS MLOPS. As organizations scale their analytics efforts, they are often hiring more
data engineers to develop complex data pipelines. Likewise, as machine learning models go into
production, organizations see the need for a new role—that of MLOps. This team is responsible
for deploying models into production and monitoring them thereafter. The MLOps team can be
overlooked, but successful organizations typically have an MLOps team to help validate models and
put them (and other AI apps) into production. MLOps teams are often skilled in newer open source

tdwi.org  5
TDWI Analytics Maturity Model Assessment Guide

technologies such as Spark and Python, as well as in understanding the output of models built using
commercial tools.
OPEN SOURCE. Open source has become quite popular for analytics because it is a low-cost
source community for innovation. This appeals to many data scientists and analytics application
developers—especially those who like to code. In fact, TDWI research indicates that organizations
are looking to open source projects such as R and Python to be a big component of their analytics
strategies, especially for data science initiatives. Many analytics vendors already incorporate support
for R and Python into their packages.
THE CLOUD. The cloud has become a major platform for data and analytics. At TDWI, we see
organizations moving more of their data management to the cloud. Cloud data warehouses and data
lakes are becoming increasingly popular.
Organizations like the cloud because it provides scalability, elasticity, and flexibility. When you need
to perform analytics processing on a large data set, the cloud enables you to procure as much storage
and compute services as necessary. When you are finished with the analysis, you can release these
services so you are no longer paying for more capacity than you need. Likewise, with the virtual
resources of the cloud, incorporating new data sources and setting up analytics sandboxes or data
labs in a cloud data warehouse environment can be fast and easy.
One newer trend is the use of multicloud deployments. Multicloud is where organizations use
multiple cloud providers (two or more cloud service providers) to provide infrastructure, applications
and key business functions. It’s basically a services model. Instead of a business using one vendor
for everything (cloud hosting, storage, and the full application stack), in a multicloud configuration,
enterprises use several. At TDWI, we typically see more mature organizations utilizing the cloud for
data management as part of a multiplatform strategy.
UNIFYING DATA PLATFORMS. As organizations collect and analyze increasing amounts of diverse data
from disparate sources, they often want to unify that data for trusted analytics. There are numerous
approaches. Some organizations move their data to a cloud data warehouse for analytics. Others look
to a data fabric that might be enabled by data virtualization. Still others look for a unified technology
stack that includes data management (and ETL) and analytics. The key is to develop a high-quality
source of data. As part of this, more organizations are also looking to technologies such as data
catalogs to help access trusted data.

6  TDWI RESEARCH
 The Context for Benchmark Scores

TDWI Analytics Maturity Model: The Context for

Benchmark Scores
Analytics maturity can be described as the evolution of an organization to integrate, manage, and
leverage all relevant internal and external data sources into key decision points. It means creating
an ecosystem that enables insight and action. In other words, analytics maturity is not simply
about having some technology in place; it involves technologies, resources, data management,
governance, and organizational components. It can take years to deploy and instill an analytics
culture in an organization.

Model Dimensions

Organizational Resource Data Infrastructure Analytics Governance

Maturity Maturity Maturity Maturity Maturity

• Leadership • Funding • Diversity, volume, and • Scope of capabilities • Data governance

• Culture • Talent/skills speed • Automation/ processes and tooling
• Impact • Roles/responsibilities • Data access augmented • Model governance
• Data integration and • Deployment and processes and tooling
• Strategy • Training
management delivery approaches • Governance roles
• Data architecture • Innovation • Security/privacy

Figure 1. TDWI Analytics Maturity Model dimensions.

