Testing research methods

I found an interesting article on CRM Daily that seems to up-end one commonly held belief about personalization. Normally, personalized pieces (Dear Bob) tend to get slightly better response rates than non-personalized mailings (Dear Neighbor). The next logical step would be try and deepen that sense of relationship by making the sender more human. The theory would be that response rats will improve if the potential respondent feels like they somehow "know" the one making the offer.

One researcher tested this theory by putting a photo of the "researcher" on the cover letter of his piece. He found that recipients who would have seen a picture (he used 4 different photos) responded at the same rate as those who would not have seen any photo at all.

My only issue with the research is that it does not address the most difficult hurdle in all of direct-mail: Getting the person to open the envelope. My theory is that his response rates weren't different because what's inside the envelope is irrelevant until they open the envelope. If the recipient opens the piece, they have self-selected themselves into a group that already has a higher propensity to respond to whatever is inside.

Is his research completely useless? No, because it shows that personalization does not necessarily improve response rates for those that get past the first obstacle. But I would really like to see if personalizing the outside makes a difference or not.

Link to the article on CRM Daily: Putting a Face on Market Research.