Catalog Doctor: Rules to Live By
10. Run a "no test" test every few years. That is, segment your list into two groups and mail them identical catalogs with identical offers. A no-test test will let you see the effect of random probabilities. You'll often see what appears to be a statistically significant difference between the two (identical) test arms. That random probability factor can help you better assess actual results of "real" tests. No-test test results will also help you and your team better understand why you should test at least twice before rolling out a new feature/functionality. Testing a second time will help eliminate the random probability portion of the results.
Susan J. McIntyre is Founder and Chief Strategist of McIntyre Direct, a catalog agency and consultancy in Portland, Oregon offering complete creative, strategic, circulation and production services since 1991. Susan's broad experience with cataloging in multi-channel environments, plus her common-sense, bottom-line approach, have won clients from Vermont Country Store to Nautilus to C.C. Filson. A three-time ECHO award winner, McIntyre has addressed marketers in Europe, Australia and New Zealand, has written and been quoted in publications worldwide, and is a regular columnist for Retail Online Integration magazine and ACMA. She can be reached at 503-286-1400 or susan@mcintyredirect.com.