The TDWI Analytics Maturity Model Assessment is made up of 52 questions across the five
categories that form the dimensions of the TDWI Analytics Maturity Model (see Figure 1). These
dimensions are:
• ORGANIZATION. Organizational factors can often make or break an analytics effort. To what
extent does the organizational strategy, culture, and leadership support and evangelize a
successful analytics program? Are analytics used for decision making across the organization?
What percent of teams use analytics for day-to-day decision making? Is it easy to deploy
analytics into production? Is there a culture of trust in the analytics? A willingness to try new
things? A unified strategy?
• RESOURCES. Performing analytics requires skills and budget. How are analytics projects funded?
How is adoption facilitated? How does the organization grow talent and build skills? How does
it organize to execute? Does it have the talent and skills needed for modern analytics including
data literacy, model building and deployment, and data engineering? Is it struggling to maintain
the infrastructure? Working around the clock to maintain it?
• DATA INFRASTRUCTURE. Data is a key component of any analytics initiative and more
advanced organizations often utilize more diverse data types and volumes. What kind of
data does the organization use for analysis? High volume? High frequency? High integrity?
Is it accessible? How is the data integrated? Is the architecture distributed? Utilizing cloud
services? Is the architecture coherent? Can the infrastructure handle the growth in users? Can
it scale by use case/seasonality? Can it scale on demand to fit users’ needs? Is it monolithic or
microservices-based?

tdwi.org  7
TDWI Analytics Maturity Model Assessment Guide

• ANALYTICS. What is the scope of your analytics? This includes the kinds of analytics utilized
(e.g., machine learning, real-time analytics, etc.) and how analytics results are delivered in the
organization. It also includes the percent of people with access to analytics. What percent of
people in the organization perform their own analytics? Are analytics embedded into business
processes? Are predictive analytics or machine learning models built? Deployed into production?
Are automated analytics part of the picture? Is the company innovating with analytics?
• GOVERNANCE. How coherent is your company’s data governance strategy in support of
your analytics program? Is the company able to balance users’ data discovery and analytics
explorations against IT processes and policies effectively? Is there collaboration between business
and IT? Is the organization able to ensure the proper access and use of data? Are advanced
models governed? What guardrails and controls are put in place to ensure that “correct” models
are deployed into production? How are they tracked? Are the right roles in place from both
business and IT? Is the organization utilizing modern tooling such as data catalogs to help
build trusted data assets? Are security and privacy measures deployed for personally identifiable
information and other sensitive data types?

Stages of Maturity
The TDWI Analytics Maturity Model consists of five stages: Nascent, Early, Established, Mature
and Advanced/Visionary. As organizations move through these stages, they should be gaining more
value from their analytics investments. Figure 2 illustrates these stages.

Figure 2. Stages of maturity in the TDWI Analytics Maturity Model

This Guide provides an overview of each of the stages of the TDWI Analytics Maturity Model. These
descriptions provide context to use to interpret your scores when you take the assessment.

8  TDWI RESEARCH
 Overview of the Stages

Overview of the Stages

Stage 1: Nascent

The Nascent stage represents a pre–analytics environment. In this stage, most companies are not
utilizing analytics except perhaps for spreadsheets. There is no real support for analytics, although
there are pockets of people spread throughout the company who may be interested in the potential
value of analytics and who may be trying out some analytics software. Generally, in the Nascent
stage, the culture is not analytics-friendly—that is, the culture is not data-driven and decisions are
made based on gut instinct rather than facts.
ORGANIZATION. In the organization at the Nascent stage, most executives are unaware of (or don’t
care about) the power of analytics to help drive action. IT and the business typically do not work
together to make analytics happen, although either group may want to. Often, IT is not necessarily
interested in helping the business get access to data unless it is under IT control because they believe
they know best. Business may have developed its own systems for various activities that might be
useful for analysis, but the analyst is typically a spreadsheet super user. There is no strategy to move
forward because the business doesn’t care enough to do so. In this spreadsheet culture, there may be
pockets of people who are trying to get educated on newer kinds of analytics because they know the
old way won’t help them remain competitive.
DATA INFRASTRUCTURE. At this stage, the typical Nascent organization does not have a sound data
management strategy in place. IT may believe that it has some sort of data management strategy—
perhaps with some databases or data sets that they use for reporting—but they haven’t really thought
about how the company data could support analytics. Data quality and consistency may be poor.
Data volumes are typically low or in disconnected silos, and any data is generally structured data
only. On the business side, organizations are asking IT for data or reports, or if they have their own
data sources for a business activity, they are typically siloed for analysis, which is done by bringing
spreadsheets together. It can be a long undertaking to assemble a data set for analysis.
ANALYTICS. In the Nascent organization, the spreadsheet typically rules. Most of the analysis is
slicing and dicing. Dashboards may be developed in a spreadsheet and emailed to those “who need
to know.” If the company has an analytics group, it is generally within a department or line of
business and targeted at a specific function such as marketing. In other words, analytics is occurring
in pockets and silos in the organization, and one group does not know what the other is doing.
Best practices are not shared and there is no effort at creating a broader analytics culture involving
nontraditional users of analytics. However, in many Nascent organizations, employees are getting
tired of not having access to data to make decisions. They want analytics tools and access to trusted
data so they can answer important business questions.

tdwi.org  9
TDWI Analytics Maturity Model Assessment Guide

Stage 2: Early

As the company moves out of the Nascent stage, it is starting to do its analytics homework. Staff
may be reading up on the topic and perhaps attending webinars or conferences. One or more
organizations may have invested in some analytics technology such as single instances of self-service
data discovery and visualization tooling, a database, a data mart, or a data warehouse for managed
reporting. People are starting to understand the power of analysis for improving decisions and
ultimately business outcomes. Key characteristics of the Early stage organization include:
ORGANIZATION. In the Early stage, an executive sponsor may step up to the plate to try to drive the
analytics discussion in the company. S/he may be frustrated because decisions are being made in the
absence of data. Or, the executive comes from a company that is more data-driven. In this stage, one
business unit might be starting to use commercially available tools or open source tooling to do some
analysis. Other groups will see the possibilities and might start to get on board. Analytics discussions
are beginning, although a culture around analytics has not yet formed.
DATA INFRASTRUCTURE. Often, in the Early maturity stage, business and/or IT leaders are
envisioning the potential value of combining multiple sources of data for analysis. The company
has realized it needs some sort of data infrastructure to support its efforts. It may be that too many
errors have been made with the existing data set or someone is pointing out the errors in the data.
Whatever the reason, at this stage the company is struggling, so it may have some sort of data mart
or data warehouse on premises. Data is still primarily structured and volumes are low (e.g., less
than a terabyte).
ANALYTICS. In this stage of maturity, reports and dashboards rule. Analytics are still rudimentary,
but pockets of advancement are starting to occur. Business analysts are realizing the power of self-
service visualization, for example, and may have invested in visualization tools to understand their
data. Marketing or other departments are starting to ask questions that need answers. Some parts of
the organization may have hired a few data scientists to start to look at specific problems that require
more advanced analytics, such as predictive analytics or machine learning (PAML).

