Page 59 - The Real Work Of Data Science Turning Data Into Information, Better Decisions, And Stronger Organizations by Ron S. Kenett, Thomas C. Redman (z-lib.org)_Neat

46 The Real Work of Data Science

Traits of the Antis

We have also distilled six bad habits that stymie managers and companies from taking full
advantage of their data. We call these the traits of the “anti‐datas.” They include:

• preferring one’s intuition over the data to an unhealthy degree;
• rigging the decision‐making system (more on this in Chapter 11);
• second‐guessing others;
• becoming consumed by “analysis paralysis”;
• engaging in “groupthink”; and
• having deep misconceptions about data quality and/or exhibiting an unhealthy arrogance
concerning the quality of one’s data.

Again, most of these traits are self‐explanatory, but we wish to expand on two. First, it is
important to remember that the data can only take a decision‐maker so far. Then, their intui-
tions must take over. And good decision‐makers work hard to train their intuitions. At the
same time, we have all met managers who say things like, “I’ve been working in this industry
25 years and I’ve seen it all. I know I can trust my gut.” They are proud of their experience
and are skeptical of anything new. Interestingly, we find many managers who behave this
way to be solid in most respects – they care about their companies and people. They desper-
ately want to do the right things, and they are smart. But they go to great lengths to ignore,
downplay, or subvert any evidence that suggests a better way. Some even reinterpret the data
to reinforce their long‐held mental models. The near‐certain result is decisions that are
increasingly out‐of‐date.
Second is second‐guessing. In its worst form, second‐guessing involves withholding poten-
tially useful information, then pouncing the minute a decision goes wrong. In some ways, it is
natural for people competing for that next promotion to engage in second‐guessing. Further,
The Forty‐Eight Laws of Power (Greene and Elffers 1998) advises those seeking power to
withhold information. One observes this trait all the time in overly political individuals and
companies. Data scientists should treat second‐guessing as a political reality. It underscores
the importance of understanding the full range of decision‐making strengths, weaknesses, and
biases of those you advise.

Implications
It bears mentioning that, at either the company or individual level, becoming data‐driven
requires deep cultural commitment, self‐reflection, training, and lots of hard work. Why
should a data scientist or CAO care? We have already mentioned the three most important
reasons. While everyone, at both the individual and company level, likes to fashion themselves
as “data‐driven,” the reality is far different. Decision‐makers are only human, and you should
help them understand their strengths and weaknesses. Start by taking a hard look in the mirror.
Find someone who will tell you the truth; then rate each other on all 18 traits. You will almost
certainly find it a sobering experience.
Second, we encourage you to work to advance your company’s decision‐making capabil-
ities over the long term. As we’ve already noted, many people readily admit that “statistics
was my least favorite course in college.” So, start simply. We have summarized some exercises
that have yielded good results for us in Chapter 12.