Certainly there wouldn't be a 100% match between "would be worth advertising to" and "would pay to remove adverts", but the problem is nobody (afaik) has any da

Certainly there wouldn't be a 100% match between "would be worth advertising to" and "would pay to remove adverts", but the problem is nobody (afaik) has any data on to what extent it might correlate, and so are scared by it. I wonder if companies like the Guardian have thought about trying to compare between the two audiences, i.e. how people who now subscribe interacted with adverts before subscring vs. audience that doesn't pay. Thinking of some of the tracking tools that could be used... it wouldn't be impossible (though wouldn't be imperfect data either), but it would be a rather complicated thing to set-up so it wouldn't surprise me if they don't have it in place.