Data catalog strategy

Operationalizing data
governance to achieve
data-driven transformation
 Data catalog strategy | Operationalizing data governance to achieve data-driven transformation

Executive summary 05
Data-driven transformation as this
decade’s grand challenge 06
Action areas for data-driven transformation
Data governance operationalization as central quick win
Data catalog at a glance 10
Core capabilities and features
Classification of data catalog archetypes
The agony of choice: data catalog
adoption as a decision-making issue
Data catalog strategy approach 15
Make your transformation a success with the right
data catalog strategy
When to start data cataloging
during your transformation?
Case study:
An energy technology company relies on its data
catalog strategy to drive transformation post carve-out 22
Value contributions and way forward 24
Your contacts 26
References 26

03
Investing in Germany | A guide for Chinese businesses

04
 Data catalog strategy | Operationalizing data governance to achieve data-driven transformation

Executive summary
Turning your organization into a data-driven company requires
Finally, we will briefly outline how this strategy adds value in the
transformational success across multiple data-related action areas,
form of quick wins for data-driven transformation as well as other
including data strategy, data value creation, data governance, data
financial impacts and organizational support. In particular, a coher-
culture, and data architecture. While only those transformations
ent data catalog strategy helps to reach the individual transforma-
that dedicate sufficient effort to each of these areas will ultimately
tion goals of responsible C-suite and data governance leaders as
succeed, data governance plays a pivotal role in transformations
outlined in following.
focused on the use and democratization of high-quality data. On the
one hand, data governance is built on data strategy and therefore
Chief information officer (CIO)
focused on creating value from data. On the other hand, it must
Gain a comprehensive overview of the success of your IT strategy
create the organizational, behavioral, and technological framework
that enables you to draft a roadmap for the evolving IT architecture
needed to treat data as crucial enterprise asset. And since every
and landscape, optimize existing IT processes and assess how avail-
data-driven transformation journey requires serious patience and
able IT and data assets will impact the business environment.
sustained management buy-in, it is key to deliver valuable and
tangi- ble quick wins for the company right from the start.
Chief data officer (CDO)
Achieve data-driven transformation at scale, embed data gov-
The ability to operationalize your data governance model can make
ernance within the organization and set up a data marketplace
or break any transformation endeavor, and that requires, among
to monetize corporate data assets and add measurable business
other tools, dedicated metadata management solutions realized by
value.
data catalogs. Equipped with capabilities for data inventory, quality,
governance, lineage, and privacy, data catalogs help document and
Chief information security officer (CISO)
retrieve data stored across the enterprise. There are numerous data
Ensure your data complies with all regulations and introduce a
catalog archetypes that can be realized for metadata management,
resource-based data security strategy that protects enterprise
ranging from disintegrated spreadsheet-based solutions for siloed
data assets and safeguards your company’s reputation by mitigat-
usage in specific business contexts to highly integrated systems
ing data breaches and security threats.
serving on an enterprise scale. Finding the right archetypes for your
needs – and the data cataloging systems to go with it – remains a
Chief technology officer (CTO)
challenge for many. In this whitepaper, we outline our preferred
Assess different catalog solutions for state-of-the-art data infra-
method for developing your data catalog strategy from solution
structure and identify features as well as synergies for using and
screening, scoping, and sourcing to building the guardrails that will
reusing metadata to avoid excessive infrastructure spending and
enable you to put that strategy into action. We will also discuss the
technical debt over the long term.
impacts, risks, and best next actions to kick-start your data catalog
strategy, whether you take a greenfield, brownfield, or bluefield
Head of data governance
approach to transformation. On the basis of a case study from one
Evolve governance capabilities through holistic metadata man-
of our clients in the energy technology sector, we will show the ben-
agement and rely on data literacy, data democratization, and
efits of our method in this brownfield transformation, both during
self-service access to make data governance more visible and
and post-transformation.
impactful within the transformation process.

0
 Investing in Germany | A guide for Chinese

Data-driven transformation
as this decade’s grand
challenge
In today’s highly volatile business environment, companies have to Action areas for data-driven transformation
make decisions amid a lot of uncertainty to face the increasingly As complexity continues to grow and data volumes, formats, sour-
complex market structure and the varying purposes as well as ces, and requirements become more diverse, companies have
definitions of the stakeholders involved. Demand for domain- started transforming their legacy data organizations and systems to
specific, high-quality, and intelligence-enriched data is growing as keep up with this new data-driven era. C-suite leaders place special
leaders aspire to more efficient and more effective lines of action emphasis on data-driven practices and management principles to
at every level. However, many companies in general and their IT enable an agile response to the ever-changing business environment.
departments in particular are overwhelmed by the massive volume Data-driven transformation promotes the purposeful use and
of data they have to generate and maintain. One primary driver for monetization of data across the enterprise as today’s tie-breaking
this is an immature data management practice evolved over time. business resource, but myriad individual initiatives are needed to
While insufficiently designed and operated data governance organ- deliver the necessary changes at all levels of the organization to
izations keep growing the disconnect between business and IT, master this challenge [1]. Regardless of the enterprise’s maturity
legacy technologies as well as disintegrated systems do not allow to level, there are five key action areas for any data-driven transforma-
enhance the value of corporate data. Many enterprises start coun- tion (Figure 1).
tering these developments by large-scale digitalization initiatives to
manage and deliver a data-driven transformation, hoping that will First, every data-driven company needs a data strategy that estab-
make them more resilient in our new and unfamiliar ‘data era’. lishes organizational guardrails and technological principles for

Fig. 1 – Action areas for data-driven transformation, including data governance as a pivot point
between conceptualization and implementation

…aligns with corporate and IT

strategy to define and share
Operationalizing data
a common vision.
governance…

…prioritizes the most valuable · …serves as an anchor

Change Management & Communication

data and analytics use cases point between design and

Foster

for future monetization. imple- mentation,

· …defines the data strategy

…provides the framework for the
and creates business value,
utilization and democratization of
high-quality data.
· …shapes the data culture and
designs the data architecture,
…helps establish a mindset within and
the organization that sees
enterprise data as an asset. · …delivers quick wins on the
Follow

transformation scoreboard.
…designs, builds, secures and
operates the technology
infrastructure.
0
 Data catalog strategy | Operationalizing data governance to achieve data-driven

managing, acquiring, processing, storing, securing, sharing, and

Fourth, every transformation should evolve a data culture to make
using enterprise data. To formulate a common vision for data man-
company stakeholders think of data as an asset. Early mobilization
agement practices, it is important to align the data strategy closely
and continuous training of the staff leverage data-driven business
with the corporate strategy in general and the IT strategy in par-
processes and contribute a common understanding of data as
ticular. Second, data value creation is concerned with the selection,
a crucial business resource. Fifth, data architecture shapes the
prioritization, and agile delivery of data-driven use cases to add
technological foundation in terms of required data platforms,
tangible merit to the firm. Eligible high-priority use cases pay direct
technical security standards, and IT infrastructures. Implementing
dividends to the business either through an increase in revenue
the concepts of this holistic technology foundation blueprint finally
or a decrease in costs, demonstrating the meaningful financial
emerges the data-driven company. To involve stakeholders in the
and organizational impact of a data-driven transformation. These
proposed action areas, the whole transformation must be accom-
tangible outcomes range from improving cross-departmental
panied by comprehensive change management and communica-
collaboration by breaking down data silos to measurably monetiz-
tion initiatives.
ing data by charging third parties to use it. Third, data governance
defines required organizational capabilities, roles, processes,
Since most business stakeholders see data-driven transformation
tools, policies, standards, and indicators to facilitate utilization
as a long-drawn-out and turbulent journey (Figure 2), it may take
and democratization of corporate data [2]. The data governance
time to yield a discernible return on investment. The program is
body is also responsible for the quality of corporate data across its
therefore only likely to succeed if it has the full support of man-
full lifecycle, giving it a central figure in any data-driven company.
agement from start to finish. To secure management buy-in, the

