top of page
  • Writer's pictureAdam Votava

Does empathy play a role in being data-driven?

An important soft skill in the world of hard numbers

It was demonstrated almost ten years ago, that data-driven decision making is superior to HiPPO. Since then, being data-driven has become an ambition for many companies and individuals alike.

“Companies in the top third of their industry in the use of data-driven decision making were, on average, 5% more productive and 6% more profitable than their competitors.” — Andrew McAfee and Erik Brynjolfsson

In order to be able to make data-driven decisions, one requires a certain level of data literacy, though I prefer the term: data fluency. One needs to be able to think in data. But don’t be mistaken into thinking it’s a purely technical domain!

When striving to be data-driven, people can sometimes get overly excited and start using data as a dogmatic proof to their point— “just look at the data!”

It is not uncommon, that people presented with same data points make completely different conclusions. Actually, we see this very often — take Covid-19 and the debates on measures, restrictions, or vaccinations, etc.

And I’m sure that even in your professional life you’ve been in heated discussions among smart, experienced, and respected colleagues, who were making vastly different judgements from the same evidence available to them.

So, what’s going on? Data doesn’t lie! Right?

Let me offer my two cents here. I believe there are four main reasons:

  1. Our brain is being constantly tricked by all sorts of cognitive biases, such as (to name a few) anchoring, confirmation bias, status quo bias, gambler’s fallacy, survivorship bias, IKEA effect, clustering bias, and — worse of all — blind spot bias. Is it even fair to expect people be rational?

  2. Data is an imperfect reflection of reality. Our complex world cannot ever be perfectly represented in finite-dimensional data. No matter how hard we try, there will always be important factors missing, many complexities simplified, and the quality of available data can never be taken for granted. Aren’t we like the inmates in Plato’s cave?

  3. There are often many competing goals. What is it that we are solving for? Are we after short-term or long-term benefits? Do we want to drive company revenue or sales of a particular product? Which statistical error is worse: Type I or Type II? Misalignment on the goals can lead to great misunderstandings.

  4. Microeconomics teaches us, that choices made by individuals are influenced by their preferences. And because the preferences of each one of us are different, we make different choices and decisions. We might be so scared of certain outcomes that we want to prevent them at all costs, no matter how unlikely they are. Some of us may consider a $10,000 road bike a great deal, others wouldn’t pay $100.

Looking at these four points, it certainly doesn’t feel like a given that two people — both making decisions based on data — will reach the same conclusion.

And when they don’t, it doesn’t mean that one of them isn’t data-driven enough! One might be subject to a bias, misunderstand the data, optimise for a wrong goal or simply have different preferences. But both are convinced about their conclusion.

What could help in these situations? Being empathetic and open-minded!

  • Ask questions. What is the problem we are solving? Are we missing any important data? What could be wrong with the data? These questions don’t always have to be phrased as questions — ‘there seems to be something missing here’, or ‘let me try to summarise my understanding of the objective’ — might do the trick.

  • Be genuinely interested in what drives the other side. What could be their objectives? What is their experience? What are they preferences? ‘It feels like you have a different opinion. Can you talk me through it?’

  • Admit that you’re biased. It is perfectly plausible that you have a blind spot. Or that you are missing something obvious. Can’t others help you identify the blind spot? ‘Can you try to shoot holes in my analysis?’

  • Support people with different opinions. Sometimes, there are more people in the room and an opinion minority might not have enough space to voice their concerns. Create a space for them. ‘She has a point! Let’s discuss it.’

And there are many more! We often live in our own bubble — surrounded by people with similar experience, opinions, and preferences. The best way to learn, practice and master empathy is to have many hard conversations, seek talking to people with different opinions. And read. A lot.

The more experience you have, the more you have seen, the more varied approaches you have tried, the less susceptible you are to a tunnel vision and being trapped in your own (small) world.

I would argue, that being open-minded and empathetic can help clarify the differences, find an alignment and ultimately make the best possible decision. Being genuinely interested in what has let others to different conclusion is extremely useful skill.

You might even identify a flaw in your own reasoning.


This article was first published on Medium - Towards Data Science

Photo by Md Mahdi on Unsplash


bottom of page