How to Retain Data Scientists
Free Food and Aeron Chairs vs. Tools & Trainings
Today, we live in a world where employers are willing to invest a lot when it comes to employee comfort in a literal sense, and especially for data scientists’ comfort, particularly when they start asking the question: why are data scientists leaving their jobs? Quite simply, they believe that investing in fancy perks will make data scientists stay.
Data scientists by nature are curious and entered the field in order to make a difference and to have impact. If all they’re doing is playing with data in a sandbox and they never get to see their projects have impact on real data, news flash: they aren’t going to be happy in their job.
And yet despite this, when it comes to employee productivity — namely data team productivity — employers are hesitant to make any investments at all (or they fail to consider the ROI in this area completely).
To break this down for a minute, let’s take a look at some (very nice) benefits that certain companies — especially the ones attracting the best data talent — have:
- Free meals: Assuming 200 work days per year (which is low, even in countries that don’t take much vacation) and $10 a day for meals, that’s $2,000 per year per person spent on this perk.
- Nice location: The average office costs $8000 per employee per year (with huge variability from one country to another). Kick that up 50 percent for a fancy spot, and an attractive office + location is $4,000 per employee per year.
- Fancy chair/ergonomic arrangement: Add at least $2,000 per employee to provide them with furniture (that lasts only about five years).
And after all that (a $6,000 to $8,000 investment per employee per year), talent is still leaving. So what kind of investments could be more efficient? And more importantly, how can your company make investments that make sense?
Let's try do dig into the mind of a data scientist. Now, full disclosure: I haven’t been a data scientist for quite some time (and back in the day, we weren’t even called data scientists yet), but I can hopefully still relate to it enough to give some insight:
Data scientists are driven by knowledge, like learning a new language or a new technique. Training, classes, or workshops on those new techniques could make sense, which means an investment in classes, seminars, or the like. Do you have your own data science academy inside your organization? If not, why not?
Data scientists are driven by efficiency, which means they don't like to do things twice if they don't need to. So what can you do to help them avoid repeat work? Invest in tooling, proper methodology, discipline, and a knowledge repository.
Data scientists are driven by focus. They value uninterrupted time and the ability to dig deep into subjects. If your data science team serves as the de facto interface with data, they will have the very frustrating feeling that they're not making progress.
Overall, my belief is that investment in data team productivity— tools that make it easy to collaborate (e.g., lifting much of the data preparation burden off of data scientists) — is the best investment to retain a data science team. From the perspective of your data scientists, data science is happening now, and they don't have time to lose in your company. If you are not well equipped to make them learn new things, they will leave.
These tools bring both a business and HR impact — companies that want to get ahead and retain the best talent in the competitive data science space will take the time in what remains of 2018 to calculate the ROI of data team productivity.
That being said, if you really want to retain your best data scientists, maybe you don't have to choose between free food and good tools. Maybe you can just do both.