No doubt you’ve heard about a shortage of data analytics specialists.https://o1.qnsr.com/log/p.gif?;n=203;c=204663295;s=11915;x=7936;f=201904081034270;u=j;z=TIMESTAMP;a=20410779;e=iThe data’s getting a bit long in the tooth, but a 2011 McKinsey Global Institute study predicted a shortfall of about 150,000 people with the needed analytic skills to manage Big Data analytics.
That may not be the biggest problem facing analytics, however. An equally important, but less cited, finding in that study was the predicted shortfall of 1.5 million business people who could leverage that data, notes a recent Harvard Business Review blog post.
“There is widespread recognition of the shortage of analytical professionals,” writes Robert Morison, co-author of “Analytics at Work: Smarter Decisions, Better Results.”
“Lesser appreciated is the fact that most organizations are also way short on analytical amateurs.”
He uses the term “analytical amateurs” to refer to business professionals and leaders who may not be hired specifically for analyzing data, but who nonetheless can appreciate how analytics can be used and what its limitations are.
In other words, he’s talking about the “power users” for data.
Here’s why this matters: Even if you don’t care two hoots about Big Data, it’s clear that organizations are shifting to a more data-driven world. For many, that will first mean developing the in-house ability to effectively use data, whether you’re investing in Big Data or sticking with the Little Data.
Data, no matter how large, no matter how small, will matter.
Once you’ve accepted that premise, here’s another reason it matters: There’s a bit of a debate about the best way to put the data heads to work. Do you:
A. Centralize data analytics, filtering your data (especially Big Data) through one Center of Excellence, which then analyzes the data and acts like a publishing clearinghouse for the results.
B. Put the data professionals to work within the business units.
There are pros and cons to each approach, according to a recent VentureBeat article, “How companies decide what to do with those darned data scientists.”
A panel of data experts discussed the question during VentureBeat’s 2013 DataBeat/Data Science Summit.
This panel featured data-heavy organizations like Intuit, LinkedIn, Humana and the multinational conglomerate GE. You know, the kind of companies that will get a first-draft pick at graduating data scientists.
So for this panel, the question over whether to centralize the data science group came down to the question of what would work best for the business goals.
For instance, Intuit, with its data-savvy staff, took a cross-disciplinary collaborative approach; GE set up an official hub of data scientists (not my words); and LinkedIn settled on a combination that creates a central approach to analyzing and publishing data, but requires data scientists to participate on product teams.
“If they’re not there as part of the discussion with product, the opportunities to leverage that (data don’t) really come up,” explained LinkedIn’s social network senior director of data science.
Any of those approaches might work well for large or well-funded organizations with a strong, business-focused need for analytics.
But for the rest, Morison’s approach actually makes the most sense, because it’s more flexible and addresses both the data scientist and the data power user gap.
Morison recommends a cross-training approach that couples the data scientists with business colleagues.
“The most effective employee development happens on-the-job, day-to-day, often one-on-one,” he writes.
The data scientists learn more about the business and its needs, while the business users learn more about data, including:
- The importance of data quality
- How to find the data they need
- How to use the BI and visualization tools
- An awareness of the limitations of data
Morison points to Procter and Gamble’s approach as one way to marry the two. Instead of centralizing or decentralizing the data scientists, the company established a rotation program that puts business people into analytical roles so they can learn first-hand from the data analytics professionals.