The voice of data among C-Level executives — A book review of The Chief Data Officer’s Playbook

I had my birthday a couple of months ago. It was the beginning of the quarter and the work schedule was fairly compact — I didn’t have the luxury of taking a day off, so I was still working and had workshops for quarterly planning on my birthday.

One of my VP data knew about my birthday, and decided on the spot that we should have a celebration — so we went to a neat rooftop bar close to the office and had drinks there.

Having a group of colleagues drinking together usually leads to more work talk. Given that the participants that night are mostly the data team, the conversation is more about data. Most of the data discussion is intense — let it be data engineering, data science, or data visualization, but the area of data leadership remains an under-charted ocean.

Yet, this book came up while we brought up this topic:

The Chief Data Officer's Playbook

Being intrigued by the topic, I decided to buy a book for myself as a birthday gift.

TL;DR of my takeaways

If you are super busy, then the following key points are the summary of my takeaway from the book:

• CDO is an executive, much like other members of the C-Suits: CEO, CTO, CFO, COO, etc.
• Engaging business is the key to the keys
• Relationship building is a vital skill for a CDO, whether upward, peer, or (in)direct reports
• Expectation management especially upward & for non-technical laymen, is one of the key parts of the CDO job
• Business Relevance suppresses shiny, cutting-edge tech
• BAU (Bussines as usual) is more important than you think

So, what is the book about?

In Amazon, the description is short & sweet:

This book addresses the changes that have taken place in ‘data’, in the role of the ‘CDO’, and the expectations and ambitions of organisations.

The CDO (Chief Data Officer) role is the executive leader of the data role. Traditionally speaking, an organization would have a CEO, COO, and CFO. As for tech companies, there often come with CTO for Tech and CPO for Product. The rationale behind this is what the organization (or, to be completely honest, the C level) actually sees as the utmost importance: Any traditional company would indeed have business operations and the finance components constituting the very foundation of the organization. as for tech companies, CTOs and CPOs are given the mission of taking care of newly emerging critical components: Tech and Product.

Given that in the context, data has always resided as a “by-product” of other functions. I’ve seen data under various functions: under operation (which usually implies data maturity is not quite there yet), Finance (a bit better but also implies that the organization is not fully aware of the potential of data yet), or Tech (which is a more mainstream approach).

The mentality behind having data reporting to any functions mentioned above is that “data would serve as a subset of that function”: Being under Operation implies data would be used to optimize business process efficiency; Being under Finance implies data would be used for cost control; Being under Tech means data are to be used for tech development, and that is the closest position of senior management seeing data as an investment rather than another cost center.

And now comes CDO — the most senior data leader who sits among the other C-level and directly reports to the CEO. What does it entail? What does it say about the way the organization sees the value of data? What are some unique challenges and opportunities for CDOs? The book shares a light of the distinct journey to a wider audience.

Table of Content

The following is the table of content of the book. This shall give a pretty good overview understanding of what is covered:

1. The accidental entrepreneur
2. A reflection on the first 300 days
3. Why does any organization need a Chief Data Officer?
4. The secret ingredients of a Chief Data Officer
5. The first 100 days
6. Delivering a data strategy in the cauldron of BAU
7. Avoiding the hype cycle
8. Relating to the rest of the business, especially the C-Suite
9. The Chief Data Officer as a disruptor
10. Building the Chief Data Officer team
11. The next 300 days
12. The different generations of Chief Data Officers
13. What type of Chief Data Officer are you?
14. How to present yourself as a Chief Data Officer
15. The Chief Data Officer and the technology
16. The hoarding mentality and how to break it
17. Data and information ethics
18. The Chief Data Officer and data governance
19. The data revolution
20. Advice to business owners, CEOs, and the board

Why does an organization need a CDO?

A CDO is not a traditional role. An organization wouldn’t be aware of their need of having one until the following data issue accumulates to a certain scale:

• No “Lessons learned”, even after mistakes made upon a complex project completion
• No proper ETL after projects, only a bunch of Excel files/Google Sheets or other spreadsheet solutions — it sucks up huge resources & time and leads to commercial impact
• Fragmented & misaligned information across different spreadsheets without proper version control
• Messy Data Lineage (i.e. what the heck are behind reported KPIs?)
• Data sitting in silos across different departments, and there is no effective way to share data
• The CEO wants to run a data-driven organization
• Data talents spend more time firefighting data issues rather than actually delivering value

All of the above are happening to all large & complex organizations. These reasons are driving the need of having an executive who solely focuses on data.

