I used to work at the FT, and we tested a number of hypotheses: 1) faster page loads means better engagement 2) everyone hates popups. 3) everyone hates autopl

I used to work at the FT, and we tested a number of hypotheses: 1) faster page loads means better engagement 2) everyone hates popups. 3) everyone hates autoplay turns out that faster page loads _do_ make user happy, normal non tech savvy users happy. (source: http://engineroom.ft.com/2016/04/04/a-faster-ft-com/ http://engineroom.ft.com/2016/04/04/a-faster-ft-com/) I don't think the popup experiment has a write up, but people _hate_ popups, even more when they are not relevant. I was part of the auto play experiment, and whilst a few people really like them, it saves them clicking, the vast majority hate it with a passion. Auto play is a thing because it boosts ad impressions between 25-80% So for ad based revenue, yes, I can see why people do this. For subscribing sites, it kills your bottom line. So empirically they either rely too much on ads, or are idiots who don't have evidence to base decisions on.