10  TDWI RESEARCH
 Overview of the Stages

Stage 3: Established

During the Established phase, the company is putting analytics tools and methodologies in place.
The enterprise typically has a data warehouse and is utilizing it for reporting or dashboard needs. The
company has moved toward self-service but is not necessarily data literate. IT may own the data and
the business owns the analytics. The two groups may start working together.
ORGANIZATION. In this stage, the company is putting a solid analytics strategy in place. IT and the
business may be working together as a team on selected projects. Often, this is because an executive
or someone with clout has realized that IT is not helping the business and vice versa. There may be
a business crisis or changes detected in customer preferences and behavior that are driving this. The
team is working to determine the right business problems to solve and how to incorporate analytics
into the decision-making process. Analytics projects led by individuals or single departments in the
company could be bearing fruit, and there is interest in building prototypes to see how analytics
could be advanced and extended further across the enterprise. As the organization gets excited about
the prospects for how analytics could improve decisions and outcomes more broadly, more people
start to come on board.
RESOURCES. At this stage, the organization most likely has formed a group that is responsible for
analytics in IT. This may be in a center of excellence (CoE) and marks the start of setting up data
literacy programs to help business users. Additionally, the organization is starting to hire more data
scientists and data engineers to help with more advanced complicated analytics. A virtuous cycle
starts to take hold.
DATA INFRASTRUCTURE. During the Established maturity stage, the organization typically has some
sort of data warehouse or data mart(s) in place—typically still for structured data only, although
it is looking at other platforms for new kinds of data. The company may be considering or even
evaluating proofs of concept for deploying data warehouses or data lakes in the cloud to increase
flexibility and reduce costs for some of its analytics initiatives. It understands that its on-premises
data warehouse probably cannot support more complex analytics. It is moving toward a more hybrid
architecture with more data (perhaps 1-2 TB) that it needs to manage.
ANALYTICS. In this stage, organizations are utilizing reports, dashboards, and some kind of self-
service tools that enable data visualization. Typically, at this stage the organization is thinking
about making the move to utilize predictive analytics more widely for use cases in IT, operations, or
marketing. Skills start to become an issue here because there is substantial handholding involved in
self-service and more data scientists are needed to perform more advanced analytics.
GOVERNANCE. At this point along the maturity path, a data governance team has formed but there
are no real guardrails for analytics governance. Some organizations might have a steering committee
overseeing data governance, with department representatives providing progress reports and
compliance updates. However, most do not.

tdwi.org  11
TDWI Analytics Maturity Model Assessment Guide

The Chasm

As organizations try to move from the Established stage to become more mature, they must
overcome a series of typical hurdles. There is the obvious challenge of obtaining the right skill
set. There may also be political issues. For example, one organization may have been driving the
company’s analytics effort and has encouraged other departments to join in. However, when it comes
time to extend the platform or put more stringent standards and governance in place, departments
begin to fight for control over who owns the data. They may quarrel about which vision should be
followed. Often times, analytics isn’t available in a way that encourages adoption.
To successfully cross the chasm, you’ll need to address the following challenges:
ORGANIZATION. Often it is the cultural and political issues that can stop analytics from becoming
more pervasive throughout the organization. For example, a company trying to operationalize
analytics as part of a business process (which is a sign of maturity) might hit a roadblock.
Operationalizing analytics—especially embedding it so that others can consume it downstream—
takes time because users must build trust first. It can take time for even the concept of self-service to
permeate through an organization because of cultural issues. If a company does not have a culture of
making decisions based on data, workers must adapt to new ways of doing things.
RESOURCES. Many early analytics projects are driven by a visionary executive champion. Of course,
it is critical to establish wins with these projects in order to secure funding. This funding needs to be
both IT- and business-driven. Business involvement is needed because analytics projects must have
business value with tangible business outcomes.
A big barrier for analytics projects moving past the Chasm stage is skills development for new
technologies such as self-service, NoSQL databases, or more advanced analytics. The skill issue
comes up most often as a barrier to moving forward with analytics. If the company can afford it, it
will hire staff to fill the skills gap. Many companies are using a combination of approaches such as
hiring from the outside as well as training from within. Some organizations are utilizing internal user
groups to help build skill sets.
DATA INFRASTRUCTURE. To become more mature, the enterprise must develop a solid data
management plan. Data management must adjust to different and often dynamic requirements,
data availability rhythms, and performance needed to support “democratized” self-service analytics.
Additionally, more advanced analytics often requires a platform where data scientists can iterate
and build models. This may involve new data types. This is often quite different from standard and
heavily scheduled BI reporting.
Data must be shared across the organization, and even more important, it must be trusted. That
means that companies will need to build toward a more unified architecture to provide data access
and ensure that data integrity for analytics is in place.
Companies will also need to ensure that they can scale their data management efforts. Many look
to the cloud to do this, utilizing cloud data warehouses or data lakes. It is key that any data lake an
organization deploys not become a data swamp. That means governance (see below) is critical.

