Chapter 4 Reading guide

Rohith Modem

Anderson University

Analytics and the Enterprise

April Adams

Feb 27, 2026

from the Enterprise IT function?

According to Sah (2022), data and analytics initiatives are run in a totally different way in

contrast to the conventional enterprise IT initiatives. The majority of IT projects like ERP or

CRM implementation have structured processes in which requirements can be defined in

advance and then the development starts. These systems have automated business processes

that are already known, which makes the process of planning relatively predictable. On the

contrary, projects of data and analytics are exploratory. Leaders might start off asking general

questions and not precise specifications and gradually with the analysis of data they start to

understand more.

Due to this ambiguity, analytics teams need the flexibility, exploration, prototyping and

high-speed collaboration with business partners. Classical IT departments are known to have

high levels of governance, have long approval cycles, and cumbersome development models

likely to slug innovation. By integrating analytics into IT, it is in danger of being seen as a

technical support feature, as opposed to strategic capability.

The separation of data and analytics with enterprise IT enables the function to be more

business-focused in its priorities and to be more agile. It focuses on value creation, insight

creation, and decision support as opposed to maintaining the system. According to Sah

(2022), this structural difference makes it more responsive, leads to better stakeholder trust,

and allows analytics to work as a strategic enabler rather than a back-office technology unit.

2. What are the 3 different organizational models recommended in the textbook?

Describe each in your own words (no cut and paste!).

As Sah (2022) notes, enterprises usually have three models available to them:
Decentralized Model

This design features regions or business teams that have their own analytics team that are

independent of one another. Local teams are free to address the needs in regions. Although

this is favorable to responsiveness, it can lead to the inconsistency of standards, tools, and

data definitions across the enterprise.

Centralized Model

The analytics activities in this model are controlled by a single core team, usually the one

that is located at headquarters. This facilitates standardization, economies of scale and a more

powerful governance. Nonetheless, it could not easily respond rapidly to local needs that are

unique to the area (Sah, 2022).

Federated Model

This model is an integration of centralized control and regional flexibility. Enterprise

wide standards and common capabilities are handled by a central team and local needs by

regional teams. It focuses on finding the balance between consistency and autonomy, but

high coordination within central teams and local teams is required.

3. Of the 3 models above, which one does the textbook author recommend as best? In

your own words, why is it supposed to be better? Do you agree?

According to Sah (2022), the federated model is suggested by the author to large and

complex global enterprises. This structure is a combination of centralized control and

decentralized flexibility. It enables common investments of technology, common standards

and utilization of solutions and yet remain responsive in the locality.

allows collaborating across the enterprise. It might cause tension between the central and

regional teams, but good governance and leadership would be able to control them.

I agree with the suggestion. Global alignment and local agility are necessary in large

organizations. The extreme of a centralized or decentralized structure will most probably

bring about inflexibility or disunity. The federated model offers an equilibrium model when

well managed (Sah, 2022).

4. What are the 5 towers of an ideal data and analytics organization? In your own

words, describe each.

According to Sah (2022), five logical towers are described:

Governance Tower

In charge of strategy, demand management, roadmap planning, risk management, change

management and performance monitoring. It provides conformity with enterprise objectives.

Business Tower

Serves as the liaison between stakeholders in the business and the technical departments.

Members are aware of business priorities and analytics. They co-innovate and make solutions

provide quantifiable value.

Technology and Architecture Tower

Determines enterprise data architecture, manages standards and promotes data

governance. It provides well-designed, scalable and secure analytics infrastructure.

Solution Delivery Tower

Manages analytics initiatives through design to implementation. This team develops and

tests applications and passes them over to them to maintain.

Manages and maintains deployed analytics solutions. It operates stability of operations,

ticketing, improvements, and improvement.

5. Describe in your own words why data governance is so complex.

Data governance is a complex concept since the enterprise information is in different

operations and systems in various geographies. The way all business functions produce and

use data differs and in most cases, has different definitions and quality standards. According

to Sah (2022), the data governance is not a technical task; it must be owned by the business.

Data custodians are the business leaders and IT the supporting role. Such mutual

responsibility complicates the organization.

Data Governance entails well-set functions like data owners, data stewards and governing

council. The policies should specify the creation, storage, access, protection, and utilization

of data. Also, businesses have to deal with regulation, privacy needs, and data security

threats. It is more challenging to align these components in international business.

Cultural resistance is another problem. Departments can be reluctant to share information

or use common definitions. The executive sponsorship is necessary since the governance

initiatives frequently presuppose the behavioral and process modifications. Devoid of

effective leadership determination and cross functional cooperation, governance systems can

be on paper but not working in the field. Due to all these reasons, data governance is a

technical and organizational change project (Sah, 2022).

6. In your own words, how are Solution delivery and Service delivery related?

Solution delivery and service delivery are two phases of the analytics lifecycle that are

closely related to each other. The solution delivery tower is dedicated to the design,
construction, testing, and implementation of analytics application in accordance with the

approved business requirements. This stage is project-based and it is commonplace to have

irregular workloads depending on the quantity and size of initiatives being undertaken (Sah,

2022).

When a solution has been deployed and its users have adopted it, the service delivery

tower becomes the responsible party. This department makes sure that the application is

functional even after. They have responsibilities to monitor the performance of the systems,

schedule data refresh process, address user problems, make minor improvements, and create

operational stability.

The association between the two towers is acute. Service teams will not be effective in

maintaining solutions without proper knowledge transfer, documentation and

communication. In other organizations, and particularly those who embrace the DevOps

practices, the two roles can be merged into one cross-functional team. Nevertheless, in these

situations, the difference between creating new capabilities and sustaining the ones is still

significant.

Sah (2022) emphasizes that the transition between these towers should be well-timed so

that the sustainability, continuous improvement and business value fulfillment due to

analytics investments can be obtained in a long-term perspective.

7. From the week in the life section of chapter 4, in your own words, which aspect

seems to be the most challenging? While this answer can be subjective, the

reasoning behind your answer is the crucial part of this answer.

Balancing between competing priorities and continuing innovation and operational

excellence seems to be the most difficult part. Sah (2022) points out that the analytics team is
required to process incoming demands, business impact, project prioritization, development

work, and business maintenance, user training, business governance, and business value

measurement simultaneously. This leaves a balancing act at all times. Different functional

stakeholders can make urgent requests and each of them can prove to be of high priority. The

team needs to make dispassionate decision making on priorities without ruining the

relationships. Simultaneously they have to keep on experimenting with new analytical

concepts and testing new solutions as they would push the future value (Sah, 2022).
 Reference

Sah, P. (2022). Defining enterprise data and analytics strategy. Springer.