Fig. 2 – Operationalizing data governance ensures tangible quick wins for the transformation and its related action areas

A B
A Strategy definition
B Strategy refinement

C Minimum viable
C D
products
D Business value
Change Management & Communication

at scale
Foster

E Operating model
E
G F Operationalization
F G Organizational
embedding

H Stakeholder
H
J mobilization
I I Training &
enablement
J Data-driven
action system
Follow

K L

I II

0
Investing in Germany | A guide for Chinese
III IV K

I
Phase
n
i
t
i
a
l

d
a
t
a

f
o
u
n
d
a
t
i
o
n
L

D
a
t
a

m
a
n
a
g
e
-
m
e
n
t
s
y
s
t
e
m
s
…

0
 Data catalog strategy | Operationalizing data governance to achieve data-driven

program has to consistently add value through incremental deliv-

Another model would be a multi-style adaptive governance model
erables, as outlined in a holistic roadmap covering realistic time-
as found in the data mesh trend [7] currently being adopted by
lines and critical milestones for each phase of the transformation
many companies. Initiatives based on data mesh principles pro-
program in line with the available resources. It should also focus
pose a new way to manage and use enterprise data by introducing
on delivering valuable results as quick wins early on in the process.
shared responsibilities. In conceptual terms, data mesh decen-
The first phase includes foundational yet transformational initi-
tralizes the ownership of data across different company domains
atives that add significant value to the company; in other words,
and regards data as a distributed product. In technical terms, data
initial pilot projects in critical action areas alongside high-priority
mesh relies on self-service data platforms and a federated data
use cases. While moving along both the transformation roadmap
governance model in order to manage data products across the
and the learning curve, subsequent phases will keep refining these
different domains. Federated governance enables decentralized
first quick wins. Although this approach relies on an agile approach
teams to determine the quality and use of their data products as
throughout the transformation program, in-depth strategic planning
they see fit, but always in line with the governance standards set by
will ensure to reach valuable milestones on the way to becoming a
a centralized authority. This is best achieved through a multi-style
data-driven company.
adaptive governance model following “agile control”.

Data governance operationalization as central quick win

Regardless of choosing a single-style or multi-style approach,
Many data-driven transformations are doomed to fail if enterprises
operationalizing the governance model is the next hurdle for
do not give data governance the priority it deserves in the overall
prospective data-driven companies. It can take several years to
program. Having a functioning data governance body is important
complete a data-driven transformation, so it is vital to focus on the
for all of today’s enterprises, and even more so for data-driven
early roll-out and operation of a stable data governance model to
companies. Among the proposed action areas for data-driven
capture and manage enterprise data moving forward. Once the
transformation, data governance serves as a broker when it comes
data governance model design is defined, e.g., in terms of required
to executing the data strategy, creating value with data, and putting
roles, processes, and operating model, it is essential to install
both the behavioral data culture as well as the technological data
required information systems to achieve it. In many cases, this
architecture in place [3].
means putting software systems for managing corporate master
data, metadata, and data quality in place [8]. The data catalog is
Setting up a data governance body can be difficult to achieve. With
conceived to be the most vital tool during this operationalization
the focus solely on data management issues relating to compliance,
phase as it provides the central platform to build and run the
regulation, and monitoring, they often lack the flexibility to respond
gov- ernance model of choice.
to the dynamic market environment. This leads to complaints from
stakeholders faced rigid data governance rules, especially when it
prevents them from utilizing enterprise data as their needs change
[4]. These bodies may also overstep their boundaries if they define
their scope too broadly, fail to focus on specific use cases, or lack
sufficient budget, talent, or skillsets.

The many different initiatives required for data-driven transforma-

tion suggest that the specific data governance model companies
choose will depend on the individual business context [5]. To
deal with the changing requirements for data-driven companies,
Gartner introduced the concept of “adaptive data and analytics
governance” (or, “adaptive governance”) to select and deploy the
right data governance model [6]. Traditional governance empha-
sizes “control” through regulations and compliance, while adap-
tive governance focuses on “outcomes”, “agility”, and “autonomy”
(Figure 3). Adaptive governance also allows for combining
individ- ual styles to meet further structural requirements. For
instance, a smart factory could tailor an adaptive governance
model for “autonomous outcomes” for its supply chains and
involved data prosumers like cyber-physical systems and digital
twins.

0
 Investing in Germany | A guide for Chinese

Fig. 3 – Gartner's adaptive governance styles, including sample governance models

Autonomous AI-driven decisions to create business value in real-time scenarios

Distributed governance to sustain the quality and use of individual data products
Agility
Governance

Outcome Harmonzied risks and returns for processes with a high

business impact

Control Rigid command-and-control governed by centralized

rules, policies and standards

Examples
Adaptive Data Governance

Top-down compliance systems for governing highly

Traditional Governance regulated domains like banking and finance

Decentralized teams held to central governance standards as proposed by data mesh concept
Federated Governance
Governance

Analytics artifacts defined, executed and used in line with business standards and expectations
Analytics Governance

Broader scope for traditional governance body fostering

Information Governance the use of data to achieve business goals

1
 Data catalog strategy | Operationalizing data governance to achieve data-driven

Data catalog at a glance

A data catalog is the single point of truth for all corporate data
assets to foster data literacy, enable data democratization, and
support cross-sectional collaboration between business and IT.
In practical terms, the data catalog makes metadata management
possible, which for several decades has been a prerequisite for
Data inventory
data governance and other data management practices. Metadata
The data catalog supports both the manual and the automated
management enables to structure, store, and locate information
capture of data assets classified as technical, operational, business,
about corporate data assets, e.g., roles and responsibilities, stor-
social, processual, administrative, and other kind of metadata from
age locations, and usage policies. Managed metadata is essential
various sources. There are metadata management features in
to creating a common understanding of corporate data assets
data catalog tools for metadata integration, modeling, population,
between different units and contexts.
curation, and evaluation. Stakeholders of various backgrounds can
also use the data catalog as a software-based marketplace to rate,
In many companies, however, the metadata management process
review, and retrieve data assets.
is often inadequate to meet today’s business challenges. Sustain-
ing a viable inventory for corporate data assets has often been
neglected due to streamlined budgets, tight project deadlines, and
limited manpower. Faced with a lack of integrated, reusable data
as well as insufficient data quality, data was often kept in local data
silos, severely limiting overall business performance. Following
the goals of every data-driven transformation to integrate and Data quality
reuse corporate data, managing data as an asset and establishing To warrant the quality of enterprise data within the information
high data quality are two factors critical to success. Consequently, systems, a data catalog provides features for alerting, reporting, and
underpinning an individually defined data governance model with functions for predictive, continuous, as well as self-service monito-
a suitable data catalog solution is a considerable transformative ring in order to sustain trusted insights. For that, the data catalog
landmark. offers functionalities to define static and adaptive data quality rules
created and maintained by responsible stakeholders.
Core capabilities and features
First and foremost, the data catalog allows to create a register of all
available data assets, thereby adding value for all stakeholders by
giving them self-service access to corporate data. It offers features
that automate the collection, management, and distribution of
metadata and enables stakeholders to locate, use, understand,
and monetize corporate data in their respective business domains. Data lineage
There are also workflow-driven self-service functions in the data For discovering the full data lifecycle, the data catalog tracks data
catalog that give responsible stakeholders the ability to manage, flowing from their origins in respective data sources to the places
maintain, and monitor the quality of data. In summary, the five core of their transformation, alteration, consumption, and deletion.
capabilities data inventory, data quality, data lineage, data govern- Views on this data lineage are highly customized and provided as
ance and data privacy are defined as follows: either business or technical lineage. Stakeholders can then use
both lineage types in impact analyses to track errors, implement
process changes and perform system migrations. By allowing
enterprises to reorganize and streamline their organizational data
flows, data lineage can help reduce complexity as well as costs.
The full traceability of technical and business data assets also helps
enterprises better understand their reuse potential in various
contexts.

