How to pay your data scientists to increase retention

Posted by

Promising data-minded individuals are easy to spot from the distance: curious detailed-oriented thinkers, energized by uncertainty, open to opposed opinions and willing work in team iteratively to produce insightful recommendations, but still with feet on the ground.

Download pdf

The analytics job market is nascent and employers compete fiercely for the best people. Sadly, salary benchmarks are unavailable or do not reflect the market since the common practice is not to pay a benchmark but to just give a rise. Simply paying a premium to staff is typically not sustainable.

There are many aspects involving retention like sense of fit, learning and work atmosphere. But, this article focuses on innovative remuneration beyond just increasing the basic salary.

One of the challenges HR departments have with data professionals is that they require a completely different set of incentives from other more traditional employees groups. Creating multiple remuneration sets requires a sophisticated HR methodology and leadership as well as a holistic communication plan so that the work-force understands why multiple plans are required.

Increasing basic salary improves acquisition but not retention

The traditional way of structuring pay focuses on the level of the pay from a relatively static point of view. There is a minimum pay in the band and a maximum pay. What most HR managers do is to move the minimum and maximum a higher based on benchmarks provided by recruiting companies.

An approach focused on pay level largely addresses acquisition. Companies increase the offered salary to be more competitive in acquiring talent.  Retention, which is a bigger concern for most companies and more difficult to solve, is about the growth of that pay level, rather than the level statically.

An approach focused on pay level largely addresses acquisition. Retention is about the growth of that pay level, rather than the level statistically.

Companies should structure a remuneration program which explicitly handles growth in a very different way than the traditional usual approach. The starting level matters less than the growth.

Ways of including this growth perspective include skillset allowances upon achieving the technical certifications, or progressive bonuses upon reaching key milestones, such as succeeding the probation period of 1 or 2 years.  This is equivalent to the anniversary bonuses C-Levels sometimes have

Understanding millennials

Chief Marketing Officers are becoming vocal with the teams about the need to understand millennials. This is critical for remuneration plans as well. The insight to understand is that millennials do not plan for the long term future. As a result, long term incentives do not have the same drawing power. Anything beyond 2-3 years is often irrelevant for them.

Thus, companies should aim to retain people for 2-3 years, rather than not beyond that. This is an additional reason to accelerate those long term incentives we have into the first years so that the total remuneration grows. Some of the measure companies can take to this end include:

  • Starting long-term incentives earlier instead than after a few years
  • Providing better paid shorter term contracts rather than more humble longer term or indefinitely
  • Or, completely eliminating long term incentives in order to fund skillset allowances, anniversary bonuses, or progressive bonuses.

Understanding salary benchmarks

Companies usually buy salary benchmark obtained from recruiting companies. Unfortunately, benchmarks for analytics do not exist in all markets. The reality is that in analytics the benchmark is not reliable in any emerging market.

The analytics job market is still nascent in many countries and employers compete fiercely for the best people. Unfortunately, salary benchmarks are unavailable, or do not reflect the market since the common practice is not to pay a benchmark figure, but to just give a raise.

In the field of economics, there is difference between being a price taker and being a price maker. The same happens in the job market. When the market is nascent, all companies are price makers. When the market is mature, most companies, except maybe the largest ones, are price takers.  The analytics job market is still in a price-making stage rather than a price-taking one in most emerging markets

Not only data scientists are scarce

It is a very common view that in analytics, only data scientists matter, and are a scarce resource. Analytics professionals in fact encompass a very wide group, including data scientists, data engineers, data architects and internal consultants. There should be no doubt that data engineers, data architects and analytics consultants are as scarce, and as important, as data scientists. Therefore  incentives should involve a broader group than only data scientists.

Additionally, providing a special remuneration approach to some data professionals and not to others exacerbates the silos that are commonly in place, for example between data engineers and data scientists. Setting different incentive mechanisms for different people has the risk of hindering that, and producing additional employee churn.

Connecting capability development with remuneration

Capability development is required to retain curious analytical talent. Creating a training program for data science and engineering is the first step. The program needs to offer a certification for every course and each company needs to define the certifications that will be required to be recognized internally.

Online curriculums such as Coursera, Udemy or Udacity are becoming popular because they are an inexpensive way of providing exceptional quality training and certifications for beginner or intermediate students

It is necessary to connect remuneration to the certification programs. Certifications increase the market value of a data professional. If his or her value of within the company does not increase, the probability of losing him or her spikes. As a result, it is very important to create skillset allowances and to communicate them broadly to the employee base.

More widely, every employee who obtains an certification should be able to get the skillset allowances even admitting non-analytics employees to be in the target group, not only data scientists.  For example, it should be open to consultants or marketers on the same terms as for data scientists.

Disclaimer: Opinions in the article do not represent the ones endorsed by the author’s employer.

One comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s