Do You Know What Really is Driving Your Demand?

Before the advent of the Internet, catalogers could trace 80 to 85 percent of their business to a specific source or key code. They knew where the business was coming from and could make sense of the results as detailed by the source code report, and they didn’t need to match back Internet and non-traceable results to a specific code. They simply could allocate non-traceable results proportionally across all key codes.

But today, with an average 35 percent of business coming through the Internet, and the same 15 to 20 percent non-traceable factor as before, you’re lucky to trace 50 percent of your sales to a specific source code.

This month I’ll discuss the importance of understanding what really is driving today’s order/revenue demand, and I’ll consider treatment of Internet-only buyers. I’ll answer frequently asked questions surrounding Internet buyers as an important component to the multichannel catalog marketing mix.

Results Can Be Deceiving

If you’re like most catalogers, you probably have considered reducing mailings to Web-only buyers as a way to save money, because undoubtedly when you look at your source-code report, it appears these buyers perform poorly.

The results of zero-to-12-month Internet buyers (as reflected on the source-code report) often don’t look good — at least on the surface. It’s logical to assume, then, they shouldn’t be mailed a catalog, at least not as frequently as catalog buyers who fall within the same RFM housefile segments. But is this really wise?

Before you decide if Internet buyers should be mailed, have your service bureau do a matchback, a process in which your order file is “matched back” against your recent mail tapes to give credit to the proper source code. It’ll tell you where business is coming from and which key codes should be given credit for the sale — even Web orders. Matchbacks are becoming commonplace, and for good reason. They’re a necessary part of cataloging.

The chart “Source Code Results” outlines our findings from a recent test and matchback. You’ll see in it the value of mailing to Internet-only buyers. Indeed, reducing circulation to Web buyers can have a negative impact on your demand revenue and profit contribution.

For most catalogers, Web buyers rank in the bottom 10 cells for housefile when reported by key code. But after a matchback, they often rank in the top 10 cells. The clearest example is in a holiday mailing. There were no Web cells in the top five when ranked. But after the matchback, there were three Web-only source codes in the top five housefile segments, proving that Web buyers are indeed very strong responders. They get a catalog, order via the Web, and we lose traceability. But that doesn’t mean they’re not buying. This confirms that the catalog is the biggest driver of traffic to the Web, and these customers must be mailed a catalog regardless of what channel they use to place their orders.

Last year we did a split test and created a hold-out panel (see chart). We found that it absolutely paid to mail the Internet buyers a catalog. We created two panels of about 25,000. One panel was mailed seven times the other only once. The net contribution for the seven-time group was $169,089 compared with the one-time group of $109,318. So the additional six mailings to Web buyers generated another $59,771 in profit contribution.

In years past, when the non-traceable factor ranged from 15 to 20 percent, you could allocate back the unattributed across all source codes on a proportional basis. But this method of allocating non-traceable results won’t work today. Here’s why: Matchbacks have shown it’s inappropriate to give equal weight to a housefile, inquiries, co-ops and rented lists. Every time we perform a matchback, we find that 50 to 75 percent of the Internet results should be allocated to the cataloger’s own housefile! Another 10 to 20 percent should be allocated to outside rented lists (this varies based on how much prospecting a company does). The point is that true allocation is far from proportional.

In our study, Web buyers were split into two equal panels. In drop No. 1, panel one and two were mailed. In drops Nos. 2 to 7, panel one was mailed, but panel two was not. The results reflect the full-season activity on the codes.

Other Benefits

You may find lists comprised of buyers who are heavy Web users. Such lists may look as if they don’t perform well, when in actuality they’re comprised of profitable customers. In this case, without a matchback you’d never know that. Also, you may assume you can’t find lists that perform well, when you actually have a few winners to add to your continuations. Or you could have lists that look like they’re falling off, or your total rentals look like they’re trending downward. This might be just a result of heavier Web sales.

Without knowing the true performance of all segments mailed, you could make false conclusions that influence your marketing strategy. A matchback will keep you on track and give you the confidence to make sound decisions.

Stephen R. Lett is president of Lett Direct, a catalog consulting firm specializing in circulation planning, forecasting and analysis. Call (302) 537-0375, or visit www.lettdirect.com.