Letter to the CEO: Roger Martin

Dear CEO

The most pressing challenge facing the CEOs of today’s corporation is incursion into corporate decision making of data analytics, commonly referred to using the fashionable moniker ‘Big Data.’ The state of play is that data analytics is considered fully above reproach: something that modern CEOs simply must embrace. If a CEO doesn’t show unqualified reverence for data analytics, it is assumed that the CEO is a Neanderthal and/or Luddite. What has changed is that data analytics has migrated from the fringes of CEO life to the very epicenter. It is now the hottest thing in business. CEOs are increasingly faced with an endless string of well-meaning but unreflective data analytics enthusiasts telling them that the ‘data prove that X is true’ or the ‘correct decision based on the data analytics is to do Y.’ The absolutely dominant prevailing wisdom is that CEOs should thank the messenger profusely and affirm that the decision based on data analytics is right. CEOs should instead ask the messenger the following five questions:

Question 1: From what era does all data in the world come?

Answer: From the past. There is no data about the future – yet.

Question 2: What is the full extent of what data analytics tell us?

Answer: What has been operative in the past based on how the world has worked in the past.

Question 3: What is our implicit assumption each and every time we use data analytics to decide what to do going forward?

Answer: We implicitly assume that the future will be a direct extrapolation of the past. It will either identical to the past or an extrapolation of the observed past trend into the future.

Question 4: What is the probability of making choices to create a future that is different from the past using data analytics?

Answer: Zero. Data analytics has zero ability to chart out a course that is anything other than an extrapolation of the past into the future.

Question 5: What is the probability that making a choice about the future based on data analytics will turn out badly?

Answer: High. Last time I checked, frequently the future turns out to be unexpectedly different from the past – annoyingly so, in fact.

The strong likelihood is that the big data enthusiast will not be able to answer any of the five questions, be baffled by the nature of the questions, and declare the CEO to be ‘anti-analytics.’ But by asking the questions and insisting on answers that demonstrate that data analytics are appropriate for the situation in question – and data analytics is appropriate when the future is likely to mirror the past – the CEO will be saving the company from the modern day vandals.

Instead, CEOs need to use the only methodology that has ever been useful in making decisions about the future: first, imagine possibilities and second, pick the one for which the most compelling argument can be made. In deciding which is backed by the most compelling argument, CEOs should indeed take into account all data that can be crunched. But in addition, CEOs should also use imagination, judgment, and experience of numerous data points from the past that the data analysts wouldn’t consider ‘objective data’ to decide in what way to shape the future – like all the great CEOs in the history of business have done.

In doing so, CEOs will have to accept widespread ridicule in their organizations among the legion of big data enthusiasts, who will say that their CEO lacks rigor and makes decisions on ‘gut feel’ and is ‘old school.’ But these enthusiasts are likely to be blind and to have never asked questions concerning the logical limits of their methodology. So CEOs need to stand strong and make decisions that can create a better future for their organization and for humanity.

Sincerely,

Roger L Martin

This is an excerpt from Strategy@Work, a Brightline and Thinkers50 collaboration bringing together the very best thinking and insights in the field of strategy and beyond. This letter first appeared in the Thinkers50 book, Dear CEO (Bloomsbury, 2017).