06 Feb 2018 Search

(I work for Mozilla. None of this is secret. None of this is official Mozilla policy. Not speaking for Mozilla here.)

Guess what? According to Emil Protalinski at VentureBeat, the browser wars are back on.

Google is doubling down on the user experience by focusing on ads and performance, an opportunity I’ve argued its competitors have completely missed.

Good point. Jonathan Mendez has some good background on that.

The IAB road blocked the W3C Do Not Track initiative in 2012 that was led by a cross functional group that most importantly included the browser makers. In hindsight this was the only real chance for the industry to solve consumer needs around data privacy and advertising technology. The IAB wanted self-regulation. In the end, DNT died as the IAB hoped.

As third-party tracking made the ad experience crappier and crappier, browser makers tried to play nice. Browser makers tried to work in the open and build consensus.

That didn’t work, which shouldn’t be a surprise. Imagine if email providers had decided to build consensus with spammers about spam filtering rules. The spammers would have been all like, “It replaces the principle of consumer choice with an arrogant ‘Hotmail knows best’ system.” Any sensible email provider would ignore the spammers but listen to deliverability concerns from senders of legit opt-in newsletters. Spammers depend on sneaking around the user’s intent to get their stuff through, so email providers that want to get and keep users should stay on the user’s side. Fortunately for legit mail senders and recipients, that’s what happened.

On the web, though, not so much.

But now Apple Safari has Intelligent Tracking Prevention. Industry consensus achieved? No way. Safari’s developers put users first and, like the man said, if you’re not first you’re last.

And now Google is doing their own thing. Some positive parts about it, but by focusing on filtering annoying types of ad units they’re closer to the Adblock Plus “Acceptable Ads” racket than to a real solution. So it’s better to let Ben Williams at Adblock Plus explain that one. I still don’t get how it is that so many otherwise capable people come up with “let’s filter superficial annoyances and not fundamental issues” and “let’s shake down legit publishers for cash” as solutions to the web advertising problem, though. Especially when $16 billion in adfraud is just sitting there. It’s almost as if the Lumascape doesn’t care about fraud because it’s priced in so it comes out of the publisher’s share anyway.

So with all the money going to fraud and the intermediaries that facilitate it, local digital news publishers are looking for money in other places and writing off ads. That’s good news for the surviving web ad optimists (like me) because any time Management stops caring about something you get a big opportunity to do something transformative.

Small victories

The web advertising problem looks big, but I want to think positive about it.

  • billions of web users

  • visiting hundreds of web sites

  • with tens of third-party trackers per site.

That’s trillions of opportunities for tiny victories against adfraud.

Right now most browsers and most fraudbots are hard to tell apart. Both maintain a single “cookie jar” across trusted and untrusted sites, and both are subject to fingerprinting.

For fraudbots, cross-site trackability is a feature. A fraudbot can only produce valuable ad impressions on a fraud site if it is somehow trackable from a legit site.

For browsers, cross-site trackability is a bug, for two reasons.

  • Leaking activity from one context to another violates widely held user norms.

  • Because users enjoy ad-supported content, it is in the interest of users to reduce the fraction of ad budgets that go to fraud and intermediaries.

Browsers don’t have the solve the whole web advertising problem to make a meaningful difference. As soon as a trustworthy site’s real users look different enough from fraudbots, because fraudbots make themselves more trackable than users running tracking-protected browsers do, then low-reputation and fraud sites claiming to offer the same audience will have a harder and harder time trying to sell impressions to agencies that can see it’s not the same people.

Of course, the browser market share numbers will still over-represent any undetected fraudbots and under-represent the “conscious chooser” users who choose to turn on extra tracking protection options. But that’s an opportunity for creative ad agencies that can buy underpriced post-creepy ad impressions and stay away from overvalued or worthless bot impressions. I expect that data on who has legit users—made more accurate by including tracking protection measurements—will be proprietary to certain agencies and brands that are going after customer segments with high tracking protection adoption, at least for a while.