Data Governance

While an asset grows to a certain scale, it needs to be properly managed. Data is no exception — and considering the hyper-growth of the volume of data, now is even more critical than ever to have proper Data Governance in place.

The book quoted the definition from TechTarget (aka whatis.com):

Data governance (DG) is the process of managing the availability, usability, integrity and security of the data in enterprise systems, based on internal data standards and policies that also control data usage. Effective data governance ensures that data is consistent and trustworthy and doesn’t get misused. It’s increasingly critical as organizations face new data privacy regulations and rely more and more on data analytics to help optimize operations and drive business decision-making.

And the key components for Data Governance:

• Policies: Key ways to work with data & deliver insights
• Process: Details of how the principles and policies will be applied and enforced
• Organizational Design: Assignment of ownership of data and responsibility for data
• Data Architecture & Design: articulation of location, lineage, and relationship for key data
• Technology: The tools required to manage data and provide the standardized reporting

There are also a couple of simple principles coming in handy while addressing Data Governance in an organization:

• Consistency of data without unnecessary duplication
• The quality is proactively assessed and standards
• Ownership and accountability are defined across the data life cycle and recorded data catalogs
• Business alignment which ensures that data is regarded and treated as a key business asset
• Access to relevant users, kept secure through access control without locking data down for no purpose
• Providing trusted insight

Organizations are advised to get inspiration and take whichever is applicable to their state of data.

Data Strategy

A majority of CDO’s mission is to deliver value from data while the organization continues to operate with the current data (Business As Usual, or BAU in short). It is one of the most difficult tasks for a CDO, and it is essentially repairing a plane on the fly.

The BAU in the organizations are typical:

• Legacy data environment: silos of data, duplicated data
• Legacy systems, burning platforms, bespoke developments
• Legacy business process: many workarounds, too dependent on end-user-compute & spreadsheets
• Multiple suppliers of software
• Legacy IT department: focus on building stuff instead of delivering value, internal networks as opposed to cloud
• Legacy “transformation” process: based on project governance & waterfall, struggling with Agile & Innovation, etc.

There are also 2 folds of Data Strategy: Immediate Data Strategy, and Target Data Strategy. For either strategy, it is vital to bear in mind that:

• Leverage internal communications to sell the vision; don’t allow a vacuum to form
• Seek every opportunity to communicate the vision; do not be frightened of becoming a data bore, think cheerleader instead
• Socialize the data vision and the changes that could be coming, especially the controversial ideas; locate the data champions to support you
• Engage the organization’s leadership and find your senior sponsors; they will be crucial
• Explain it; If you can’t explain it, you’re doing something wrong — “it’s me not you”
• Win hearts and minds; Often a logical argument or business case is not enough to win the day; Use your storytelling ability to the maximum

The focus of the Immediate Data Strategy is tactical:

1. Delivery support for BAU
2. Gain quick wins
3. Temp fixes
4. Prep the way for the second part: Target Data Strategy

The Immediate Data Strategy is more about listening to the organization’s pain about data and trying to deliver high-profile quick wins. And the goal would be to build up the narrative and vision of the Target Data Strategy.

The 6 key elements of the Immediate Data Strategy are:

1. Stability & rationalization of the existing data environment
2. Data Culture & Governance
3. The existing & immediate data & IT development initiatives
4. Data exploitation & integration, getting value from the data assets, and finding ways
5. Data Performance, quality, integrity, assurance, and provenance
6. Data Security (e.g. GDPR)

On the other hand, the Target Data Strategy is more long-term:

• The focus would be more on the Information Strategy, rather than just Data Strategy
• Swift much more on the value side and what data is used for
• Look at the Information vision and the principles that underpin it, while also looking at how to deliver this

In comparison, the Target Data Strategy is more abstract and more contextualized. It would vary much from organization and organization, and it takes deep experience and understanding of the organization to formulate one.

CDOs — Generations & Type

There are several generations of CDOs: The First Generation CDO (i.e. FCDO), the Second Generation CDO (i.e. SCDO), and so on.

