So You Finally Hired a Data Scientist

So, Marketing leaders, you have a new data scientist on your team?

First of all, congratulations!

Data scientists are rare and expensive, which means hiring one is a true feat for most marketing leaders. Not to mention, you’ve unlocked a host of new possibilities. Leveraging predictive? Making use of AI? This new frontier is exciting.

Yet, about once a week, I have a client who asks “How do we make the most of our data scientists?” It’s a legitimate question. Because of a failure to realize promised improvements, my colleague Lizzy Foo Kune predicts that by 2023, 60% of CMOs will slash the size of their marketing analytics departments by 50%.

It’s not that the talent isn’t needed or that the use cases are poorly scoped. The thing is, for many organizations, other barriers emerge after they get the talent they want and need. Some of these include:

• Data scientists spend more time fixing data than analyzing it. In a Gartner for Marketing Leaders survey in 2018, marketers who had dedicated data science resources reported they were more likely to be performing tasks such as data visualization (48%) and data preparation (46%) rather than advanced activities such as modeling and machine learning.
• Equipping data scientists requires additional investment. What started as a case to hire a data scientist can quickly emerge into a need for additional tool sets, third party data, etc. Lots of conversations about deeper customer insights and advanced capabilities go from “a data scientist can…” to “our data scientists could…” for lack of greater resources.
• The organization’s most pressing needs aren’t appealing to data scientists. Organizations often hire data scientists to complete high-profile projects. Sometimes these projects are wildly successful. Sometimes they’re not. In either scenario, after the project’s completion, what’s next isn’t clear. Progressive organizations assess what’s most needed for the organization, and in turning away from what’s glitzy, many find they need a less specialized skill set.

If you find yourself confronting these barriers, don’t despair. It’s not that leading organizations don’t have these problems; they adapt. One of the easiest ways to free up data scientists’ time is to assess which other team members have a basic understanding of data science and can take on “data cleansing” activities, using How to Improve Data Quality (subs. required). Additionally, before contemplating new investments, consider what additional capabilities are available through tools you already have. Use this white paper, the 4 Types of Analytical Tools to Support Insight Generation (subs. required).