Scroll depth without a fingerprint

Scroll depth is one of the few engagement metrics that reliably changes what people write. If nobody reaches the third section, the third section needs to move or go.

It also has a bad reputation, because the tools that measure it tend to measure everything else at the same time: window size, device orientation, pointer movement, all of it combining into something that identifies a browser rather uniquely.

None of that is necessary. Quarters are enough. Record whether somebody reached twenty-five, fifty, seventy-five or a hundred per cent, and send that as a single number attached to the page rather than the person.

Four buckets cannot fingerprint anybody, and in two years of using it we have never once wished we had finer detail. “Half the readers never see the conclusion” is the finding, and it does not get more actionable with decimal places.