Site engagement, consent management, bargains
06 November 2018
One common growth hacking pattern on social and collaboration sites is to build user profiles incrementally. Capture just enough info to get the user logged back in, then get them started using the site. As they get into it, prompt them to fill in more and more profile information. You have probably seen this on new sites where you have to make an account. FIXME: list of good examples here.
People don't want to give up a bunch of information up front before they see how good the site is. And, I suppose, if the site is good enough that the person thinks they'll spend more time on it, they're more likely to provide correct information than all the residents of "asdf" born on January 1, 1970.
But news sites don't take this approach. Instead of trading a little value for a little information, repeatedly, you get one big dialog asking you to give up all your information before you even read the first story.
Does the same incremental approach that applies to data collection for social and collaboration sites also apply to news sites? Preliminary results from Global Consent Manager tend to indicate that yes, it does.
So here's the bargain. Right now, the web ad business is set up to bid on ad impressions that come with third-party data, way more than for impressions without third-party data. So a trackable bot impression on a fraud site can produce more ad revenue for the fraud operator than an impression from a privacy-sensitive user running Firefox Nightly or Apple Safari produces for a legit site.
Yes, even though the privacy-sensitive user is more likely to be human and interested in buying something related to the topic of the site.
The opportunity to get a bargain is: instead of relying on conventional programmatic ad buying, if you do a little extra work to understand the audience of specific sites, you can reach more of the humans you're interested in.
Not every Firefox user who shows up on the Road and Track site is going to buy a car this year, but $1 worth of ad impressions there is likely to reach more human car buyers than $1 spent programmatically—because you get a higher fraction of humans for a lower price. and third-party data on who's a likely car buyer is bogus anyway, but that's another story.
This opportunity is likely to go away as more agencies figure it out, but right now it's a great chance to get humans cheaper than bots.