Wednesday, January 27, 2016

Turn Aging Behavioral Data From Problem to Opportunity

Database marketers focused on quick response to customer behavior tend to discount aging or expired behavioral data, creating an ongoing "data atrophy" problem. And that's a mistake in our experience. We agree with veteran database marketer Stephen Yu's recent Target Marketing magazine post, which argues that marketers need to see their aging and expired behavioral data as an opportunity rather than a problem. Issues arise because while targeting is improved when demographic or "firm-ographic" data is combined with behavioral data (transactions and clicks, for example), behavioral data is both harder to collect than geo-demographic data, which can be appended to fill gaps, and has a shorter shelf-life. The value of a hotline list evaporates quickly, and delayed response to real-time mobile or online actions can misfire, even backfire. But aging behavioral data still has value, and formerly hot data can be warmed up--especially if handled appropriately as Yu suggests. One way is to go from simple time stamps to measurements of intervals between events. How many weeks have elapsed since the last purchase? What are the average number of days between transactions? What is the average number of weeks between new product release and actual purchase? Marketers should also measure by channel to catch when an in-store or catalog buyer becomes an online buyer, and for which items. Yu points out that by collecting, maintaining and transforming historical behavioral data, marketers can use it for more effective targeting and personalization. Scored behavioral data become predictors in models identifying “cutting-edge buyers,” “bargain seekers,” “online buyers of repeat items,” “infrequent high-value customers,” “frequent small-item buyers,” for example. Yu concludes: "Today’s data become historical data in a blink, but we still have a lot to mine there. And such mining is possible, only if we arrange the data properly and let it age gracefully using statistical techniques. That is the way to personalize messages constantly for everyone, instead of reacting to real-time data only sporadically for a fraction of your audience." For the whole post, see

No comments:

Post a Comment