Five analytics priorities for 2016

Looking to sharpen your analytics programs in the new year? Follow this cheat sheet on how to win with analytics in 2016.

The International Institute for Analytics' (IIA) annual list of analytics predictions included something new this year: In its recent webinar, members of the advisory and research organization in Portland, Ore., also offered a list of analytics priorities as a companion piece to help guide CIOs and analytics leaders in the new year.

"We talked to companies extensively in our network and sought out opinions from attendees of our Chief Analytics Officer Summit that we had in October," said Dan Magestro, a former analytics professional who worked at Cardinal Health, JPMorgan Chase & Co. and Nationwide Insurance, and who now heads up research efforts at the IIA. "And we combined that with our own perspective ... on where the industry is going and on how to best win with analytics."

Why muddy a predictions list with must-dos? According to Robert Morison, analytics consultant, researcher at the IIA and webinar moderator, analytics are maturing. "And with that maturity, our priorities are naturally quite focused on the business side of the house and putting analytics to work directly," he said. "[The list of priorities is] also making sure you're building an analytical organization and capability to help make that happen -- not just today, but to build a muscle to be even better at it tomorrow."

The IIA's five analytics priorities are as follows:

No. 1: Use analytics to develop business strategy

Businesses are still struggling to get the analytics buy-in they need, according to Magestro. In 2016, those leading analytics efforts should prioritize analytics projects that tie directly to strategic business initiatives, which "will help the analytics work and the analytics stewards increase their relevance."

Tom Davenport, co-founder of the International Institute for AnalyticsTom Davenport

But aligning to business priorities isn't enough; CIOs also need to align analytics to strategy development and the strategic planning process, which is another area analytics leaders have struggled with, Magestro and other IIA researchers agreed.

"You've got to think a big underlying cause of this is just senior executives who aren't analytics-oriented in the first place," said Tom Davenport, the president's distinguished professor of information technology and management at Babson College in Babson Park, Mass., and co-founder of the IIA. "If they were, they would insist on this sort of thing. They'd insist that a strategy be analytically verified."

No. 2: Tap existing analytics strengths across the enterprise

Different departments often acquire competency in analytics at different rates. Rather than have every department reinvent the wheel, the CIO's list of analytics priorities in 2016 should include connecting the talent dots across the organization.

At financial institutions, for example, cybersecurity teams may be well ahead of the rest of the organization on building data lakes and real-time analytics. "If you want to put that combination of capabilities to work, for example, on the customer side, you may already have great experience in place and a precedent for doing that sort of thing," Morison said.

Robert Morison, researcher, International Institute for AnalyticsRobert Morison

Leaning on in-house talent is a good utilization of resources that can encourage coordination and cross-functional analytics projects, unlocking new insights that could be important to the business, Magestro said. It can also send strong signals to an established team. "Even if they're sitting in a different function, those leaders ... are going to enjoy the increased relevance they can have and the increased perspective they can have," he said.

No. 3: Define and prioritize analytics projects with the business

Clear processes on how a team prioritizes, defines and executes on analytics projects can set project parameters, set expectations and build consensus. "Project management, project discipline, prioritization discipline -- that whole muscle -- is part of the journey of building analytics to be a true practice, a true capability within companies," Magestro said.

Dan Magestro, researcher, International Institute for AnalyticsDan Magestro

He suggested analytics leaders start by taking a look at their project intake process. "This is really around better capturing the business problem in order to connect the work to the decision," so analytics professionals have a clear focus as to what the ultimate goal is, Magestro said. And he encouraged leaders to come up with "a scheme to prioritize projects in a formal way" for a couple of reasons: to ensure efficient allocation of resources and to build consensus among senior leaders on what key projects to tackle.

But formalizing project management practices could encourage "shadow analytics" -- the analytics version of shadow IT -- if some groups feel they're getting the short shrift. Davenport called shadow analytics "uncoordinated" and "rouge;" Magestro, on the other hand, said shadow analytics could be a positive development for companies. "The more people doing it, the more people invested in it, the better for our organization," he said.

No. 4: Retain analytics talent with career development programs

Analytics and data science talent are hot commodities, which can make retention efforts difficult. "Just being able to backfill analytics talent as it leaves or moves can be challenging, because you don't have a talent pipeline established," Magestro said.

Along with building a pipeline, it's important that executive leadership recognize and support the unique nature of the role, experts said. Gone are the days of the back office statistician or quant, the foundation from which the role of the data scientist evolved; today, analytics and data science professionals are often expected to possess business acumen, as well as technical skills, a combination that can create complex career paths, Magestro said. "Analytics folks are more cross-industry mobile than other skill sets," he said. "So someone from retail can move into healthcare can move into financial services. That's natural and even sought after."

Magestro provided three recommendations on how companies can support analytics professionals:

  1. Recognize their leadership potential.
  2. Provide training in business skills. Business basics, such as how to own the room, how to communicate with the business or how to develop strategy, provide skills on how to move from the technical to the business side of the house -- and vice versa, he said.
  3. Provide additional support to analytics professionals who step into management roles. Magestro said they'll likely have more difficulty than someone with a business background. But they'll bring critical thinking skills and experience with data to the role, which "will benefit the organization far beyond the investment required to do that," he said.

No. 5: Measure the value of analytics

CIOs and analytics leaders should tease out the value of analytics, which is often lumped in with other processes. "Recognizing that analytics is its own muscle within an organization, I think viewing it that way, under that same lens will help increase the buy-in with leaders in the organization," Magestro said.

Doing so will enable CIOs and analytics leaders to make a case when they need to justify an investment in talent and tools, or when asking for additional resources. Magestro provided three ways to measure the value of analytics:

  1. Measure the direct benefit of the insights: Return on investment (ROI) could come from an increase in revenue or a decrease in spend from identifying inefficiencies.
  2. Measure the accumulated knowledge gained: Harder to quantify than ROI, but something that's still "very real," Magestro said. Even if a project turns out to be a bust, the results become part of a library of analytics knowledge that will influence additional work. "The way I like to say it is every hour that gets spent on analytics work makes the company smarter," he said.
  3. Measure the analytic efficiencies gained: Having an in-house analytics team cuts down on possibility of duplicating work across an organization; plus, it means companies don't have to outsource the work to an expensive firm. "There's a cost of not doing the work that needs to be called out," Magestro said.

Davenport provided one additional piece of advice: Be good at "anecdote management."

"Have a bunch of well-crafted anecdotes at your disposal so you can show the kinds of things you've done," he said. He compared it to having a marketing function for analytics projects.

Welcome to The Data Mill, a weekly column devoted to all things data. Heard something newsy (or gossipy)? Email me or find me on Twitter @TT_Nicole.

Next Steps

More on analytics priorities

Gartner's building blocks for big data strategy

Crafting an enterprise data analytics strategy

Hone your data analytics strategy

Dig Deeper on Enterprise business intelligence software and big data