by Anonymous

Last year I was “volunteered” by management to conduct a subscriber survey of our publication, which had not been done in four or five years. The previous person who conducted marketing research would survey using the tried and true method:

  1. Design the questionnaire
  2. Put it through layers of bureaucratic approval
  3. Print it
  4. Mail it
  5. Wait for responses to trickle back in
  6. Send the stack of filled questionnaires to a contract data entry person
  7. Import the results into SPSS
  8. Print out ugly, boring tables
  9. Put these ugly tables into ugly, boring reports
  10. Don’t share the findings with the sales force in any timely manner

I decided to break the mold—I didn’t want the pain of such a manual system in 2006 A.D.

I found an ASP-based survey tool and proceeded to create my questionnaire in it, with the hope of e-mailing it out to subscribers, having them click on a link to take the survey, and have the results instantly arranged into graphs.

When I went to let colleagues know I was doing it, the editor-in-chief and circulation VP had a cow. They went to my boss saying that I didn’t know what I was doing, and that they were afraid the results would be different from the traditional snail-mail results, and because the number of e-mails we had was so small, the pool we had to solicit from would not produce a valid survey.

I argued that the only reason we didn’t have a higher proportion of e-mails was because we had never required them so we were behind the curve on gathering them. Besides, our readership was educated, and specifically, a large portion are active duty military and their employer, Uncle Sam, is sure to give them e-mails even if they don’t have a home personal email account. There would be no significant difference in results because so many people nowadays have Internet access.

Well, my boss made me compromise—I would send out about 7,500 e-mails with links, and about 7,500 letters asking them to go to a link, also giving them the option to call a toll free number and request a hard copy questionnaire to be sent to them with a business reply envelope to send it back at no cost to them. By then I had a tad of doubt in myself and caved in.

Only 1% of the snail mail recipients asked for hard copy questionnaires; I had to wait for these to trickle back and enter them into the survey myself. There were an adequate number of recipients who made the effort to get from their mailbox to a computer screen and participate; while the email-originated responses were at a higher rate than snail mail-originated responses (10.3% vs 5.8%), there were slight but not significant differences in demographics (the snail mail responders tended to be older, have older children, and have slightly more household income), but more importantly, the subjective answers regarding perceptions and preferences about our publication were identical.

So I was right after all—most of the snail mail respondents took the online survey—so we didn’t have email addresses because we failed to ask for them, not because there was this huge chunk of audience was bunch of Luddites. The only Luddites were the editor and circulation guy. Just to keep them happy I had to spend an extra $5K for this one survey.


Posted August 6, 2007


Do you have a story to tell where you were able to “do it wrong quickly”? E-mail your entry to me and maybe it will appear on this site, too. For more tips on how to super-charge your Internet marketing, order Do It Wrong Quickly and Search Engine Marketing, Inc. today.