What do you mean, we?

Enough with the "we" stuff about fixing web advertising.

This is not a "we" problem. "We" can't promise to replace "ads that provoke blocking" with "better-performing ads", because ads that provoke blocking are the high-performing ads. As a web user, you're not seeing crap ads because the advertisers want to waste money and annoy you. You're seeing them because they test well.

Crappy, annoying, deceptive ads get clicks.

The terrible stuff on the web is there because it works.

Everyone agrees that "we" need to get rid of "bad" ads. Naturally, "we" is defined as "you" and "bad" is "not the ads that work for me." But because the same qualities that get response also provoke blocking, there's no equilibrium strategy here.

Imagine that all the right-thinking people agreed to L.E.A.N. or some other set of self-regulation terms. No auto-playing videos, no NSFW animations, no fake error dialogs.

The more that self-regulation limits crappy/click-getting practices, the more incentive for any advertiser who is willing to bend the rules and offer a little more money to run an ad that's a little bit creepier, a little more attention-getting or finger-fumble-attracting.

Incentives for bad practices are there because users can be tracked from site to site. That marginally extra-annoying advertiser will be able to find a publisher with marginal reputation, who claims to be able to reach the desired users and is willing to accept the ad. And self-regulation breaks down, or never really gets going in the first place. Cross-site tracking gives everyone an incentive to do advertising that gets clicks today and provokes ad blocking tomorrow.

So there's no "we" solution. The fixes for the web advertising problem will have to happen one user at a time. Every user who becomes harder to track from site to site helps give high-reputation sites a little more market power to enforce ad standards.

Publishers and brands need action from users

In today's web advertising, high-reputation and low-reputation publishers compete to reach the same users. And high-reputation brands are hard to tell apart from low-reputation ones.

High-reputation publishers and brands win when users get less trackable, but users have to be the ones to take the action.

So instead of putting everything in terms of "we", it's time to think about reciprocity and measuring the benefit from each additional tracking-protected user. Instead of hippy-dippy "we" stuff, relying on everyone to cooperate, let's talk exchange of value. Big Data is not just a tool to help with low-reputation strategies. Data-driven projects can help with high-reputation strategies, too.

Questions might include:

  • Which customers gain the most value to me when they're protected from tracking by low-reputation competitors? (For an HMO, what's the net present value of protecting a customer from quack diet ads? For a car insurer, how much is it worth to keep the most profitable customers from being picked off based on their social media usage?)

  • Which categories of readers are most valuable to the best advertisers on my site? How much does it cost me when adtech intermediaries can follow them elsewhere? What's a cost-effective tracking protection solution that I can offer them, to keep them from being reachable on low-value sites?

I'm not against "we need to work together" messages in situations where a cooperative solution is really workable or necessary. But for fixing web ads? Time to give it up.

Don Marti · #