You Found a Data Scientist -- Now What? The Case for Building a Center of Excellence
Hiring a data scientist doesn't mean your work is done. Building a center of excellence and ensuring good relations with IT are both valuable steps toward a successful analytics program.
- By Shawn Rogers
- September 12, 2016
Contrary to popular belief, data scientists do exist. Yes, they're hard to find, with global demand far outpacing supply. Yes, they're expensive, and as we sit here today, most companies either can't find or can't afford a data scientist.
Although everything you've read and heard is accurate, that doesn't mean data scientists are unicorns. They're not mythical. They do exist and there are companies making the investment necessary to bring them into the fold -- and those numbers will only grow as more universities develop programs and curricula to support the explosion of interest in analytics.
However, data scientists are not silver bullets. Hiring a data scientist doesn't mean your work is done. It doesn't automatically transform your company and its culture to one built around analytics. It certainly doesn't guarantee a treasure trove of business-altering insights. Despite the significant hype around these high-impact hires, the addition of a data scientist is really just one (albeit valuable) piece in the overall analytics puzzle.
Analytics Takes a Village
Succeeding with analytics takes a village. It requires a wide range of skills and resources that go well beyond the hiring of a data scientist. Realizing this, more companies are moving toward the adoption of an analytics center of excellence. This centralized resource places the burden of analytics on a well-balanced team of employees -- including, if you have one, a data scientist -- with the right combination of technical skills, business perspective, and data access.
Think for a second about what it is you're hoping to get from a data scientist -- someone with the skills (business, statistical, and analytical) to make sense of data and turn it into actionable insights. As good as your data scientist might be, he or she can't reach that goal if a large swath of your data is still in inaccessible silos.
Your data scientist can't succeed if the disparate nature of your organization leads to the creation of multiple "analyst" reports pushing multiple versions of the truth -- or if poor analytics practices and culture continue to proliferate throughout the organization.
Building a Center of Excellence
Centers of excellence can address all of these challenges and more. Centers of excellence allow organizations to centralize data, improve data access, eliminate redundant efforts, and drive adoption of analytics best practices throughout the business.
Especially for larger companies with more complex reporting and analytics needs, centers of excellence can drive insights for the entire organization. They can help identify the data-related investments that have already been made while pointing out the gaps that still exist -- both things the newly hired data scientist might not be aware of.
Perhaps most important, a center of excellence can help democratize relevant capabilities that, although popular in one department, might be completely foreign to another. In the process, you'll ensure that you're not building processes and capabilities that aren't used. We've all been in a meeting where someone in sales points out how great it would be if they only had a certain capability, only to hear someone in marketing say they've had that capability for years.
Centers of excellence can break down these silos and ensure that your analytics initiatives focus on the smartest and most impactful things you can do for your business.
When Centers of Excellence Aren't Feasible
Just as not every company has the resources to hire a data scientist, not every company has the resources (human or financial) necessary to build a formal center of excellence. There are still steps those organizations can take to better position a data scientist for success.
Put yourself in the shoes of an incoming data scientist for a moment. Though you're long on statistical and data prowess, you're probably short on insight into historical business challenges, domain expertise, and pressing areas of immediate need. As an organization, it's important not to just drop your newly hired data scientist into the deep end of the pool with no direction.
Maybe you can't afford to start up a center of excellence, but you can set your data scientist on a course for success by identifying the specific business questions you're hoping to answer as a result of his or her work. Remember, where to start and how to start are critical questions for any analytics professional, data scientist or otherwise. A data scientist asked to boil the ocean is going to fail just the same as any other analytics professional would.
Work with your data scientist to assess your current landscape. Identify what you already do with data and use that as your launching point for a new project. Start small, start with something that's critical for your business, and let your data scientist grow into the role from there.
IT as Enabler
Another critical step any company can take to empower its newly hired data scientist is to ensure a collaborative working environment with IT. The success or failure of any data scientist depends on a company's ability to provide needed data. For IT, this means becoming an enabler rather than a gatekeeper.
Although the importance of security, control, and governance cannot be lost, enablement is about matching the speed of business -- the same speed a data scientist is expected to work at. If IT cannot (or will not) hold up its end of the bargain, neither can your data scientist.
If They Come, You Still Have to Build it
If there's one closing takeaway here, it's that if they come, you still have to build. Data scientists don't solve your analytics challenges, change your company culture, and uncover business-altering insights simply by walking through the door. The onus is still on the organization to put the structure in place that will allow these bright minds to flourish. Fortunately, doing so doesn't have to be nearly as challenging as finding a data scientist was in the first place.
Shawn Rogers is chief research officer at Dell Statistica. Shawn is an internationally recognized thought leader, speaker, author, and instructor on the topics of IoT, big data, analytics, cloud, data integration, data warehousing, and social analytics. Shawn has more than 20 years of hands-on IT experience. Prior to joining Dell he was vice president of research for business intelligence and analytics at Enterprise Management Associates, an analyst firm. He co-founded the BeyeNETWORK, a global publication covering business intelligence, data warehousing, and analytics and was also a partner at DMReview magazine.