The FCDOs are the very pioneers of data leadership in the organization. All of a sudden, the organization has invested in a key asset (FCDO) and something is expected to happen right away. The FCDO needs to understand the big picture, get the lie of the land and get on.

The 4 key areas need to be focused on:

• Governance
• Information Architecture
• Engagement
• Building Capability

And of course, the quick wins.

Also, there are a couple of critical traits to have as a specialized as FCDO:

• Optimistic
• Full of energy
• Able to think around problems & solve them
• Able to face the same issues & problems as they move from position to position

As for SCDOs, the organization is sold the vision of what can be archived with great Data Governance, and all the value derived from data — and the SCDOs need to deliver them.

The push for FCDOs still exists for SCDOs. If the FCDOs have set up a good foundation, then SCDOs could rely on those and build on top.

There are also certain possible ways of SCDOs could bring additional value to the table:

• Advanced Analytics & Data Science
• Data Narrative
• Data Visualization
• Publishing Data
• Data Refining

Compared to FCDOs and SCDOs, Third Generation CDOs (TCDO) are more established than pioneering in the organization. They are more normalized, and some transit into other C-Suite positions like COO, CFO, or even CEO roles as the foundation of data is solidified enough and does not need that intensive attention from C-level.

Types of CDOs

For CDOs, there are typically different backgrounds and patterns of them:

• CDOs have a fairly strong technical background, which is similar to CTO
• CDOs being data/information-led, who major in data models, data architecture, and data lineage
• CDOs whose strengths from good general management & leadership skills
• In certain cases, there are also CDOs who are entrepreneurs/start-ups/innovators/disruptors, who could lead the business to try lots & fail fast and also create a blame-free culture
• Also, there are CDOs coming from analytics backgrounds with strengths around finding insights into data

By knowing the type of CDOs, one could:

• Play to the strength
• Identify weaker skills
• Address the labor market demand

First 100 days, and also the next 300 days for the CDO

During the first 100 days for the CDOs, there are certain topics to be addressed:

• Make the case for change
• Vision & Strategy
• POTI model
• Data Basics
• Quick Wins

Make the case for change

There have to have certain pain points bugging the organization regarding data( e.g. siloed data, same work repeated over and over) and the case needs to be articulated.

The book has proposed a Maturity Model can be used to access the data maturity in the organization:

• Strategy
• Leadership
• Corporate Governance
• Framework
• Policies
• Information Risk
• Architecture
• Organization, roles, and responsibility
• Metrics
• Skills
• Behavior
• Tools

Nevertheless, there are many more robust models across the Web. This one is more like an inspiration.

Vision & Strategy

The POTI model is proposed as an example to provide high-level direction for what needs to be set up during the first 100 days of the CDO:

• Process: What can be improved with data in the business processes?
• Organization: What kind of people & skill change is required to make the improvement happen?
• Technology: What kind of tech & tools would help for the improvement?
• Information: How would information flow after improvements?

Data Basics

The 3 main areas are summarized to focus the CDO’s effort:

• Governance: i.e. Data Governance, and the policies with it
• Information Architecture: Who are the domain knowledge experts?
• Engagement: Where is the network of evangelists to sell your messages?

Next 300 days

After the first 100 days, certain quick wins should have been hit. And there come to the next 6 points to measure against for the next steps:

1. Reporting & Analytics
2. Information Flows
3. Data Governance
4. Data Management
5. Data Organisational Design
6. Data Technology

And, there should be two important milestones hit:

1. Delivering Immediate Data Strategy
2. Drafting Target Data Strategy

Final Words

All in all, the book is fairly Europe/UK contextualized. Given the state of data of organizations, it will be fairly different to put the same ideas in the context of Asia, or the Americas, for example. The book also spends quite some paragraphs during the early stages of CDOs. It is more tailor-made for organizations that just officially kick-started their journey of capitalizing data, and there isn’t enough content for more well-established organizations in data.

Personally, I believe that data is still worth C-suite attention. It takes specialized knowledge and experience to effectively capitalize, utilize and mobilize data in the organization. Depending on the industry, the stage of the organization, maturity etc., different techniques and management skills are required to leverage data as one of the most critical assets in the organizations. And it is not something that can be easily taken over by other C-Suites without prior knowledge and experience.

Nevertheless, this book still serves as a good starting point to see data from a data leadership point of view — which is still a fairly new section in the data industry.