BPM can bring big value to big data initiatives

Everyone’s talking about big data today, and business and IT pros who are involved with BPM are certainly no exception.

What makes big data challenging, of course, is that it’s beyond the immediate analytic capabilities of most garden-variety IT resources. However, a variety of platforms can extend the useful manipulation and analysis, says Jim Harder, principal of Visual Data Group, a consultancy based in Cleveland, Ohio. That way, companies can focus on results.

Armed with big data, Harder says, “manufacturers can make better-informed decisions about what to make, hotels can better balance demand with room availability and farmers can better predict where to put their resources.”

THE BPM-BIG DATA LEARNING CURVE
Forrester Research Senior Analyst Michele Goetz says many organizations don’t understand what’s involved in big data. “They’re just getting used to the technologies needed to assess and then analyze information from these large data sources,” Goetz said in a recent ebizQ podcast. “Then they just set them out into a silo and start to play around to see what benefit they can derive.”

That may be a typical kind of learning curve for a new wrinkle in technology. But some experts and practitioners believe that BPM can be useful in helping organization to derive value from big data from the get-go.

For instance, David Handmaker, founder and CEO of the online printing company Next Day Flyers, has found that big data can be made more useful with help from BPM. Big data and BPM processes work hand in hand to help the company grow and prosper, he says. With printing facilities in Saddle Brook, N.J. and Los Angeles and more than 100,000 customer records to manage, the company relies on big data both to drive strategic decisions and help the company to run more efficiently.

“We are able to do this successfully because of our BPM approach, which helps us continually align our internal processes with the needs of our customers,” Handmaker says.

USING BIG DATA TO BOOST MARKETING EFFORTS
Much of Next Day Flyers’ work comes from the business-to-business (B2B) arena. “We have tools in place, such as MicroStrategy and SAS, which allow us to assess different types of data to improve our marketing efforts,” he says. Specific examples include:

1. Segmenting data based on “firmographic,” or market-segmentation, information, then combining that data with online behavior information to automatically determine the best message to show a specific segment at a specific touchpoint – for example, a website banner, a re-targeting ad or cross-selling and up-selling options.

2. Combining back-end behavioral data with Google clickstream data to determine ideal ad-spend levels and understanding return on investment (ROI) at an ad-group level.

3. Joining Google keyword click volume and back-end behavioral data to serve as a tool to help build forecasts for new product offerings

“We wouldn’t be where we are today without the combined results of utilizing big data,” Handmaker says. “Both are essential components in our marketing toolbox and well worth the investment of time and money.”

BIG DATA AND BPM: WHERE TO BEGIN
Before getting into big data, it’s important to consider where the data is located in the organization and whether you have a unified view of it, says Frank Palermo senior vice president of the Global Technical Solutions Group at the Virtusa consulting firm.

“Big data can help organization get insight into behaviors and analytics can help them better service clients,” Palermo says. “But it isn’t just about processing more big data; it’s about doing something actionable with the data.”

To that point, Palermo says BPM can deal with the integration side: taking the big data, no matter its source, and serving in a traffic-cop role, steering the data toward where it’s needed. BPM can help to both stream and aggregate big data as well as bringing consistency to it.

“A lot of the processing of big data is currently done without context; there is no understanding of whether you are parsing data to facilitate a selling opportunity, to understand customer service, or to pursue a fraud investigation,” Palermo says. “Understanding what business process you are supporting helps in determining which needle in the haystack you are looking for.

LINKING BIG DATA TO 'BIG PROCESS'
Armed with the mantra that big data needs big process, Forrester Research’s Clay Richardson agrees that he challenge is determining how to connect big data and achieve insights across large data sets.

“Last year, I started to look at big process and we saw that big process could sit on top of big data sources so that you could get an automated, intelligent routing process,” says Richardson, a Forrester senior analyst. In existing systems, the potential value of big data is often lost, he says.

For instance, he notes, in the oil and gas industry, refineries have often been retrofitted with sensors that provide vast quantities of operational data--but not in record time. “They have admitted to us that it often takes six to eight weeks to get a meaningful data set back to a decision maker,” Richardson says.

However, he says, with BPM, organizations can more quickly visualize data and develop trigger points or invoke events. By using BPM, Richardson says, organizations can craft solutions that allow stakeholders to build visualization, with minimal help from IT, that supports “slice and dice” activities and, in turn, makes it possible to develop event-driven apps and alerts tied to specific processes.

Still, Richardson cautions, BPM may not be a silver bullet. “The biggest issues with big data are more around data quality. At a minimum, customers need to make sure they have a master data source that can be used by the business process sources,” he says.

And while big data seems to be “the shiny new thing,” that doesn’t mean companies should pack everything into a big data initiative, Richardson says. Instead, he advises companies to focus on how to become better connected to data and gain insight.

Among the benefits that BPM can provide is better understanding of processes and being able to match data to process, Richardson says: “With BPM, when big data is introduced, it lets you make changes as problems arise.”

READER FEEDBACK: Have you used BPM to improve your big data initiatives? If so, ebizQ editors would like to hear about your experience. Contact Site Editor Anne Stuart at editor@ebizq.net.



About the Author

Alan Earls, a journalist who specializes in writing about technology and business, is based in the Boston area.

More by Alan Earls, ebizQ Contributor