0
 Data catalog strategy | Operationalizing data governance to achieve data-driven

Classification of data catalog archetypes

Over the years and during past development cycles, we have seen
different metadata management systems evolve for various areas
of application [10]. While features for managing metadata found
Data governance their way directly into operational and analytical standalone soft-
To operationalize data governance, a data catalog provides fea- ware products, data catalogs as integrated software systems for
tures for creating workflows, checking compliance of data assets professional metadata management started to evolve, too. The
toward regulations and defined business rules, managing access market for data catalog products has grown significantly in recent
rights, versioning data assets, and conducting audit trails. This years. The number of products listed as data catalog leaders in
enables key users, i.e., data domain managers, stewards, and Gartner’s Magic Quadrant for Metadata Management Solutions
custodians, to manage their data assets. The business glossary and more than doubled from four in 2016 [11] to ten in 2020 [12].
data dictionary are two major data catalog features that support
data governance. The business glossary is key to define a common According to this development of metadata management solutions
language for an organization in terms of central business terms and corresponding products, a multitude of abstract archetypes
and reference data, while the data dictionary serves as a metadata for data cataloging (henceforth archetypes) can be distinguished
repository for IT stakeholders with a technical view of enterprise by their (1) required efforts for organizational integration, (2) main
data assets. Both the business glossary and the data dictionary area of organizational usage, (3) addressed metadata scope, and
help users from different domains retrieve and use data according (4) level of metadata integration. In addition, each archetype
to their needs. A data helpdesk can provide further assistance for individu- ally fulfills the provided set of data catalog core
users to resolve issues related to specific assets. capabilities, varying from low to high. The first distinguishing
characteristic relates to the amount of effort required to embed the
data catalog within an organization. The second is based on
whether a specific archetype of data catalog is used more for an
enterprise’s business or IT division. The third criterion looks at
whether a data catalog arche- type has a passive metadata scope
focused more on documenting technical metadata or an active
Data privacy metadata scope that also includes operational, business, and social
One of the key functions of a data catalog is to protect stakehol- metadata. The fourth character- istic is about the level of metadata
ders’ privacy and personally identifiable information (PII) against integration that will depend on whether the data catalog archetype
a multitude of IT threats through role-based permissions that bundles and collects metadata from various disparate sources or
manage and control access to data assets and related resources. focuses solely on a limited set of selected sources.
As PII management is regulated by law, the data catalog must pro-
vide granular, asset-level security controls in line with legal regula- The provided classification matrix (Figure 4) incorporates nine
tions, e.g., the EU General Data Protection Regulation (GDPR) or distinct data catalog archetypes which can be realized by available
the California Consumer Privacy Act (CCPA). products for metadata management or corresponding product
features. In the following, a brief description for each of these data
As an important information systems’ class for data management, catalog archetypes is provided with close reference to the afore-
data catalogs went through three major evolutionary stages for mentioned classification criteria.
developing their core capabilities [9]. The first wave focused solely
on IT governance aspects by providing IT teams with a data inven-
tory to support their day-to-day tasks. The second wave brought in
business-side stakeholders looking to introduce data governance,
including the data stewards in the respective data domains we
know today. Now, in the third development wave, data catalogs are
focused on promoting enterprise-wide collaboration and helping
diverse user groups across the organization locate and use the
data they need.

1
 Investing in Germany | A guide for Chinese

Fig. 4 – Classification matrix for data catalog archetypes

Metadata Scope DC Capabilities1

Indicates whether the scope is to document technical assets (passive) or to
also manage further ones like business assets (active)
# I Q L G P Impact
Passive Active

9
Augmented Catalog
9
8
Enterprise Catalog
Enterprise
7 8

Indicates whether a data catalog can integrate

Arch. Catalog
Describes the effort required to implement

disparate sources or selected sources only

7
Ma

Hi
Analytics

Metadata Integration
6
Organizational Effort

Catalog
data catalog

API Catalog of
Catalog
4 Catalogs
5 5

4
Data Product
Catalog
3
3
Min

Vendor- 2
L

Specific Catalog

2
Spreadsheet
Catalog
1
1
IT Business

Organizational Usage 1
I: Data Inventory Q: Data Quality L: Data Lineage
Indicates whether the data catalog is targeted G: Data Governance P: Data Privacy
towards business or IT purposes

Spreadsheet catalog
manage the quality of quite passive metadata, though the level of
With an increasing number of data silos in today’s companies,
integration with other metadata sources is low. An enterprise may
business units and stakeholders will often maintain individual lists
choose to bundle several of these vendor-specific catalogs from
of data assets to manage the data in their own domain. These
different software applications to create a “catalog of catalogs”
offline data catalogs are typically a first attempt at collecting and
(explained in the following) and integrate previously disparate
managing passive and active metadata for a very limited scope.
metadata sources at a later date.
The level of metadata integration in this “quick and dirty” approach
is, however, quite low. On the other hand, with only minimal effort
API catalog
required, it allows stakeholders to create an initial inventory of their
With the API economy on the rise, IT managers can use an API
data assets very quickly. These catalogs are generally based on
cata- log to define, manage, and document passive metadata for
more or less customized worksheets in Microsoft Excel or other
sys- tems, interfaces, API endpoints, technical data lineage, and
spreadsheet software.
other technical data assets. Data management software designed to
manage enterprise architecture and virtualize or integrate data can
Vendor-specific catalog
set up API catalogs that are very similar to vendor-specific catalogs,
Most vendors of standard software (e.g., ERP, CRM, SRM) and
though they focus solely on managing the technical metadata men-
other software solutions come with integrated features for meta-
tioned above. As one or only a few software entities are actually
data management in their suite of tools. IT systems designed for
maintaining the technical data, the level of technical metadata inte-
operational and analytical applications usually have built-in fea-
gration with this solution is medium to high. However, setting up
tures to document the data assets used by the software. These
and running these platforms tends to be a challenging endeavor.
vendor-specific catalogs help set up decentralized inventories to

