An unprecedented volume of data inundates businesses every day—and what organizations do with it matters. When fully exploited, it enables companies to do better. They make better decisions, engage customers, and leverage emerging technologies like artificial intelligence, machine learning, and blockchain—which is why data scientists are among the most strategic positions businesses are recruiting for today. Top-notch data scientists propel companies forward by translating data into strategic insights. The best and brightest are creative, curious problem-solvers who have a rare blend of analytical and creative abilities. They understand the business problems they need to solve and how data can help. And they can explain results in terms business leaders understand.
If you’re looking to find these people to fill data scientist positions, be prepared. It’s a highly competitive hiring market: Demand for data scientists is high and expected to grow 28% by 2020, according to a recent IBM study. The pool of qualified candidates is also small, partially because it’s a relatively new field. Companies are looking for experience, with 78% of data science positions requiring at least three years’ experience and 39% of positions preferring a master’s degree or Ph.D.[i] And they’re competing for top talent against large, innovative businesses with well-established data science departments, like Amazon, Google, and Facebook. Attracting and hiring the best data scientists can challenge even the most talented search firms.
How Do You Find Them?
It’s not easy. To ensure you find the best-fit candidates for each data science position in your organization, be strategic and specific about what you need, and be prepared to change your approach when you aren’t getting the results you seek. Here are a few insights we’ve gained through helping organizations hire data scientists on their product and engineering teams:
1. Let business strategy form the foundation of your search.
There are many types of data scientist roles with a wide variety of responsibilities. For example, some work closely with product or engineering teams, while others align more closely with business operations. Positions may demand big thinkers versus applied theorists; researchers versus agile problem-solvers; people who create outputs for people versus those that create for machines.
Key and critical is the work that starts BEFORE the hiring process. Form a data-science-savvy hiring committee. Work with them to let the organization’s goals drive a clear understanding of the people you need on board by identifying the specific problems to be solved and how each role will help the company meet its objectives. This sounds obvious, but there’s not always clarity—and that complicates hiring.
Keep in mind that a good mix of data scientist types can often make a team more effective. According to Jayant Lakshmikanthan, CEO and Founder of Clara Analytics, if talent is too heavily research-focused, teams can get into analysis paralysis. But too many tactical problem-solvers may not be able to go deep enough into the problem to find the best solutions. He adds, “It helps to have a critical mass of data scientists already on board to attract the best. When there are about 10 data scientists on a team, candidates know the company values the role and that newcomers will learn a lot from their peers.”
2. Be specific and flexible about the skills and experience you need.
Drill down into the work you need people to do day-to-day to get specific about the skills you’re looking for, for example:
- Technical skills: Experience with specific tools and technologies (e.g., computer languages or platforms for data analytics, machine learning, and AI platforms) or techniques (e.g., methods specific to machine learning, AI, NLP, statistics, visualization or modeling)
- Soft skills: Abilities to communicate, collaborate, problem-solve, or lead, among many others
- Industry experience: Time working for other companies within your industry
Keep in mind it’s often easier to help people develop technical skills versus soft skills. It’s also generally easier to assess candidates for their presence. So be prepared to prioritize the need for each if it’s difficult to find qualified candidates.
Additionally, know that many highly qualified data scientists come from academia, but they may have little commercial experience. And recent graduates might appear to be a great fit for a position but not have the real-world experience you’re hoping for. In both situations, plan to immerse them in the business early to develop the knowledge they need.
3. Cast a wide net and be prepared to adjust.
Be open-minded about where you look for this coveted talent. Leverage internal networks. Stay up-to-date with new and evolving data science programs at academic institutions. Keep in mind, however, not to be overly-concerned about pedigrees—a person coming out of a state college may be a better fit for a position than a recent graduate of a STEM program at an ivy league school. “It’s more about building expertise to do the specific job at hand—decoupling the domain with the actual work itself,” says Lakshmikanthan.
Be prepared to adjust the search when needed. For example, a company in the field of natural language processing (NLP) engaged us to help find data science candidates with a nuanced set of skills and experience. We started by seeking a “double bullseye”—candidates with NLP expertise and the soft skills most important to the company. Eventually, we cast a much wider net with the client’s support and found the right person for the job—an individual from academia with passion and strong soft skills but minimal business experience. In the end, the hiring team wasn’t as concerned with the lack of a commercial background, because they believed in the candidate and were willing to help him develop over time. He has been a valuable addition to the company.
When needed, it can also be helpful to engage a search firm with prior experience filling data science roles. They will be familiar with the hiring market, can tap into their own extensive professional networks, and can enhance your ability to attract top talent in this competitive space.
Data scientists sit at the intersection of business and technology, which is where our increasingly-networked economy flourishes. Finding the best is difficult but worth the investment. They are integral to linking strategy with execution and can be catalysts in the organization by providing insights that help leaders find the best path forward. Leaders who are serious about gaining the most benefit from the ubiquitous data they collect every day should look to find the best and brightest.
[i] The Quant Crunch: How the Demand for Data Science Skills is Disrupting the Job Market; Burning Glass Technologies, Business-Higher Education Forum, and IBM; https://www.ibm.com/downloads/cas/3RL3VXGA