So You Have Decided You Need a Data Scientist: A Primer for Digital Marketers

As the head of your marketing group you know you want to bring on a data scientist in order to gain deeper insights from your data, achieve better customer profiles, improve your understanding of consumer touch points, and enhance overall campaign effectiveness. You’ve made the decision to hire. But what now?

Ever since the Harvard Business Review (HBR) called the data scientist “The Sexiest Job of the 21st Century” there has been a flood of companies wanting to hire them—and just as many people trying to fill the roles. The data scientist is a pivotal role that crosses all boundaries in a company. As HBR sums up below, they are both technologist and data product manager to all members of the team:

Data scientists realize that they face technical limitations, but they don’t allow that to bog down their search for novel solutions. As they make discoveries, they communicate what they’ve learned and suggest its implications for new business directions. Often they are creative in displaying information visually and making the patterns they find clear and compelling. They advise executives and product managers on the implications of the data for products, processes, and decisions.

Naturally, this raises two questions: How do you define your marketing team’s success now that you have a data scientist on board? And, what do you need for the data scientist to do so that your team can be successful? These are both simple yet amazingly deep questions, and both should be thoroughly thought out and discussed before you even start skimming resumes.

The 3 core questions to ask when hiring a data scientist

To answer these core questions, you actually need to start by answering some more basic questions first. Three, to be exact:

  1. Do your systems reliably capture all your data?
  2. Do you understand the data you are currently acquiring?
  3. Do you already have a set of questions or problems that you think your data can answer?

If your answer to question one is “no,” or “I don’t know,” your first data science hire should probably be a Data Wrangler. Similar to a cowboy, a Data Wrangler will manage your data acquisition and processes to ensure data is processed and internally consistent. A Wrangler will come in and review your data requirements, architecture, and processing and point out any shortcomings or potential problems. They should then be able to come up with both short and long-term solutions, allowing for a robust and consistent data acquisition.

If your answer to question one is “yes,” but two is “no,” you need a Data Miner or Algorithm Expert. Data Miners are “why” scientists. They will try to get to the root of the problem or question by asking odd but obvious questions like “Why are you asking the question you’re asking?”. They then try to find patterns in the data and build individual or derived metrics that will help focus the business in their direction and outcomes.

Your other option is the Algorithm Expert. These are the people who try to answer your questions and problems through Machine Learning, Artificial Intelligence, or Optimization techniques. They excel at answering the most difficult problems you can come up with by forcing the computer to do the heavy lifting.

Now if you answered “yes” to questions one and two but “no” to question three, you need an experienced Data Miner or Algorithm Expert. A senior/Lead data scientist can come in on day one and evaluate your situation. They will then partner with you to build out objectives and a road map for data science in the short and long-term.

Finally, if you answered “no” to question one and “yes” to question two, you are lying to yourself.  If you don’t understand and are not confident about the data coming into your system, how can you ever be sure of answers based off that data?

Final thoughts

In the end, hiring a qualified data scientist will help your marketing team no matter what type of data scientist you choose. They will help your team become more data-centric, with more accurate reports and analysis on topics like user acquisition costs, retention, and lifetime value. They will also deepen the understanding of your consumers with analysis including clustering, profiling, and messaging optimization.

Finally, your data scientist will begin to ask the questions that nobody else in marketing might have come up with like:

  • Can we predict user loss through a time series analysis?
  • What is the expected lifetime value of a customer by acquisition channel and personal information?
  • How many messages is too many messages on an individual basis?

Just remember to give your data scientist the leeway needed to be creative and inquisitive. This is the best way to get the most out of them.