1
 Data catalog strategy | Operationalizing data governance to achieve data-driven

Analytics catalog Catalog of catalogs

Stakeholders and business units focused on analytics, business Scraping together metadata from one or more catalogs establishes
intelligence (BI), and information management tend to prefer meta- a “catalog of catalogs”, a pattern that combines different kinds of
data management models that track data flows, source and target metadata from various decentralized sources into a single man-
systems, transformations, reports, KPIs, metrics and additional agement tool. A catalog of catalogs relies on proprietary software
analytics-related data assets. However, with the excessive technical to merge non-integrated or distributed data catalogs from the
and organizational overhead required, very few of the analytics archetypes introduced before. However, due to the decentralized
and BI projects in the past focused metadata management. Most storage and varying quality of the individual catalogs, the catalog of
of these projects are cloud-based today, which has increased catalogs has an average level of metadata integration for technical
momentum for metadata documentation tools. Thanks to cloud- as well as organizational data assets. In addition, maintaining a cat-
based catalogs from external vendors or cloud-native catalog alog of catalogs is difficult because all changes made to the individ-
components (e.g., Azure Purview, AWS Glue, GCP Data Catalog), ual catalog instances must be aligned with the central repository.
setting up an analytics catalog for data governance appears to be Since a catalog of catalogs is built from professional data catalog
within reach. This type of data catalog may still be quite complex to software, it features professional workflow-driven metadata man-
achieve, but it offers a high level of metadata integration to docu- agement and usage of the data inventory which also improves
ment the ingestion, transformation, use, and democratization of data governance, lineage, privacy, and quality. Using the data mesh
technical and organizational data assets for analytical purposes. paradigm can further professionalize distributed data and meta-
Both the underlying software for the data catalog and the high data management, as it establishes cultural, organizational, and
degree of metadata integration offer a great opportunity to technical solution patterns, including the federated governance
expand this archetype of catalog to an enterprise-wide solution at a of distinct data product catalogs. To integrate these decentralized
later date. catalogs, the catalog of catalogs offers the necessary functionality
to centralize metadata management [13]. The ability to sync dis-
Enterprise architecture catalog tributed metadata remains a key challenge, however, and requires
More and more IT divisions are asking for support when it comes further architectural components [14], e.g., for complex event
to managing data assets in line with the broader business and IT processing, change data capture, and audit logging, all of which
strategy. Products for modelling the enterprise architecture as well certainly add complexity to establishing this archetype.
as mapping business capabilities to the corresponding IT capa-
bilities and vice versa are instrumental in IT-business alignment. Enterprise catalog
Because these tools track how technical and organizational data There are several integrated software suites designed to set up a
assets are linked between the business and IT architecture, both data catalog at an enterprise scale and achieve enterprise-wide
passive and active metadata fall within their scope. It can be chal- data governance [12], [15]. As a centralized platform for managing
lenging to introduce a catalog tool into the enterprise architecture well-integrated metadata, the enterprise catalog offers features for
management practice, but where it succeeds, the result is a high data inventory, governance, lineage, and privacy. Data quality fea-
degree of metadata integration and a rich set of data catalog fea- tures may also come as standard, though they may be limited to a
tures to manage mainly IT-related but also business data. fixed set of static rules. Enterprise catalogs can be difficult to set-
up as many of the products available today are monolithic
Data product catalog standalone software solutions for on-premises environments. It
When it comes to the data mesh concept and its federated may also be hard to maintain the underlying software foundation
approach to data governance, success relies on the ability to estab- which could add further complexity to the data-driven
lish domain-specific ownership of distributed data products doc- transformation. While enterprise catalogs put the emphasis on
umented in decentralized data product catalogs [13]. This allows both passive and active metadata, their area of usage is mostly
local data owners and stewards to document all of the data they related to the whole busi- ness organization.
produce and consume. Each data product catalog documents the
content, structure, and APIs of business-related data to trace its Augmented catalog
full lineage, which means it targets mainly on active metadata and As data management becomes increasingly complex, the next
business users. With a federated and decentralized approach, this generation of data catalogs are expanding beyond the prevailing
archetype of catalog has a medium level of metadata integration, enterprise-class solutions to include modern platforms, technol-
though it is quite easy to integrate. Data mesh architecture allows ogies, and features. These so-called augmented catalogs offer
an enterprise to integrate federated data product catalogs into a general cloud-readiness as well as data quality features supported
central catalog, resulting in the functionality and reach of an enter- by machine learning, active metadata and “big metadata” man-
prise-wide catalog yet must function and operate along the lines of agement, embedded collaboration, and end-to-end metadata
a catalog of catalogs (see next). visibility [9]. Because cloud computing is one of the technological

1
 Investing in Germany | A guide for Chinese

key drivers in data-driven transformation, augmented catalogs are What is more, adopting a data catalog does not require a “big
also designed to make the most of such advantages as scalability bang approach” as an evolutionary approach involving stakehold-
and maintainability. This archetype of data catalog can enhance ers from several management levels is often better suited to this
data quality through machine learning, e.g., applied to metadata endeavor. A company may opt instead to start with small-scale,
matching. Augmented catalogs focus on active metadata manage- highly localized solutions like spreadsheet-based or vendor-based
ment and add value through data usage in the business context. catalogs to achieve tangible as well as useable results fast, while
Because metadata, like other types of data, is growing in volume, preparing in parallel to roll out an enterprise-wide solution with
augmented catalogs derive insights from “big metadata” to improve catalogs for managing enterprise architecture, APIs, or further cor-
an enterprise’s technical and business performance. This data porate data assets. Taking a gradual approach like this achieves a
catalog archetype also provides all user groups with embedded higher level of metadata integration but also increases integration
functionalities and workflows to promote collaboration in data and project execution effort. Consequently, it is required to weigh
asset management. Finally, augmented catalogs offer end-to-end the various impacts and trade-offs for introducing a holistic data
visibility of highly integrated metadata by creating a single point of catalog solution comprising one or more data catalogs.
truth.

The agony of choice: data catalog adoption as a

decision-making issue
The presented classification matrix helps to quickly evaluate the
technical and organizational properties of different archetypes of
data catalogs based on the functionality they provide. Further-
more, this overview offers a tangible basis for making an informed
decision on selecting required data catalog archetypes truly fitting
the scope of a transformation program.

It would be a mistake, however, to choose a data catalog providing

more functionalities than actually needed which bears the risk of
wasting company resources without achieving the transformation
goals. Since there is obviously no “one size fits all” solution, exer-
cising caution when analyzing the available options is advised. The
first key step is to assess the individual goals and needs of the
enterprise as it transforms. Each data catalog solution is as unique
as the transformation journey itself and generally requires a combi-
nation of two or even more data catalog archetypes to meet the
varying metadata management needs.

1
 Data catalog strategy | Operationalizing data governance to achieve data-driven

Data catalog
strategy approach
Data-driven companies succeed when they have the right strat-
both in detail is vital prior to the strategy project. In operational
egy, culture, governance, use cases, and architecture in place. It
terms, this will also enable you to make metadata management
is up to management to define and substantiate all of the action
initiatives an integral part of your overall transformation project
areas involved in a transformation, while also making sure they
portfolio which will help specifying and further refining your data
are all closely aligned – always with the goal of making the best
catalog solution.
use of enterprise data. Data governance provides the organizing
framework for integrating and managing enterprise data assets
The ultimate goal of your data catalog strategy is to operationalize
and as such is key to evolving as a data-driven company. To
data governance in a timely manner and with sufficient quality to
design and successfully run a data governance unit, companies
deliver tangible as well as visible quick wins in your transformation.
must implement necessary information systems. At least one of
For that to happen, you need to meet several milestones within
these applications must offer a data catalog feature to ensure all of
your data catalog strategy, from selecting vendor products and
the stakeholders involved are on the same page in terms of the
preparing for implementation to embedding the concept within
poten- tial of enterprise data assets. In our experience, it is really
the enterprise and defining the way forward. Once you have
hard work to get the process right when it comes to
selected the right products, you can start the procurement
operationalizing the data governance model as choosing the right
process and ensure your data catalog has only the metadata
or wrong data catalogs could make or break the process. To enrich
management features you truly need. This prepares you for the
and align the operationalization endeavor with all the details and
upcoming imple- mentation phase when you need to deliver
specifics of the
important data catalog MVPs and critical use cases to maximize the
broader data-driven transformation program, having a holistic data
visibility of both the transformation program and the data catalog
catalog strategy will help to succeed with the transformation in
solution. This is also the time to embed the solution within the
general and data governance in particular.
business and technology architecture in a way that directly benefits
business stakeholders. Finally, the results of this phase will
Make your transformation a success with the right
determine the way forward in the form of clearly defined, agreed
data catalog strategy
roadmaps for your metadata management with crucial milestones,
Although data-driven transformations differ in size and scope,
organizational responsibili- ties, and transformation
finding the right data catalog is a top priority for every program.
dependencies.
The complex and resource-intensive process of introducing a data
catalog involves both technical and organizational risks. Failure to
The following diagram outlines our structured approach that will
define the scope, immature data quality practices, technology and
help you achieve the goals of data catalog strategy, from screening,
infrastructure issues, and the lack of talent or resources to com-
scoping, and sourcing to setting the right data catalog solution (Fig-
plete the transformation are among the main risk factors. That is
ure 5). This process works in virtually every data-driven transfor-
why it is so crucial to define a viable data catalog strategy to obtain
mation program, though the outline below is designed as a generic
the goals of the targeted transformation and its data governance.
reference only, which you can adapt as needed for your specific
transformation. In the next section, we will shed light on the wide
The first step in your data catalog strategy is to identify what is
variety of transformation types and briefly outline the benefits
driving your metadata management objectives; in other words,
and risks of creating a data catalog strategy for existing (as yet not
your data governance requirements to meet key business needs
clearly defined) transformation programs.
and to comply with general or domain-specific legal regulations.
And since you need to align your data catalog solution, including its
strategy, with your overarching corporate and IT strategy, defining