12  TDWI RESEARCH
 Overview of the Stages

GOVERNANCE. To get across the chasm, organizations must implement strong data governance
capabilities. This includes plans to govern a multiplatform environment that includes on-premises
and cloud deployments of data warehouses and data lakes. Metadata, data lineage, and data catalogs
are often part of this plan.
Becoming more mature will also require the organization to consider an analytics governance team
that will deal with issues such as model management, deployment, and monitoring.
To cross the chasm, companies need to ensure that the right governance, data architecture, data life
cycle management security strategies, and organizational structures are in place. In our observations
to date, we have seen many organizations spend extra time in the Established stage as they near the
Chasm stage and require additional staffing to move to the next stage.

Stage 4: Mature

In an analytically mature organization, end users typically get involved, and the analytics transforms
how they do business. For instance, users may change how decisions are made by operationalizing
analytics in the organization. They will be using disparate kinds of data and even data that is semi-
or unstructured for their analytics efforts.
For most organizations to reach this stage of maturity, they might have repeatedly addressed certain
gaps in organization, infrastructure, data management, analytics, and governance.
ORGANIZATION. At this stage, the company realizes that analytics is a competitive differentiator and
has some successful projects in production. Innovation around data and analysis is a core value, and
an analytics culture prevails. The business strategy is generally a blend of top down and bottom
up, with a data infrastructure that can support this. Business and IT view themselves as a team and
have experience working together successfully on analytics projects. A growing analytics culture and
mindset pervades the organization. These organizations often think outside the box when expanding
the reach of analytics, such as by holding competitions or other events. Importantly, the organization
acts on its insights.
RESOURCES. A company at the Mature stage typically has IT and analytics teams that include data
scientists, data engineers, and MLOps (for putting models into production). There is budget to
support these roles. The organization at this stage might even have a chief analytics officer or a chief
data officer. Here, the virtuous cycle has taken hold.
DATA INFRASTRUCTURE. The more mature a company is, the better it is able to manage and make
use of its data specifically for advanced analytics, not just BI reporting. In this stage of maturity,
organizations can use many forms of data and absorb new sources of data as they emerge. The
company will most likely have a data warehouse in the cloud and possibly a data lake. Data
integration supports modern tooling. There is a coherent architecture to support data access by
multiple personas and the company is able to scale its data management infrastructure to meet its
requirements.
ANALYTICS. Any company might collect a lot of data but not make use of it—it’s a sign of maturity
when new data coming into the organization can be quickly analyzed and incorporated into the

tdwi.org  13
TDWI Analytics Maturity Model Assessment Guide

logical infrastructure. In this stage, analytics supports the organization. Typically, a company at this
juncture also has a center of excellence (CoE) that serves different parts of the organization. The
CoE includes the data science team, which might even train other groups in the use of analytics in
different forms.
At this point, analytics might be operationalized as part of a business process. In other words,
analytics might be automated or integrated with the business process. Different kinds of data are
analyzed, including unstructured data and geospatial data. Mature companies are typically using
new technologies such as machine learning and NLP. Multiple models are in production.
GOVERNANCE. A company at this stage of maturity understands that analytics, for all its benefits,
can be a liability waiting to happen, especially if data from external sources is part of the equation.
A company at this stage should be concerned with answering questions such as Whose data was it?
Whose data is it? Where did it come from? At this stage of maturity, organizations will have program
governance, with program management office (PMO) guidance for the program and a steering
committee that oversees the program from a companywide perspective. Tools such as data catalogs
are used. Additionally, a company at this stage of maturity will have some sort of model governance as
well as data governance to be able to version, validate, monitor, and update PAML models.

