Using Analytics to Squeeze Value From Big DataBy Tony Kontzer | Posted 2015-01-28 Email Print
A new generation of analytics tools is helping a growing number of companies derive bottom-line benefits from the ever-larger pools of data they're collecting.
That's the problem Grace Preyapongpisan, vice president of business intelligence for foodie site AllRecipes.com, faced up until about a year ago. With as many as 40 million unique visitors a month, the site captures billions of records, including consumer interactions and advertisement impression data. Combining those data pools to glean insight into how AllRecipes.com could be tweaking its site and serving up ads to create a better experience was proving to be a difficult task.
One particular challenge was meeting requests from the sales organization for metrics it could use during meetings with potential advertisers. If a sales team was going to meet with Quaker Oats, for instance, it might want data on breakfast trends. But Preyapongpisan's team of analysts lacked the technical skills needed to make this happen quickly. So, if the team wasn't given enough lead time, the data sets provided were weaker.
Enter analytics technology. Preyapongpisan says AllRecipes.com chose Tableau Software about a year ago in large part because of its ease of use for nontechnical users. She estimates the analysts are now able to assemble so-called "hot sheets" of data for sales calls in half the time they used to need. What's more, she's hearing from the sales staff that this capability has been a key differentiator in closing deals.
That's not all: Prior to Tableau, Preyapongpisan's team spent a full week at the start of each month preparing reports on the previous month's data. Today, that process is completed in a day, which she describes as a "huge gain in efficiency."
Not surprisingly, projects like this one have fueled demand for Tableau elsewhere in the company. "The rest of the organization is extremely hungry to get access to that," Preyapongpisan says.
451 Research's Hanselman says that the growing sophistication of big data analytics tools (and more mature offerings are on the way, he says) is helping companies develop a deeper understanding of what analytics can—and can't—do. But success on the big data front is about much more than technology. It's about understanding what you want from the data before you start analyzing it.
Simply accumulating data and then figuring out what questions to ask "is a recipe for disaster," Hanselman says. "You've got to understand what the business drivers are up front."
Guidelines for Big Data Analytics Efforts
Too often, companies put the cart before the horse with their analytics efforts, applying technology before they fully understand their data needs. Kumar Ramamurthy, vice president of enterprise information management for IT consultancy Virtusa, has this common-sense advice for getting big data analytics efforts started on the right foot:
- Make sure the people who know your data best are part of the project because "no one understands your data better than you."
- Create use cases. "If you can monetize them, do so. If not, seek help."
- Keep technology out of the discussion "until you know what needs to be done and how it needs to be done."
- Make sure to create a viable project plan and "measure progress and ROI in each iteration."
- Consider doing just a thin slice of the project to gauge success and identify roadblocks.