I saw this post last week at the blog ‘Neuromarketing’ and a longer one this week on New Scientist’s website: ‘Mind-reading marketers have ways of making you buy‘. New Scientist picked the cover of the new issue by having a company called NeuroFocus hook 19 male readers of their magazine to an electro-encephalograph (EEG) machine and then observe their reactions to three potential cover designs.
Each design was rated according to how ‘attention getting’ or ‘novel’ it was (among other factors). The cover selected as the winner was chosen for use as this month’s actual cover.
Setting aside the obvious issue of sample size, as I see it, there are a couple of other problems with this methodology: One is that the potential covers were only tested against each other. Two of them used the title ‘Shattering spacetime’ for the cover article and the third used ‘End of Spacetime’, so perhaps it’s not surprising that the one that was different from the other two (the one that actually won) was rated more ‘attention getting’ and ‘novel’.
If New Scientist wants to sell the most magazines, it should test its covers against recent covers from Astronomy or Scientific American or other magazines that appear in the same section of the rack as New Scientist, not against two cover designs that will never see the light of day.
I remember reading somewhere that Sesame Street has been doing this for decades. Before airing, each episode of ‘The Street’ (as I call it) would be shown to a test group of children along side a slideshow (called ‘The Distractor’) of all sorts of different images. If more kids paid attention to the slideshow, the scene would be reworked or dropped entirely from the episode.
The other big problem is that New Scientist should be measuring actual sales, not ‘novelty’. They claim that printing a split run with multiple different covers would be cost-prohibitive (though I’m not exactly certain why, since it would seem like extremely valuable data to have after having done the EEG testing, as a way to ensure that the “neuromarketing” methodology was useful).
Alternatively, perhaps New Scientist could have run multiple image-based ads on the web with identical designs but different covers (saying something like ‘Click here for a special in-depth look at the upcoming issue of New Scientist‘), to see which design got the highest click response rate.
It seems like it would have been worth it as a means to ‘close the loop’ on the data collected.