Stage 5: Advanced/Visionary

Only a small percent of companies currently have visionary analytics. At this stage, organizations
are executing analytics programs against a highly tuned infrastructure with well-established data
governance strategies. Well-governed but flexible data access is available for users so they can explore
data and develop visualizations in a self-service fashion and are not completely dependent on IT.
Many programs are executed as budgeted and planned initiatives from the company perspective. In
the Visionary stage, there is excitement and energy around analytics and a healthy and agile analytics
culture enables nontraditional users at middle management and even frontline positions to benefit.
ORGANIZATION . Companies at the Visionary stage share several characteristics. First, executives
have bought into analytics and view it as critical. Analytics is seen as a competitive weapon. Second,
analytics is not simply used to drive strategy or insight; instead, the businesses are always looking for
opportunities to use analytics in new ways. These companies realize that analytics is becoming more
common and they need to be thinking outside the box and looking for monetization opportunities.
Analytics is a way of daily life with these organizations, whether they are consuming the analytics or
developing it.
DATA INFRASTRUCTURE. The way you manage complexity is key to analytics maturity. The Visionary
company has deployed a coherent data infrastructure that is fully operational and can be used in the
mission-critical aspects of the business. Part of the infrastructure includes the ability to integrate new
sources of data for analytics, whether they are internal or external to the company. The infrastructure
uses what has worked in the past, including data warehousing, and leverages the cloud.
ANALYTICS. In the Visionary company, the mindset is to continually develop analytics. Typically, this
kind of company uses all kinds of data, including unstructured data and real-time data, for decision
making and for incorporation into business processes. The Visionary company can connect the dots

14  TDWI RESEARCH
 Overview of the Stages

between new data and existing assets. CoEs exist and are paying off. Teams are delivering new and
exciting forms of analytics. Some Visionary organizations build a team of business and IT staff to
explore technological innovations, test them for the business, and put the more useful ones into
production. Machine learning, deep learning, NLP, and other AI technologies are commonly in use
and in production. In fact, the Visionary company may have hundreds or even thousands of models
in production. These models are versioned, managed, and monitored.

tdwi.org  15
TDWI Analytics Maturity Model Assessment Guide

Evaluating Benchmark Scores

The TDWI Analytics Maturity Assessment has 52 questions across the five categories that form the
dimensions of the TDWI Analytics Maturity Model.
These dimensions should now seem familiar because they are the same categories we have been
referencing throughout this guide. These factors, and others, are used to explore relationships in the
data in order to help determine best practices for analytics.
Of course, organizations can be at different stages of maturity in each of these five categories,
and most are.

Scoring
The questions are either provided singly or grouped together in a matrix. Questions may be weighted
differently depending on their relative importance. Each dimension has a potential maximum score
of 20 points. Because organizations can be at different levels of maturity in the five dimensions, we
score each section separately as well as provide an overall score. The output of the assessment is a
score in each dimension and a total score.

Interpretation
Once you complete the survey, a report-based interface will show how your responses compare to
those of your peers. The breakdown of scores for each dimension is as follows:

SCORE PER DIMENSION STAGE

≤5 Nascent

6-10 Early

11-15 Established

16-19 Mature

20 Visionary

For instance, if you receive a score of 11 in the Organization dimension of the assessment, you are in
the Established stage for that particular dimension. You should expect to see your scores vary for the
different dimensions. Analytics programs don’t necessarily evolve at the same rate across all of the
dimensions. For example, your company might be more advanced in terms of bringing data sources
together than it is in analyzing them or governing this data.
The Chasm can be overlaid between the Established and Mature stages.
When you complete the assessment, you might see scores like this:

DIMENSION SCORE STAGE

Organization 10 Early

Resources 7 Early

Data Infrastructure 11 Established

Analytics 4 Nascent

Governance 7 Early

Total Score: 39
This means that you are more mature in your data infrastructure but less mature in the other areas.

