A Question for BTB Mailers

Share this article:
I have long been puzzled by the fact that so few business-to-business mailers take advantage of the best way to increase response rates -- namely, to use a penetration analysis by SIC against a national six- or eight-digit SIC master file of businesses.


I can understand why so few that run such a share of market report don't take the next step to use transactions and dollars and purchasing patterns to produce a mailing model. The reason is that such a second step is costly and, except for very large files, it does not add enough significant, profitable results to make it worthwhile.


But why, when better data is at hand do mailers insist on selecting primarily by four-digit SIC classifications, when a penetration analysis is guaranteed to produce higher response rates than any selection made by any mailer or broker or consultant? Is it because they don't believe the figures?


Share of market simply compares counts of customers by SIC against the total universe of all sites with that SIC code.


The penetration run produces a penetration directly logged on the file for each two-four-six-eight digit segment by SIC.


Only tyros make selections by two-digit SIC. It is obvious this is "too coarse for comfort." But while a given two-digit SIC may be bypassed for lack of response, a given four-digit segment of that two-digit SIC may be attractively responsive.


In the same way, a four-digit SIC overall may be poor potential, but segments within that four-digit classification can be very valuable.


So, a penetration analysis arrays in descending order percentages of penetration for all segments of your file -- which then can show top potential (prospect counts) at a given cutoff percentage. This provides potential quantity available (cumulative) at any level of penetration you wish to measure.


Also note that your file need not be coded by SIC. By merge-identification against the universe, your records are coded for the analysis. If you want such new SICs to be appended, there is a separate charge for this.


The following results illustrate the power of this method:


o A tool wholesales client of mine for years purchased (rented) every manufacturer, every year. When he was convinced to do the analysis, he saw from his own penetration, at the eight-digit level (including those two- and four-digit segments that qualified), that hardly any portion of "printing and publishing" (over 1/16th of all manufacturers) was viable. This changed the way he mailed to manufacturers.


o A major compiler of non-household mailing lists originally considered any business their market and sent catalogs to any site. But when a penetration of at least zero to five (five responses per M) was set at break-even, only 2 million of the 11 million sites on the file were mailable. This reduced the universe he wanted to mail but saved the hidden cost of mailing any part of the 9 million found below the line.


o A major computer manufacturer with a large customer file requested selection of their top 100,000 prospects. This turned out to be a penetration of more than 5 percent. In other words, this mailer already had 5 percent or more of the SICs in the SIC segments with the highest potential.


o One mailer was convinced that wholesalers were not for them -- until a penetration analysis came up with 28 list segments totaling 315,000 prospects with more than their average response range. They were right about wholesalers as a whole, but were leaving some attractive money on the table by refusing to mail to any prospects coded 50 to 51.


In general, it's not difficult to increase average response rates by 15 percent by drawing the line on the array and mailing all segments above that line which have a given penetration of "X" or more.


Business mailers do not stay enslaved at the four-digit SIC level.


Note: The larger your customer file, the more certain the result. If the customer file, however, has less than 4,000 or 5,000 records, the penetration by individual SIC segments may be quite thin -- and the quantity found of well-above-average prospects may be disappointingly small.
Share this article:

Sign up to our newsletters

Follow us on Twitter @dmnews

Latest Jobs:

Featured Listings

More in Database Marketing

What's H-appending? DiscoverOrg Taps Marketo's Webhooks

What's H-appending? DiscoverOrg Taps Marketo's Webhooks

Cloud-based marketing automation behemoth Marketo joins forces with marketing intelligence company DiscoverOrg to improve its data collection capabilities.

A Toast to Marketing Attribution

A Toast to Marketing Attribution

Vino accessories and storage company Wine Enthusiast indentifies top and underperforming affiliates using algorithmic marketing attribution.

Q&A: When (and How) to Bust Down the Data Door

Q&A: When (and How) to Bust Down the ...

Some people run into issues with trying to build the perfect solution when often an 80% solution will do, says MailChimp's chief data scientist.