Predictive analytics can help companies forecast business results with stunning accuracy – for example, how a particular group of customers might react to a targeted offer or what the potential business impact might be as a result of a marketing program.
But do data scientists always ask the right questions to get at the heart of what business leaders are attempting to accomplish? For instance, when is the right time to make an offer to a customer in order to generate the most favorable results?
Predictive analytics can clue data scientists and other users in on patterns worth examining (e.g. customers and prospects with certain common attributes were most likely to convert when an offer for a particular product was presented).
Companies that use predictive analytics more effectively than others achieve dramatically better business results. According to a recent Aberdeen Group report, “companies using predictive analytics enjoyed a 75% higher click through rate and a 73% higher sales lift” than companies that don’t. However, even among users of predictive analytics “there are significant differences in the level of business performance achieved,” according to the report.
In particular, top-performing companies identified by Aberdeen outperform their peers considerably in areas such as year-to-year changes in customer lifetime value, average response rates and average opt-out rates.
The beauty of predictive analytics is that these tools can provide decision makers with impressive qualitative and quantitative results. Healthcare industry practitioners have frequently used these tools to predict a patient’s ability to pay as well as which customer accounts have the highest likelihood of ending up in collections. Providing healthcare employees with these types of insights enables them to do their work more productively and efficiently.
When data scientists ask the right questions using predictive analytics, they’re not only able to address known business challenges but they can also uncover some of the unknowns.
As an example, a growing number of law enforcement agencies are using predictive analytics along with visual analysis and geo-spatial analysis to better predict when certain types of crime are most likely to occur (e.g., a six-month analysis of a four-block radius for a particular city reveals that muggings are mostly likely to occur between 2 p.m. and 6 p.m. on weekends, leading police commanders to increase their patrols for this area and time period).
Obviously, companies need starting points for applying predictive analytics. You should begin by using predictive analytics against a couple of top business challenges and then build upon lessons learned. As a recent PwC report on predictive analytics points out, a terrific question worth asking is this: What will the cost be to our company if management makes a “wrong” business decision?
Next steps: For more information on this topic, check out our complimentary “5-Minute Guide to Business Analytics.”
