The Role of Data Analytics in Strategy

Eric Sandosham, Ph.D.
5 min readApr 14, 2024

Should you invite your data analysts to the strategy table?

Photo by Felix Mittermeier on Unsplash

Background

I was hired into Citibank’s prestigious management trainee programme in the early 1990’s, and because of the said programme, I’ve had the privilege and advantage of both observing and participating in a number of so-called ‘strategy offsites’. It all felt so glamorous. While I rose through the ranks over the years to become the regional CAO, I would have to admit that it was not common to see my one-downs or competitor bank counterparts being invited into strategy meetings.

Should you invite your data analysts to the strategy table? That’s a question I’ve been thinking about lately. With the rise of AI, it has injected a lot more exposure and credibility to the underlying ‘support’ function of data analytics / data science; giving them a seat at the management table. In an earlier article, I wrote about the necessary behaviour-based attributes that a data analytics / data science leader needs so as to either earn or maintain their seat with the ‘big boys’. But the strategy table is a peculiar one. Not everybody gets invited. And so I dedicate my 34th weekly article to the intersection of Data Analytics and Strategy.

(I write a weekly series of articles where I call out bad thinking and bad practices in data analytics / data science which you can find here.)

Front-Middle-Back

I’m a big fan of Professor Roger Martin’s Where to Play, How to Win strategy thinking (aka Playing to Win). It’s simple, clear thinking. Professor Martin has long argued that strategy is about the art of the possible; it’s about creating an outcome that has not yet manifest and therefore, there is no historic data to model on. He believes, and I concur, that strategy is akin to designing; it requires creativity and imagination. And so, you would typically invite the more creative types to strategy meetings. Of course the CFO needs to be there because he/she controls the purse strings, but you would typically not invite the Operations or the Compliance person to such meetings. So what is the perception of the data analytics person (i.e. creative vs non-creative)? What will the data analytics person or function contribute to the strategy discussion?

Let’s start by answering the question: “Is data analytics a front, middle or back office function?” Why is this question important? Because strategy is typically thought of as a front-office activity (i.e. market shaping). Back office folks typically don’t get invited, and the middle office folks that do get invited are those that can marshal resources to increase the odds of strategy success (e.g. the CFO). While I’m over-simplifying, we can think of front office as functions that ‘face-off’ with the market, shaping the interaction engagement with the customer, e.g. Sales, Product Management, Marketing. Back office functions are those that manage the assets and keep the lights running, e.g. Operations, IT. Middle office functions are those that coordinate and direct resources, much like internal traffic police. Functions such as Finance, Legal, Audit would be classified as middle office.

There is still no firm rule as to whether data analytics is a front, middle or back office function; its classification is both mired in its own history and its perceived utility. In Citibank Singapore where I worked, data analytics emerged out of a union of marketing support (front) and operations (back), leading to a middle office classification. In some other organisations, data analytics emerged out of IT, leading to a back office classification. Rarely has (enterprise) data analytics been classified as a front office function.

According to Professor Martin, your strategy must always be trying to move your organisation towards one of 2 possible positions: as a cost differentiator (meaning you can provide the same quality of goods cheaper than anyone else; think Walmart) or as a product/service differentiator (meaning the market is willing to pay a premium for your goods; think Apple). So the question on front/middle/back office can be rephrased as “What does the data analytics function bring to the strategy table to move the needle towards cost or product/service differentiation?”

Data Analytics Differentiation

I would argue that the data analytics function brings the following value to the strategy discussion:

  1. Keeping the conversation honest.
  2. Steering the AI race.
  3. Commercialisation of data analytics capabilities.

Keeping the Conversation Honest

The data analytics person at the strategy table plays the crucial role to course-correct flawed assumptions on market research where potential demand may be inflated, market segmentation based on questionable persona-based attributes, marketing effectiveness based on fallacious practices (read the book How Brand Grow by Byron Sharp). The data analytics person counters by asking for and reviewing the ‘evidence’.

Steering the AI Race

As the most data literate person at the strategy table, the data analytics person must enrich and tamper the strategy with AI capabilities, both from an employee-facing and customer-facing perspective. Knowing what can or cannot or should not be done with AI is an important consideration that must be brought to bear in every strategy conversation as we have officially entered the Age of AI, just like when the world entered the Age of Internet (all strategy discussions had to take into account how the internet will change the competitive and consumer behaviour landscape).

Commercialisation of Data Analytics Capabilities

Modern strategy should include the re-usability and external monetisation of internal assets. I’ve written a number of articles before on how organisations should build out their data analytics capabilities in terms of data capability, computation capability, and translation capability. Some of these capabilities may have the potential to be commercialised. Just as Amazon expanded with AWS, and Reddit looking to monetise their unique conversation threads for training large language (AI) models, there may be hidden opportunities to productionise some of the organisation’s data analytics capabilities into non-competitive areas.

Conclusion

Beyond earning that proverbial seat at the (management) table, data analytics functions should further strive to insert themselves into strategy discussions. They can contribute meaningfully to the discussions with their insightful take on data interpretation (i.e. calling out questionable evidence) and AI integration. They can bring a new practical perspective to strategy development and refinement that would be well appreciated.

--

--

Eric Sandosham, Ph.D.

Founder & Partner of Red & White Consulting Partners LLP. A passionate and seasoned veteran of business analytics. Former CAO of Citibank APAC.