1
 Investing in Germany | A guide for Chinese

Fig. 5 – Deloitte’s approach to data catalog strategy

Input Data catalog strategy Output

Scree Scop Sour

n e ce
Business DC vendor
needs · Transformation · DC vision & selected
· Vendor longlist
program mission · RFI/Utility-benefit analysis
overview statements · Vendor shortlist
· Use case canvas · DC fit-gap analysis · Demos/POCs
· Requirements list · DC solution design · RFP/DC eval.,
· DG/DC health check (incl. suitable recommend. & selection
· DG/DC models)
maturity · Target
Project assessment architecture Way forward
portfolio blueprint Set
defined
· Use case prioritization
· DC roadmap & deliverables
· Scenario impact analysis
· Initial DC product backlog
· Initial service
operating model
(TOM)
· DC playbook

Corporate strategy
Ready for
implementation
Enterprise architects
Data domain managers
Data catalog product owner

IT/Governance council

Data strategists
Legal
Business/Use case owners Org. embedding
regulations
outlined
Data

Change management & communication strategy

1
 Data catalog strategy | Operationalizing data governance to achieve data-driven

Screening term, and long-term perspective to prepare for the planning and
The first step toward finding the right data catalog solution for your roll-out of your data catalog solution. Prioritizing use cases also
data governance ambitions is gaining a clear understanding of your offers an opportunity to ensure important quick wins for your
transformation goals by roughly outlining your data strategy and transformation program by contributing useful data catalog results
the ongoing initiatives in related areas. Based on this initial over- early on.
view, you will also have a good idea of the scope of your transfor-
mation program, whether it focuses on transforming your business Result: A high-level outline of the data catalog solution,
operations, IT operations, or both at the same time. Then it is time including a plan for embedding it into the company’s techni-
to get your central stakeholders involved to specify and compile cal architecture and organizational action system.
their respective use cases, so that you can better understand the
kind of metadata management support they need. All technical Sourcing
and organizational stakeholders must also define and prioritize Once you have a clear understanding of the status quo and target
their most important functional and non-functional requirements data catalog solution, the next phase is about sourcing the right
for the data catalog solution. Subsequently, an overall health check vendor products to close any identified capability gaps. You can
must reveal the current maturity level of the enterprise’s data skip this step if your company plans to set up spreadsheet or ven-
governance model and metadata management [16], all of which is dor-specific catalogs using previously purchased software. For all
formally documented in preparation for the next strategy phase. other archetypes, now is the time to make a list of vendor products
that can build the data catalog archetype you need. The prese-
Result: A deep understanding of the status quo in terms of lected and weighted requirements are useful in the utility-benefit
the role data cataloging plays in your enterprise’s overall analysis based on the product information sheets obtained from
transformation program. various vendors. Subsequently, it is required to test each of the
shortlisted products that meet your requirements for key use
Scoping cases in demonstration and proof of concept sessions with ven-
Leaders keen to make a quick decision may be misguided by the dors. Finally, conducting a business case evaluation compiles a list
sheer number of options and opt for a full-scale data catalog of top-ranked product solutions to make the final selection.
solution since it seems the most comprehensive and therefore
best choice. This is an approach doomed to fail, not only because Result: A shortlist of recommended vendor products that
of limited resources but also because you run the risk of over- meet your requirements and best close any capability gaps.
shooting the goals behind your strategy if they have not been
clearly defined. That is why it is so important to understand the Setting
scope of your data catalog solution based on the results of the Subsequent to or in parallel with product sourcing, the final strategy
previous phase. You need to draft a vision statement defining the phase is about preparing for the implementation of your chosen
organizational, financial, and technical impacts the data catalog data catalog solution. That includes consolidating information
solution is expected to have on your company and a mission state- related to crucial program events in a holistic data catalog roadmap
ment clearly outlining the operational improvements you hope to with a short-term, medium-term, and long-term perspective. As
achieve. Communicating these vision and mission statements to all the previous step may deliver different roadmap alternatives, a
stakeholders is key to secure buy-in for your initiative. In a fit-gap scenario impact analysis should identify the most viable roadmap
analysis, you will discover which additional data catalog capabilities design. Then, an initial product backlog for your data catalog will
you need to meet your requirements for data governance and lay the groundwork for a seamless start of the implementation
other action areas related to the transformation program. Based phase after the sourcing and preparation phases. In line with the
on the available resources and requirements, you can identify one resources allocated to the program, a high-level service operating
or several data catalog archetypes that will fill any organizational model will help you understand which roles and responsibilities are
as well as technical capability gaps. This short list of data catalog required to embed your data catalog solution within the organiza-
archetypes will also narrow down the list of potential vendors for tion. Finally, the data catalog playbook documents the main capabil-
product sourcing in the next step and expand the blueprint for ities and features of your target solution as well as the stakeholders
the company’s target architecture, helping you to discover and involved along with their interaction routines, design patterns and
debate all technical or organizational dependencies and critical use cases.
implications at
an early stage. Given your enterprise’s aspiration to complete the Result: Implementation preparation and defined way for-
data-driven transformation in a timely manner, it is important at ward for realizing your targeted data catalog solution.
this stage to prioritize critical use cases with a short-term, medium-

1
 Investing in Germany | A guide for Chinese

When to start data cataloging during your transformation?

The following diagram provides a brief overview of how to develop
As indicated above, adding a data catalog to your transformation
the right data catalog strategy for greenfield (initial maturity), brown-
endeavor can significantly increase complexity, so it is in every
field (defined maturity), and bluefield (advanced maturity) transfor-
com- pany’s interest to come up with a viable strategy early on.
mations (Figure 6). Based on these maturity levels, the transforma-
However, with such different business agendas, scarce resources,
tion program will have already made some progress with regard
and varying levels of maturity, no two companies will have the
to data strategy, data governance and metadata management, all
same transforma- tion priorities. That is why it is so important to
of which are important for determining the mission for your data
choose a data catalog strategy that is aligned with the individual
catalog solution. As there are distinct opportunities and risks inher-
status and the maturity level of your transformation program.
ent in the data catalog strategies in each maturity level, our method
recommends the best next actions in each case.

