Reward data scientists' good work with significant recognition

A structured rewards program for data science teams should be significant enough to make a difference, timely, and actually happen. Get advice on setting up such a program.

Image: iStock

You'd think that just the honor of being a data scientist in your prestigious company would bring them enduring happiness and joy, right? I'm being a bit sarcastic, though many leaders feel it's a privilege to be a data scientist at their company, and the mere role should suffice to keep their data scientists happy -- notwithstanding a handsome salary. It stands to reason with all the hype these days around how special it is to be a data scientist.

In truth, the novelty of landing the coveted data scientist position wears off in about three to six months, regardless of what you're paying them. So, to sustain the right behaviors from your data science team, make sure you reward them properly.

Is that all?

A structured rewards program is an important part of any functional data science team. Like everyone else, data scientists need to feel like they're appreciated for good work.

I was recently involved in a high-profile data migration that went extremely well. During this high-stress period we made it fun with a friendly competition that included some nominal prizes. After it was over, everyone involved received something a little more significant for his or her contributions, including a lot of attention from top management. Several weeks later, I still get comments from participants about how much they appreciate their reward.

The key that made this event work is the significance of the reward after the fact. This is my first piece of advice when structuring a rewards program for your data scientists: the reward should be significant enough to make a difference.

Remember when...

Secondly, be sure your data scientists are rewarded in a timely manner. Most of what data scientists do involves solving difficult problems, which is neither linear nor predictable; as such, it can be very frustrating at times. And suffice to say, solving a difficult problem provides a significant internal reward, but receiving an external reward offers a much different sense of satisfaction -- as long as it's done in a timely manner. If you wait too long, your reward will still be appreciated, but not in the same way; the association between the reward and the behavior becomes weak.

Let's say you've been struggling with cohesion between your qualitative and quantitative teams. However, six months ago they did pull together to solve a very difficult problem. Since then, you've been working with HR to figure out the best way to reward them for their teamwork. Unfortunately, it's too late. You should still reward them to show your appreciation, but you missed the boat on encouraging teamwork -- the reward/behavior association won't stick. You'll have to be more diligent in your timing next time.

The check is in the mail

Finally, you must execute on your promises. It's good to be spontaneous with rewards, but it's more effective if they're planned -- that's why you must be intentional and structured about your rewards program. This provides two benefits: It greatly increases the chances that rewards will show up in the first place, and it lets the team know that good behavior will be rewarded. There's a catch, though: If you don't follow through, your plan will backfire.

A principal component of structuring an effective rewards program is certainty. Your team must believe that they'll actually receive a reward for behaving a certain way. If you get them excited about a party that never happens, or a bonus that never hits their bank account, or a few days off after a critical, 36-hour crunch period, your credibility will be shot. One bad move can eradicate 10 good moves, so don't let that happen.


The thrill of becoming a data scientist provides an amazing sense of euphoria, excitement, and accomplishment for about six months -- after that, it's important to be intentional about how you shape and reward their behavior. Work with your organizational change management resources to structure a comprehensive rewards program that your data science team can be excited about, and then, work with your analytic manager to make sure it happens. If you don't, your data scientists may find their happiness somewhere else.