Four Reasons Why Managing Data Scientists Is Different. . . Even Unique

Data scientist
Managing smart professionals can be difficult, but managing data science professionals can be even more difficult. Cultural differences–understanding how a culture thinks–are rarely clear to most business people, other than when they surface in difficulties. So understanding these differences early on limits the potential for friction and conflict.

MIT’s Roger Stein lays out a number of issues impacting organizational relations between data scientists and other disciplines.

Dynamic and self-correcting.  As a result, timing and final project outcomes can be difficult to plan. Messy is the appropriate term. In contrast to more subjective and non-data projects, you’d think that data project timing and outcome would be quite predictable. But the opposite is the case.

Data research versus line managers. Since most line managers from non-techie business units aren’t familiar with data mining or analytic modeling, it’s tough for them to assess the quality or the timing of a project. Of course, data scientists have the same problem in reverse. Miscommunication and misunderstanding between data scientists and other business folk are par for the course. They’re what I call “classic breakdowns.” Inevitably, underlying problems are interpersonal breakdowns–not hard data. So it’s always important to make certain everyone is reading from the same sheet of music. You won’t notice things that are always in front of your eyes. . . like communication differences.

The TCQ triangle. New projects inevitably try to balance time, quality and costs. It’s nearly impossible to maximize all three of these attributes, but quality is most likely to suffer lack of attention. Stein puts it this way: If you want to do a project quickly and cheaply, quality is likely to suffer; if you want to do the project quickly and to a high stand, it will probably cost more; if you want to do the project cheaply but at high quality, it will take longer.

Collaboration is a big deal. For the best results, it’s very important for data scientists to be as clear as possible about TCQ, and also help business teams understand what’s possible, the various options and the consequences. And collaboration? That’s all about effective talking.

Stein, like 99% of business folk, largely ignores the huge role that interaction skills play in the success or failure of projects. The collapse of literacy among adults—and especially among the data-oriented—is a perfect setup for data failure. So these skills, usually ignored, must. . . absolutely must be brought to the table for companies to get the most out of data science. Interactional skills underlie all business decisioning.

, , , ,

Comments are closed.