A Facebook and Microsoft experienced Data Scientist, Mengying Li, says, “Often, the push to build out a data science function is driven by FOMO – fear of missing out. But without thoroughly understanding the data science needs of your particular business, you risk hiring someone not well-equipped to tackle your particular company’s unique goals and challenges – a mismatch that sets you back, rather than springing you forward.” Here are some strategies for when and how to hire your first data scientist.
When to Hire
- Put your mission first. If you’ve been able to understand your customer and your product roadmap without a data scientist, you don’t need to hire one to cosign your decisions. But if you need help defining your customer, where they come from and what they want, you would probably benefit from a data scientist hire.
- Do you have a data engineer? Without ENOUGH data that is organized, cleaned, optimized, correct, understandable, curated, governed, secure, measurable and archived, your first data scientist won’t be able to do their work. Good data engineers will not only prepare your data, they will probably be the first ones to know when a data scientist is needed.
- Do you have the right tools? In addition to your data engineer’s data collection and storage tool and data pipeline tool, your first data scientist hire will need a business intelligence tool and an interactive query interface.
- Do you have company-wide support? Your entire organization must embrace a data-driven culture . Mengying Li says, “If expectations are misaligned and your team doesn’t embrace data culture, data scientists might just end up like fancy window dressing, rather than driving business impact.”
- Would an external consultant be the best first step? If you have doubts or concerns about any of the above points, try using an external consultant or contractor first. With data scientist median salaries around $160K per year, being unsure of what you need and what your company will utilize effectively could be a costly mistake.
How to Hire
- Craft a job description that first defines what you want the data scientist to do. What are the goals of the position? What do you want them to accomplish? Be specific. Then detail the analytical skills, programming proficiency, machine learning expertise and communication and presentation skills they will need.
- Have a champion within the organization that understands what a data scientist can do for the organization and helps to sell it up the hierarchy as well as across the company. This champion should be an integral part of the interview process.
- Hire a recruiter that specializes in placing data scientists. These candidates are rare and expensive and if it is your first data scientist hire, you will need someone who knows the market and the variations in the candidates. They can help you with not only the technical assessment but the cultural fit as well. The best data scientists have the ability to explain and sell what they do to non-technicals within the organization. A recruiter who has worked with them over time will know the abilities of the candidates without a reference check – although you should have these done anyway!
- Candidates should be given a problem to be solved and should have an online portfolio that can be reviewed. The problem should not take more than an evening to analyze but their skill in solving the problem and in presenting the results are both critical parts of your assessment of them as a viable candidate.
One last recommendation – integrate your first data scientist hire carefully. Initially they will spend most of their time interacting with the data engineer to get the right data in the right form and battling with data tools to be able to efficiently do the right analysis. Your champion should smooth the path to good relationships with the staff who will be interested in and utilizing the results your data scientist will provide. Success with your first data scientist hire has a very significant payoff when done right.
