In response to Richard West’s article, Tidy Profit in Junk, I think there needs to be a greater emphasis on the role of predictive analytics to enable better targeting and return on investment for direct marketing. But of course, before we can perform advanced analytics we need to collect relevant data.
I don’t totally agree that targeting remains a “very” unpredictable science – there are ways to improve predictability. Data that is priced high should hold a larger number of variables if it’s of good quality. The more relevant variables we have, the more effective and reliable the results of running a predictive model will be against this data.
By finding and eliminating the correlations between the variables we can then highlight the key predictors to improve targeting and therefore increase the response rates and overall performance of the direct marketing campaign. At the same time we can simultaneously reduce the amount of data we need.
We need to work on collecting more variables. The more information on prospects and customers the better the ROI will be – in comparison to a generic mass-mailing approach. Let’s improve our data capture, get a consolidated view of all customer interactions across the business and then apply advanced analytics.
Make an investment now to save later and bin the scattergun approach, or our industry will continue to be received with caution. We may never completely eradicate junk mail, but it can be reduced.
UK country manager