Has the current fervor to pounce on every piece of available data for potential analytical uses spawned a world in which information often is collected for its own sake? Sometimes it might seem that way. But in the ever-expanding universe of "big data," predictive analytics software is one technology that can take advantage of the great variety of data accumulated by an organization as it works to model customer behavior and future business scenarios.
And using predictive analytics tools to interpret data is becoming more important to businesses: The most successful companies and rising-star enterprises sedulously employ them to help point the way forward on business strategies and operations, according to analysts who focus on advanced analytics technologies. But that doesn't happen magically, they cautioned; organizations need to take the right steps to develop effective predictive analytics programs.
In many industries, getting a leg up on the competition can be more challenging than ever -- especially if companies are set in their ways. The starting point in embracing predictive analytics should be ensuring that an organization has a proper frame of mind about using the technology, the analysts said. An open, dexterous attitude that's naturally curious, eager to learn and willing to adapt will produce the best results.
Douglas Laney, an analyst at Gartner Inc. in Stamford, Conn., thinks a predictive analytics program should begin by questioning historical business methods while searching far and wide for better ones. Companies "should not only focus on how things have been done in the past but be open to big ideas for innovations and transformations," he said. "This could mean applying measures effective in other industries to your industry." Such a mind-set should extend to the point of embracing approaches that "radically change the way business processes are done" in an organization, Laney added.
With that in mind, the mentality of the players -- particularly the business managers who are being asked to buy into the findings of predictive models -- is frequently the key variable that determines the success or failure of predictive analytics programs. A perspicacious corporate culture champions objectivity, welcomes new ideas and is naturally flexible. Conversely, a retrograde one resists change and draws heavily on existing biases and subjective formulas. "Resisting new ways of doing things is the reason most projects fail," said John Lucker, head of Deloitte Consulting LLP's advanced analytics and modeling practice.
Keep your eyes on the business prize
The grand plan of a predictive analytics deployment should also begin with a clear set of business objectives, said Thomas "Tony" Rathburn, a senior consultant at The Modeling Agency LLC, a Pittsburgh-based consulting company that focuses on data mining and predictive analytics. Then, he added, a team-oriented strategy is needed to advance those objectives. That is best constructed through substantive discussions involving program managers, predictive modelers, data analysts and business representatives.
More on predictive analytics
Learn more about the relationship between predictive analytics and big data
Watch a video Q&A highlighting how one company is using predictive analytics tools as part of its analytics strategy
Read about data scientists and their role in predictive analytics projects
So critical is the strategy development process that Eric King, president and founder of The Modeling Agency, recommends retaining "a seasoned strategic mentor" to help lead the effort and keep it on track.
Once a predictive analytics strategy is in place, it's time to begin the analysis process. Laney said "chewy" questions that probe deeply into data will unearth findings with high operational value.
The truly useful ones, he said, are multifaceted -- for example, "How can we grow new customers by 20% per year for a certain product line without cannibalizing other product lines given the range of economic forecasts, competitor trends and changing consumer demands?"
Run through predictive models, such questions can contribute in a big way to driving new business, according to Laney.
Building models is a testing process
After choosing and deploying the predictive analytics tools that best fit the job at hand, developing models is the next step. Mike Gualtieri, an analyst at Forrester Research Inc. in Cambridge, Mass., said analytics algorithms should be run on 70% of a data set to create an effective predictive model. "Then you test that model on the remaining 30%," he said.
Completed models should be regularly tested and enhanced as needed, and a set of performance metrics should be put in place for tracking their accuracy, Gualtieri added -- all part of a process for "continuous monitoring of the predictive analytics model."
Moreover, said other analysts, the entire predictive analytics process requires regular monitoring as business needs and the nature of the data being collected by an organization change. Analytics strategies and tactics that worked initially will need to be revisited and perhaps revised in order to continue achieving optimal results.
The mark of a truly successful predictive analytics program, Lucker said, is when some of the cost savings or business gains realized from an ongoing analysis project can be applied to pay for the next one so no new dollars need to be spent. "Using the value of each project to fund downstream efforts is an evolutionary approach that comes with a [built-in] return on investment," he said.
Roger du Mars is a freelance writer based in Redmond, Wash. He has written for Time, USA Today and The Boston Globe, and he was the Seoul, South Korea, bureau chief of Asiaweekand the South China Morning Post.
Follow SearchBusinessAnalytics.com on Twitter: @BizAnalytics_TT
This was first published in October 2012