Fig. 6 – Impacts, risks, and best next actions for developing a data catalog strategy for
greenfield, brownfield, or bluefield transformations

Greenfield Transformation Brownfield Transformation Bluefield Transformation

• Data strategy concept outlined/ • Data strategy defined • Data strategy refined
drafted • DG org. structures largely in place • DG operating model widely estab-
• Initial DG concept sketched • Immature siloes available for meta- lished
Progress data mgmt. • Legacy DC solution available

• Build metadata management solu- • Adapt metadata mgmt. solution to • Migrate & adapt existing solution in
tion from scratch DT program’s context the cloud
• Impact and accompany DG develop- • Evolve DG through underlying tech- • Stabilize & further extend the existing
Mission ment as co-creator nology and features DG body’s role

• Use synergies of targeted DC • Mature understanding of overarching • Expanded DG roie through sophisti-
solution as co-creator for the DT transformation framework to drive cated technology base to achieve
program, esp. in large-scale trans- selection & build-up of target DC DC use cases & workflows
Impact formations • More resources available for • Where DG & DC solution has a high
• Operationalization of DG early on data ownership & stewardship level of maturity, the effort & resour-
• Foster stakeholder acceptance • Improved cost and time management ces required to define, implement, &
through steep learning curves for for DC strategy and implementation migrate the DC strategy are relatively
the DC solution low

• Possible “mismatch” of DC models • Increased number of offline/shadow • Lifting & shifting of the existing solu-
as transformation program advances DC may generate extra overheads tion during DC (cloud) migration will
along the learning curve and redundant work for metadata not mitigate existing flaws in source
Risks
• High dependencies on DG design, mgmt. systems, DG TOM & organizational
which is still evolving • Ungoverned systems along the design
• Resources unavailable for transformation roadmap until • Refined data strategy may require
data ownership & seri- ous DC comes into play rebuilding existing DC
stewardship
• Assess maturity levels of both trans- • Assess maturity levels of both trans-
• Complete data strategy formation program & DG body formation program & DG body
• Create transformation roadmap • Only if needed, create temporary • Determine whether existing DC
Next
strongly aligned with data strategy DCs as individual spreadsheet/ solution and its features will deliver
and focused on early quick wins vendor-specific solutions for crucial on transformation goals
• Consider initiatives to modernize business contexts & data domains • Source and select viable cloud-
operational & analytical data • Find & address suitable points to based DC solutions
systems expand the data-driven transforma- • Consider Gartner’s 5R cloud migra-
• Define the data mgmt. structure, incl. tion program with a DC solution as tion strategy framework to find the
DG & TOM soon as possible right approach, e.g., decide between
• Allocate sufficient resources for rehosting and rebuilding the legacy
governance during DT to design DC solution
& manage the DC

1
 Data catalog strategy | Operationalizing data governance to achieve data-driven

DC = data catalog DG = data governance DT = data-driven transformation

1
 Investing in Germany | A guide for Chinese

Greenfield Brownfield
Greenfield transformations are characterized by unstable, imma- Brownfield transformations have already made some progress
ture, or even unavailable data management organizations. This is towards becoming a data-driven company, including successfully
especially true for data strategy and governance, both of which defining both data strategy and data governance, but they have
usually are in their infancy. These transformation programs must yet to determine the nature and scope of the data catalog. Starting
therefore build a metadata management solution from scratch on a brownfield usually means having a more mature understand-
to help the company adapt to a changing market environment. It ing of key transformation issues, so it is important to adapt the
makes sense in this case to start developing a data catalog strategy targeted data catalog solution to further develop the underlying
as early as possible that is designed to accompany the data-driven governance structure, including the capabilities required.
transformation from start to finish. The data catalog should be
seen as an integral part of the greenfield transformation program A better understanding of the transformation in general and its
and a co-creator in terms of operationalizing the data governance governance goals in particular should make it easier to find the
design. right data catalog archetype to close capability gaps in metadata
management. Brownfield transformations generally also have more
To be sure, developing and implementing a data catalog strategy resources available to make responsible decisions on data govern-
does not come for free and adds further workload. Large-scale ance and other new systems and structures. It is up to responsible
transformations in particular benefit from early adoption of a decision-makers to establish a data governance body, which goes
data catalog solution because it makes it easier to manage the hand-in-hand with finding the right data catalog strategy in a timely
constantly growing data systems and structures within the trans- manner and at a reasonable cost. And with more resources avail-
formation journey. Early, gradual introduction of a data catalog able, there is also more potential to fill the required roles, e.g., for
creates a steep learning curve for the respective stakeholders, data ownership and stewardship, which will mean more seamless
which will also increase their buy-in for the initiative going forward. implementation and operation of the data catalog.
However, companies starting on a greenfield transformation with
so many unknowns (especially goals for use cases, data strategy The data catalog solution kicks in rather late in brownfield trans-
and data governance) might choose the wrong archetype of data formations, and there is a risk that a lack of maturity may have led
catalog. Given the low level of maturity in the enterprise, they lack to uncontrolled operational and analytical systems being built and
much-needed talent to take on critical roles and responsibilities, run during the initial transformation phases. These enterprises
e.g., for data ownership and stewardship, which could slow pro- may have also made decisions regarding metadata management
gress in the data catalog strategy. that involve standalone and distributed data catalogs as ven-
dor-specific or spreadsheet catalogs, which require more migration
We advise these companies to execute the entire data strategy, and integration effort later. The best next step in this scenario is to
including the data management design in terms of data govern- assess the maturity of the transformation program with respect
ance and the target operating model, in order to better understand to the already established data governance body. This assessment
the consequences and mitigate the risks of greenfield transforma- will help the transformation team find and understand where and
tion. They should also have a transformation roadmap detailing how to progress with an aligned data catalog solution. Temporary
concrete deadlines for deliverables as transformational quick wins, data catalogs for highly disparate metadata management should
chief among them operationalizing data governance with a data ideally be limited to crucial business contexts only. While this
catalog. Other key initiatives in greenfield transformations include avoids the additional organizational effort and redundant work
modernizing or newly designing operational as well as analytical needed to integrate these shadow catalogs later on, they can serve
data systems and allocating sufficient resources to data govern- as viable use cases while a more professional data catalog solution
ance design and data catalog management. is being developed. It is vital here as well to start developing a data
catalog strategy as soon as possible.

2
Investing in Germany | A guide for Chinese businesses
Data catalog strategy | Operationalizing data governance to achieve data-driven

20
2
 Data catalog strategy | Operationalizing data governance to achieve data-driven

Bluefield
So, when to start? Now!
Finally, bluefield transformations focus on migrating the existing
In summary, there is no definitively right time and approach for
technology foundation by lifting and shifting data catalogs and
developing your data catalog strategy, as this is highly depend-
other on-premises systems to the cloud. Enterprises tend to
ent on the individual context, size, goals, timeline, budget and
embark on this kind of transformation while they are refining their
resources of a company trying to become more data-driven. It
data strategy, which means bluefield transformations may also
seems worthwhile in general to develop a coherent data catalog
involve upgrading from one data catalog archetype to another to
strategy as soon as possible, as it enables companies to better
supply any additional capabilities required by the updated strategy.
understand the impact any particular solution will have on
Transformations at this maturity level usually already have a
their data-driven transformation.
well-established foundation for data cataloging, including mature
data governance, viable use cases and functional workflows and
processes for the existing solution.

As data governance and data catalog operations are already

very mature, the number of resources required to develop and
implement a data catalog strategy in bluefield transformations is
relatively low. Lifting and shifting the existing data catalog to the
cloud cannot, however, compensate for hidden vulnerabilities in
connected source systems, governance processes or established
organizational structures. As a result, if the data strategy requires
drastic changes and serious modifications to the data catalog solu-
tion as it stands, it is probably best to rebuild the legacy solution.
The best next action for bluefield transformations is to re-assess
the maturity levels of both the transformation program and the
data governance project to understand the impact of the
required changes. The existing data catalog solution should be
included in this re-assessment to determine whether it brings the
enterprise closer to its transformation and governance goals. It is
vital to start sourcing and selecting a viable cloud-based data
catalog solution as soon as possible. On the basis of Gartner’s 5R
framework for cloud migration, the data catalog strategy for the
bluefield will determine whether an enterprise will be rehosting or
rebuilding its legacy data catalog solution.

