How to Spot Data Imposters – A Field Guide

Data is currently one of the most bandied-about words in business.

And data scientist is the “sexiest job of the 21st century” according to the Harvard Business Review.

It’s a hot topic, and it seems that everybody is jumping on the bandwagon.

This trend invites a lot of data imposters – people who seem to say all the right things to get you to buy something from them.

To nobly help protect you, the consumer, from being steered the wrong direction, MindEcology has created this simple 7-point field guide for avoiding hiring data imposters.

You know you are talking to a master data scientist if:

1. They speak plainly: People who know data don’t feel the need to impress you with their knowledge of jargon and buzzwords. In fact, they excel at getting right to the key concept that you need to complete the picture by using plain English, with technical terms used only when absolutely necessary.

2. They have subject matter expertise: A data scientist can hold a degree in math from MIT, but without subject matter expertise, it’s all theory. True mastery involves also knowing one or more areas of expertise where data skills can be applied, such as marketing, industrial automation, operational research, fish & wildlife population tracking, product quality testing, etc. Only theory + application makes data science a viable pursuit.

3. They understand statistics: The ability to operate a sophisticated data analytics or modeling tool is impressive, but one needs a solid background in statistics to make sense of results and to compare the validity of competing models. Without stats, they are flying blind.

4. They are multi-tool-fluent, tool-agnostic: There are many ways to analyze data, from Excel to desktop-based modeling apps to SaaS to cloud-based systems to open source tools powered by Python or R. There are also many types of models and analytical techniques, often useful for approaching the same problem. If your data scientist has a small toolkit, she’s likely to be susceptible try using a hammer to drive in a screw.

5. They don’t under-explain: People who don’t fully understand what they are doing might try to tell you to “leave it to them, they got it” when really they are just masking their lack of understanding behind the mystique of expertise. A good data scientist will keep you abreast of what she is doing – if and when you want to know.

6. They don’t over-explain: By the same token, a data scientist will not try to impress you by plowing you with textbook-like detail on what she is doing behind the scenes. Think of a good airline pilot: they let you know when they are taking off and landing but they don’t tell you about every lever they pull or button they push along the way.

7. They are collaborative rather than silos: A good data scientist works well with others, including clients, database administrators, business strategy experts, external data providers, finance people, software experts, their bosses, and people who report to them.

Some of these items are easier to spot than others, but they are ALL essential for having a productive relationship with your data scientist.