By Mike Follett, CEO at Lumen Research
Like many female scientists, Elisabeth Vrba doesn’t get the credit she deserves. She came up with the theory of *exaptation* in partnership with Stephen Jay Gould in 1982. Her name is right there as the co-author of the paper! But the world remembers Gould, not Vrba.
Vrba’s paper describes how biological traits evolved for one function can be later repurposed for another. Feathers, for instance, may have evolved for insulation before being adapted for flight. The bones in the inner ear of mammals: probably initially used as jaw bones, they have been co-opted for hearing. Our mouths are not *made* for talking — but they do a lot of it nowadays.
In ad tech, we face a similar challenge: how do we measure attention when our industry wasn’t designed to do so?
Rather than building from scratch, we can repurpose existing signals—developed for other tasks—as indicators of attention.
Take *viewability signals*: originally designed to cut down on ad fraud, they can now be repurposed as a key indicator of attention.
*Audience signals* — first developed for targeting – can also tell us about attention likelihood. Salience is a function of relevance to the viewer, so better targeted ads tend to get more attention.
Then there are *interaction signals*. Mouse movement, scroll depth, and scroll speed – originally UX metrics – can be exapted to gauge active engagement. If a user pauses their scroll over an ad, chances are they’re paying attention.
We don’t need to reinvent the wheel; we need to reinterpret and recombine the data we already have. Ciaran O’Kane, when discussing Experian, calls this ‘super signals aggregation’. Elisabeth Vrba would call it exaptation.
If you have an ad tech dataset that you think could be usefully exapted to predict attention, please get in touch: Lumen Research would love to partner with you.
Sources: