How to Make Big Data Work For You

This is a guest column by Jacqueline Paige of Smith Hanley Associates, LLC, as part of our Analytics Advantage series.

Big data has been touted as the new route to easy success. Companies have been made to feel like they’re way behind the times if they haven’t gone after big data aggressively. The problem? Too many companies jump into big-data projects and end up with nothing to show for their efforts.

Why?

1. Lack of Focus

Too many companies are seeking data for data’s sake. Too many companies have started big-data initiatives without first asking themselves: What is our goal when harvesting this information?

Before starting a big-data project, ask yourself: What do you want to learn? What questions are you trying to answer? Are you flexible enough to tweak goals if the data takes you to unexpected places?

2. Lack of Talent

Once you get that data, you need to know what to do with it—you need people to analyze it. And in the current marketplace, there is a huge demand for skilled data analysts that shows no sign of lessening. The McKinsey Global Institute has estimated U.S. demand for employees skilled in data analysis could outstrip supply by 50% to 60% by 2018.

How can you get around this skills gap? Start preparing your internal talent pool as soon as possible. Try giving existing analytics specialists new training in big-data skills such as the programming language Python. Or check out local educational institutions for recent grads from certification programs with names like data science, decision sciences and machine learning.

3. Lack of Organization

It sounds basic, but before starting a big-data project, your company should do a “data audit” to ensure the information you want to mine is collected in a single database and format. Data can’t be helpful if it can’t be found.

4. Lack of Ownership or Lack of Buy-In

Big data initiatives must have top-level support and must be funded appropriately. Somebody at the C-level needs to take charge; many companies are hiring “chief analytics officers” or putting someone in a similar role to bring a top-down approach. This can also help avoid organizational friction, such as territorial spats between departments over who owns a project.

5. Lack of Realistic Goals

Don’t start off with expensive or high-risk big-data initiatives if you can help it. Sure, complex problems require complex approaches, like building software from scratch or adding a data center—but often you can get more bang for your buck by trying smaller projects with narrower goals. Achieving tangible results builds confidence within the organization and gets skeptics on board.

 

##  ##  ##

 

Jacque has been recruiting statisticians and quantitative analysts for positions in marketing since 1989.  At Smith Hanley, she is seeing continued demand for data analysts and a shortage of top talent.  Contact her if you’d like to learn more about big data and what she can do to help your company tackle it.

Your Comment