16  TDWI RESEARCH
 Evaluating Benchmark Scores

Summary
The TDWI Analytics Maturity Assessment provides a quick way for organizations to assess their
maturity in analytics and compare themselves in an objective way against others with analytics
initiatives. The assessment is based on the TDWI Analytics Maturity Model, which consists of five
maturity stages plus a chasm between stages 3 and 4.
The assessment serves as an initial measure of your analytics maturity. Its 52 questions across five
categories merely touch the surface of all of the complexities involved in building out your analytics
ecosystem. To gauge precisely where you are, it may also make sense to work with an independent
source to validate your progress.

tdwi.org  17
TDWI Analytics Maturity Model Assessment Guide

Research Sponsors

aws.amazon.com www.matillion.com
For 14 years, Amazon Web Services has been the world’s most Matillion is data transformation for cloud data warehouses. Matillion is
comprehensive and broadly adopted cloud platform. AWS offers over 175 fully purpose-built for Snowflake, Google BigQuery, and Amazon Redshift,
featured services for compute, storage, databases, networking, analytics, enabling businesses to achieve new levels of simplicity, speed, scale, and
robotics, machine learning and artificial intelligence (AI), Internet of Things savings. Trusted by companies of all sizes to meet their data integration and
(IoT), mobile, security, hybrid, virtual and augmented reality (VR and AR), transformation needs, Matillion products are highly rated across the AWS,
media, and application development, deployment, and management from 76 GCP, and Microsoft Azure Marketplaces. Dual-headquartered in Manchester,
Availability Zones (AZs) within 24 geographic regions, with announced plans U.K. and Denver, Colorado, Matillion also has offices in New York City and
for nine more Availability Zones and three more AWS Regions in Indonesia, Seattle. Learn more about how you can unlock the potential of your data with
Japan, and Spain. Millions of customers—including the fastest-growing Matillion’s cloud-based approach to data transformation. Visit us at www.
start-ups, largest enterprises, and leading government agencies—trust AWS matillion.com.
to power their infrastructure, become more agile, and lower costs. To learn
more about AWS, visit aws.amazon.com.

www.snowflake.com www.tableau.com
Snowflake’s cloud data platform shatters the barriers that have prevented Tableau helps people and organizations become more data-driven as the
organizations of all sizes from unleashing the true value from their data. More trusted leader in analytics. The Tableau platform provides the breadth and
than 3,000 customers deploy Snowflake to advance their businesses beyond depth of capabilities to serve the needs of even the largest global enterprises
what was once possible by deriving all the insights from all their data by all in a seamless, integrated experience. Tableau is designed to fit, not dictate,
their business users. Snowflake equips organizations with a single, integrated your data strategy and adapts to your environment with unmatched flexibility
platform that offers the data warehouse built for the cloud; instant, secure, and choice while meeting the toughest governance and security requirements.
and governed access to their entire network of data; and a core architecture People love using Tableau because it is both powerful and intuitive—and
to enable many types of data workloads, including a single platform for offers a fundamentally different user experience by empowering people of all
developing modern data applications. Snowflake: Data without limits. Find out skill levels to explore and analyze data using visuals and natural language.
more at Snowflake.com. Tableau has become the standard language of analytics for modern business
users and continues to lead the industry with the most passionate and
engaged user community in analytics, a customer base with millions of users
at more than 86,000 organizations, and a deep commitment to customer-
focused innovation. Learn more at www.tableau.com.

18  TDWI RESEARCH
 TDWI Research provides research and advice for data
professionals worldwide. TDWI Research focuses exclusively
on data management and analytics issues and teams up with
industry thought leaders and practitioners to deliver both
broad and deep understanding of the business and technical
challenges surrounding the deployment and use of data
management and analytics solutions. TDWI Research offers
in-depth research reports, commentary, inquiry services, and
topical conferences as well as strategic planning services to
user and vendor organizations.

T 425.277.9126
555 S Renton Village Place, Ste. 700 F 425.687.2842
Renton, WA 98057-3295 E [email protected] tdwi.org