Every business appreciates the importance of data and analytics. Even companies that have the most obsolete “data capture, store, and analyze” processes know they have to catch up or bow down. It’s not uncommon for today’s SMBs to budget generously for processes, programs, and technologies that bring them the power of data. Where’s the problem, then? For most companies, their lofty “data” programs end up merely creating pipelines to organize and accumulate data. To churn data, make data from disparate sources talk, and to build single versions of truth across business, you need the magic of a whole new kind of science. That’s data science, and these magicians are data scientists.
Hire data scientists? Why now? Can’t it wait?
Here’s a little stat to put things in perspective. A recently published McKinsey report suggested that by the end of 2018, there would only be 200,000 available candidates for 490,000 open data scientist positions.
It’s clear — data scientists are hard to find, and consequently expensive to hire, and difficult to retain. On the other side, businesses must treat this as their lottery ticket, wherein they could simply get a leg up on their competition by hiring data scientists now and jumping a few months ahead. Next up, let’s tell you how you can hire data scientists.
Phase 1 — Do your homework
Imagine. You decide to hire the best salesman in town. On Day 1, your instructions are clear — make us money. Will this work? No, 100 out of 100 times it won’t. Just like your decision to hire a salesman must be based on a conclusion that your product’s market potential and territory potential are underutilized.
Similarly, don’t expect a data scientist to walk in with a halo (or a magic wand). The decision to hire a data scientist must be made after establishing a case for the expensive hiring exercise. Ideally, your problems should be close enough to these, for you to look for a data scientist to solve them:
- You need someone who you can rely on to build organization strategy toward leveraging analytics for business success.
- You want someone to lead projects where routine problems are rooted in mathematics, statistics, probability theory, and computer programming.
- You want someone who knows which business questions can be answered using data, as well as the ones that can’t and should not!
- You need better processes that help employees work in a manner such that the data trails of their activities are easy to log and analyze.
- You desperately want someone with a keen understanding of statistical modeling to build highly tailor-made algorithms.
You get the idea, right?
Phase 2 — Know where to look
Remember we mentioned how data scientists are hard to find? Well, that should give you an idea of how difficult this phase might be for you. It’s important to give yourself ample time rather than risk a hasty hire. Conventional channels of hiring aside, you will need to be as innovative and resourceful as you want your data scientists to be.
For many businesses that have mature data governance processes and analytics champions in place, it’s too hard to resist the temptation of promoting an employee to the role of data scientist. That has its merits and demerits, and that’s an entirely different discussion.
The CIO or the CDO are best placed (if they’ve been with your business for 2+ years at least) to determine whether the current state of data-related processes is robust enough to consider an internal hire.
If you’re looking toward the skills market, you need to know where to look. Some tips:
- Create a business account on the top five most popular on-site blogging platforms dedicated to data science, and build a community.
- Join newsletters from online data science magazines to be in the thick of things from the market.
- Join online academic communities and foster relationships with academia offering credible data science degrees.
- Check for credible data science boot camps that can bring out some really exceptional talent.
Note: Your preparedness for your organization’s data science journey will improve mightily via these platforms. However, remember to always contribute, and focus on community-building, rather than pitching your organization’s job offers.
Phase 3 — Know what to ask
The “data science” talk is fresh, at least in the context of applying data-driven insights for measurable business benefits, across the organization. So, it’s crucial that you entrust senior managers with the responsibility of conducting interviews and interactions at advanced stages of the hiring.
It’s generally prudent to keep the hiring program flexible, to give yourself a better chance to evaluate the multi-disciplinary expertise of candidates.
Another golden rule to remember, while hiring data scientists is: “talent first, tech later.” The difference a data scientist can bring to a business is centered on his/her talents such as analytical thinking, strong communication and people skills, and domain experience. The technical aspects, as important as they are, can be honed via in-house training and sponsored external certifications.
Phase 4 — Know what to offer
Ironically, expert data scientists are themselves careful and keen enough to find the perfect fit. Who wants to land in a cage with peering eyes from all sides, hoping for the next miracle!
When you get the good gut-feel about a few prospective candidates, you don’t want to miss out. If they’re as good as you think, these are the nonmonetary conveniences and support mechanisms that will help you close the deal.
- Support in getting cross-functional teams on the same platform, and securing their support for data science projects.
- Support from business teams to assist the data scientist in identifying pain points and potential points of value addition.
- Full flexibility in running pilot tests to dynamically evaluate projects in environments that closely match real life.
Data-driven insights could help you create better products, expand the business to more territories, improve process KPIs, sell more in less time, and, in general, do more with less. Data won’t do that on its own; you need data science on board and data scientists to steer this ship.
Featured image: Pixabay