2
 Investing in Germany | A guide for Chinese

Case study
An energy technology company relies on its data catalog
strategy to drive transformation post carve-out

Our client is an internationally renowned engineering company

In line with the data catalog’s vision and mission, we drafted a
delivering advanced energy technology products and services. The
solution blueprint that fit into the target architecture.
customer is a system-critical supplier for the emerging energy tran-
Furthermore, we outlined a first service operating model with the
sition. After a carve-out as part of the recent group-wide restruc-
necessary guardrails for embedding the catalog within the
turing, the company became a separate and standalone subsidiary
organization. This initial design knowledge allowed us to
with its own stock listing. This led to challenges for corporate IT
simultaneously source the right software product and prepare for
currently offering insufficient data services due to an immature IT
implementation of the data catalog. Then, inquiries were sent to
landscape comprising a large number of disparate systems, pro-
potential vendors detailing all of the client’s requirements. Most
cesses, and structures for data management. To better compete
promising vendors were asked to provide product and proof-of-
on today’s market, the customer is pursuing a data strategy fit for
concept demonstrations to show how their data catalog solutions
the ongoing digitalization of the economy by utilizing enterprise
would impact the most important use cases.
data as an asset. Like so many of its peers, the client wants to
become a data-driven company.
We documented the results of the demonstrations in our utility-
benefit analysis tool to support the client in their decision-making
To implement their target data strategy, the company launched
process and helped them identify the most viable software solu-
a transformation program on the brownfield to restructure
tions. While the client’s procurement team started investigating
and harmonize the existing data architecture and to create a
the software shortlist, the customer asked for advice on how best
data
to prepare for implementation of the data catalog solution with a
culture that enables stakeholders to unlock the potential of data as
focus on operationalizing data governance. We used the top-priority
an asset. The company has started to build a new organizational
use cases for the data catalog to come up with a strategic roadmap
and architectural model based on the data mesh paradigm. The
for implementation, including key milestones and core deliverables.
new model will foster the democratization, sharing and reuse of
On the basis of a scenario impact analysis, we determined which
enterprise data by establishing shared responsibilities and ena-
roadmap design to recommend as the most viable way forward.
bling users to create data products. As a key concept of data mesh,
We also created an initial data catalog product backlog and a data
the federated data governance model is at the heart of this trans-
catalog playbook so that we could hit the ground running once
formation program. The customer has also launched a full-scale
implementation started.
modernization of its enterprise resource planning system with the
introduction of SAP S4/HANA. By request from the client’s chief
Our initial implementation efforts took an agile approach, focus-
data officer, we proposed a data catalog solution that would serve
ing on a single top-priority use case to deliver a minimum viable
as a key component in operationalizing the federated governance
product that was useful to the client. Consistent with available
and ensure all individual initiatives are closely aligned.
resources and capabilities, our main use case centered on the data
product creation process which had just been introduced as a first
Our experts at Deloitte started by developing a custom data cata-
step towards the data mesh architecture the client was developing
log strategy to chart the most promising way forward. We worked
in parallel. The software product the client selected came with a
with representatives from the client’s data strategy and IT/govern-
jumpstart program designed to install the platform and security
ance councils to compile key use cases and requirements for the
for the enterprise data catalog, followed by the onboarding of the
data catalog that also provided valuable input for our as-is maturity
technical metadata sources required for the use case. Then, in
assessment. Based on a fit-gap analysis, we proposed an enter-
close alignment with the business glossary and data dictionary, we
prise data catalog solution that we felt best met all of the client’s
worked with the client to finalize the design of the underlying meta-
demands in the medium-term. The fit-gap analysis also revealed
data model and adapted existing user interfaces to give data users
that the client needed an enterprise architecture catalog and API
the self-service platform they need to shop for data.
catalog.

2
 Data catalog strategy | Operationalizing data governance to achieve data-driven

A training program evolved in parallel, with materials and work-

shops designed to prepare data catalog users for their respective
roles. The whole endeavor, from strategy design, sourcing, selec-
tion, and purchase to preparing and implementing the MVP took
roughly one year, with the MVP go-live as a major milestone in the
client’s data-driven transformation journey.

Our data catalog strategy benefits the client in three ways: through
the transparent development, implementation, and iterative devel-
opment of the data catalog solution. First, the strategy increased
the visibility of the data catalog within the overall transformation
program by defining a clear way forward and providing key project
management parameters, including timelines, milestones, and core
deliverables. This groundwork enabled us to select, source, and
purchase the software as well as rollout a usable MVP implemen-
tation in quick succession efficiently and effectively. Second, thanks
to our strategy, we knew that the enterprise catalog was the right
fit for the client’s data mesh architecture, the defined data govern-
ance, and related use cases like data product creation. The fact
that we were able to embed the solution within the organization
early on unlocked tremendous synergies and ensured a successful
rollout at various company levels. Third, our agile implementation
approach provided economies of scale that enable the client to
add more use cases in subsequent design cycles. By starting with
key data catalog use cases, we delivered real benefit to the cus-
tomer and helped sustain management buy-in.

2
 Investing in Germany | A guide for Chinese

Value contributions
and way forward
In direct response to changing market environments, data tech-
• Embedding within the organization
nologies, and organizational paradigms, the real challenge facing
Data-driven transformation changes several company functions
today’s companies is how to mitigate the risks associated with
ranging from management control to operational execution. That
these impacts. Large-scale data-driven transformations have
is why it is so important for the data catalog strategy to look at
moved to the top of corporate agendas as a result or are already
the transformation in its entirety and ensure the solution
underway as companies look to take a more data-centric approach.
embed- ded within your organization is both effective and
As a major lever to unlock the value of enterprise data and drive
valuable. This includes a holistic data catalog roadmap that acts
data quality, retrieval, and reuse, the data catalog is an
as a central lever when it comes to addressing changes from and
indispensable part of
to other levels of the data-driven company that are evolving in
any transformation endeavor. Having the right strategy in place will
parallel.
ensure that your data catalog solution is closely aligned with the
The service operating model and playbook are also important
transformation program and its data governance design. As we have
for defining the roles responsible for delivering certain services,
outlined in this whitepaper, the key advantages of a data catalog
including the onboarding of stakeholders to the data cataloging
strategy and the role it plays in creating value are as follows.
space.

• Focus on high-quality use cases

• Quick wins for the transformation
Professional software products can build solutions according
Becoming a data-driven company takes time, and the tangible
to different data catalog archetypes. By specifying your com-
returns, generally through productivity gains and data mone-
pany’s exact requirements for the transformation journey, the
tization, usually pay off at a later point in time. However, if your
data catalog strategy points you in the right direction in terms
transformation program takes an incremental approach to ope-
of managing metadata in the data-driven company. Viable use
rationalizing data governance, a robust data catalog strategy can
cases for each different data catalog archetype are the guiding
deliver visible and viable results by improving the resilience and
principles for this strategic planning endeavor. They are the key
capabilities of the data-driven company. And thanks to the quick
drivers for a holistic metadata management solution that meets
wins that keep demonstrating the benefits of data governance as
your company’s needs today and in the future. While each trans-
it evolves, you can be sure of management buy-in throughout the
formation and data governance scenario will have its own requi-
entire transformation journey.
red use cases, we offer a more generic but highly valuable set of
use case templates in our approach to support every context.
Even the best plan is worthless without execution, so we need to
think of the data catalog strategy as the first step in an actionable
• Minimize costs, maximize return
implementation. To design and scale the right data catalog solution
In the short run, developing your data catalog strategy will
(Figure 7), you need to include the provided core deliverables like
obviously generate further costs for your organization as well
roadmap, use cases, service operating model, playbook, initial data
as your transformation program. However, when you select the
catalog backlog, and other conceptual assets in your strategic plan-
most viable solution in a timely manner, this targeted investment
ning. We make sure you have the ideal data catalog strategy for the
will yield returns by minimizing your overall spend. Deloitte’s best
subsequent execution and delivery phases when you implement the
practice approach for defining a data catalog strategy, selecting
target solution as required. By maintaining strong partnerships with
the right software, and thoroughly preparing for implementation
major data catalog vendors and providing a team with the
will significantly streamline project timelines and ensure the
necessary skills to deliver the solution you need, Deloitte is your
right solution for your particular transformation. In the long run,
one-stop shop for building and implementing a data catalog solution
choosing the right data catalog solution will significantly reduce
of any size and scope.
technical debts by delivering vital capabilities for your data-driven
company. This lowers both transaction costs for users shopping
for data as well as opportunity costs of meeting the changing
demands of data-driven use cases.

2
 Data catalog strategy | Operationalizing data governance to achieve data-driven

Fig. 7 – Deloitte’s approach to data catalog strategy, including additional phases for solution implementation and scaling

Data catalog strategy Data catalog implementation

ShapeScale
Scree Scop Sour Product installation· Additional demand
n e ce Infrastructure securityintake
DC use case MVP· DC roadmap
· Transformation · DC vision & · Vendor longlist implementationrefinement
program mission · RFI/Utility-benefit Testing, roll-out &· DC product backlog
overview statements analysis go-liverefinement
· Use case canvas · DC fit-gap analysis · Vendor shortlist Training concept· Organizational
· Requirements list · DC solution design Hypercare &embedding managed services· Training solution &
· Demos/POCs
practices
· DG/DC health check (incl. suitable · RFP/DC eval., recom-
· DG/DC models) mend. & selection
maturity · Target
assessment architecture
blueprint Set
· Use case prioritization
· DC roadmap
&
deliverables
· Scenario
impact
analysis
· Initial DC
productbacklog
· Initial service

Data domain
Enterprise managers
architects
Data catalog product owner
IT/Governance council
Data strategists
Business/Use case owners
Data Stewards/Custodians
Data Stewards/Custodians
Change management & communication strategy

2
 Investing in Germany | A guide for Chinese

Your contacts

Andreas Klein
Dr. Benjamin Klör
Partner
Manager
Tel: +49 173 2876392
Tel: +49 151 54484579
anklein@deloitte.de
bkloer@deloitte.de

Andreas is leading Deloitte’s Technology Strategy offering and the

Benjamin is working as a Manager for Technology Strategy &
CIO Program in Germany. He is a Partner from Deloitte’s Enterprise
Transformation at Deloitte. He is Topic Lead for Data Architecture
Performance practice focusing on digital/ technology strategies,
in TS&T’s market offering capability Architecture Vision. He acted in
IT target operating models, IT organization, tech cost optimization,
different roles to coordinate and deliver transformation projects
business process and IT sourcing and related complex transforma-
in various industries for building the data-driven company.
tions. Andreas gained experiences in diverse industries including
Benjamin’s professional interests are focused on the full lifecycle of
Financial Services, Manufacturing/ Automotive, Technology, Media,
data management, including data strategy definition, data gover-
and Telecommunications as well as Life Sciences & Health Care.
nance design, as well as data value creation by the means of data
Overall, he has worked over 15 years in management consulting.
monetization and modernization.

References
[1] R. Bean, "Why Is It So Hard to Become a Data-Driven Company?," 2021. [Online]. Available: https://hbr.org/2021/02/why-is-it-so-hard-to-
be- come-a-data-driven-company. [Accessed 16 01 2023].
[2] O. Sohail, P. Sharma and B. Ciric, "Data governance for next-generation platforms," Deloitte, 2018.
[3] G. Vial, "Data Governance in the 21st-Century Organization," 2020. [Online]. Available: https://sloanreview.mit.edu/article/data-govern-
ance-in-the-21st-century-organization/. [Accessed 10 1 2023].
[4] Forrester Research, Inc., "New Data Management Models Are Essential To Operate In The Cloud," Forrester Research, Inc., 2022.
[5] MIT Technology Review Insights, "Using governance to spur, not stall, data access for analytics," 2022. [Online]. Available:
https://www.technol- ogyreview.com/2022/09/15/1059242/using-governance-to-spur-not-stall-data-access-for-analytics/. [Accessed 27 01 2023].
[6] S. Judah and R. Gulzar, "Adaptive Data and Analytics Governance to Achieve Digital Business Success," 2020. [Online]. Available:
https://www. gartner.com/document/3987755. [Accessed 11 1 2023].
[7] Z. Dehghani, Data Mesh: Delivering Data-Driven Value at Scale, O'Reilly Media, Inc., 2022.
[8] A. Granzen, F. Giron, M. Goetz and S. Sjoblom, "The Forrester Wave™: Data Governance Solutions, Q3 2021," Forrester Research, Inc., 2021.
[9] S. Prukalpa, "Data Catalog 3.0: Modern Metadata for the Modern Data Stack," 2021. [Online]. Available:
https://towardsdatascience.com/data- catalog-3-0-modern-metadata-for-the-modern-data-stack-ec621f593dcf. [Accessed 12 1 2023].
[10] N. Jahnke and B. Otto, "Data Catalogs in the Enterprise: Applications and Integration," Datenbank Spektrum, 2023.
[11] G. De Simoni and R. Edjlali, "Magic Quadrant for Metadata Management Solutions," 2016. [Online]. Available:
https://www.gartner.com/docu- ment/code/297145. [Accessed 14 1 2023].
[12] G. De Simoni, M. Beyer, A. Jain and A. Dayley, "Magic Quadrant for Metadata Management Solutions," 2020. [Online]. Available:
https://www. gartner.com/document/3993025. [Accessed 14 1 2023].
[13] E. M. Broda, "Data Mesh Patterns: Enterprise Data Product Catalog," 2022. [Online]. Available:
https://towardsdatascience.com/ data-mesh-pattern-enterprise-data-product-catalog-ba4bf072d7c3. [Accessed 16 1 2023].
[14] E. M. Broda, "Data Mesh Architecture Patterns," 2022. [Online]. Available: https://towardsdatascience.com/data-mesh-architecture-
pat- terns-98cc1014f251. [Accessed 27 1 2023].
[15] M. Goetz, S. Sridharan, C. Provost and P. Dostie, "The Forrester Wave™: Enterprise Data Catalogs For DataOps, Q2 2022," Forrester
Research, Inc., 2022.
[16] O. Bieh-Zimmert, M. Engel and S. Kraus, "Cataloging Data - A capability maturity model for data catalogs," Deloitte, 2018.

2
Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited (DTTL), its global network of member firms, and their
related entities (collectively, the “Deloitte organization”). DTTL (also referred to as “Deloitte Global”) and each of its member
firms and related entities are legally separate and independent entities, which cannot obligate or bind each other in respect
of third parties. DTTL and each DTTL member firm and related entity is liable only for its own acts and omissions, and not
those of each other. DTTL does not provide services to clients. Please see www.deloitte.com/de/UeberUns to learn more.

Deloitte provides industry-leading audit and assurance, tax and legal, consulting, financial advisory, and risk advisory
services to nearly 90% of the Fortune Global 500® and thousands of private companies. Legal advisory services in Germany
are provided by Deloitte Legal. Our people deliver measurable and lasting results that help reinforce public trust in capital
markets, enable clients to transform and thrive, and lead the way toward a stronger economy, a more equitable society and
a sustainable world. Building on its 175-plus year history, Deloitte spans more than 150 countries and territories. Learn how
Deloitte’s approximately 415,000 people worldwide make an impact that matters at www.deloitte.com/de.

This communication contains general information only, and none of Deloitte Consulting GmbH or Deloitte Touche Tohmatsu
Limited (DTTL), its global network of member firms or their related entities (collectively, the “Deloitte organization”) is, by
means of this communication, rendering professional advice or services. Before making any decision or taking any action
that may affect your finances or your business, you should consult a qualified professional adviser.

No representations, warranties or undertakings (express or implied) are given as to the accuracy or completeness of the
information in this communication, and none of DTTL, its member firms, related entities, employees or agents shall be liable
or responsible for any loss or damage whatsoever arising directly or indirectly in connection with any person relying on
this communication. DTTL and each of its member firms, and their related entities, are legally separate and independent
entities.

Issue 08/2023