Ads.txt is an initiative the IAB Tech Lab launched earlier this year to ensure greater transparency in the programmatic marketplace by creating a record of who is authorized to sell or re-sell ad inventory.
The top aim of Ads.txt is to crack down on domain spoofing, a widespread, wily and wickedly expensive problem in programmatic. Domain spoofing was spoken about quietly for a long time, but fraudulent players’ efforts have intensified so much—think the highly public Methbot operation, where fraudsters were netting an estimated $3-$5 million per day in late 2016—that the issue has become a very popular discussion point.
Domain spoofers will create domains that appear at first glance to belong to premium sites, but they’re actually counterfeit domains, and their traffic is non-human traffic. To avoid buying from spoofed domains, buyers have historically created whitelists and blacklists. It’s not hard to create those lists, but it’s a pretty manual process, which can be time-consuming, and fraudsters register new fake domains faster than lists of legit sellers can be maintained.
Another consequence of Ads.txt is that by specifying who is authorized to sell a publisher’s inventory, it could limit arbitrage in the programmatic space. Arbitrage happens frequently, but compared to domain spoofing, not everyone is quite so eager to talk about it (especially companies who benefit from it).
The way it works is quite simple. The Ads.txt file is a little text file the publisher’s webmaster posts to the site. In that file, you have a list of every selling partner the publisher deals with—think SSPs and ad exchanges—and who is authorized to sell their inventory. Each seller also has a distinct ID listed in the file. That ID matches the seller’s ID in the bid request. The publisher can specify whether the seller in question is a direct partner with the pub or an authorized reseller. Ads.txt also allows for multiple sellers to be authorized to sell the same inventory, which might come up for publishers who syndicate their content out.
Buyers can search for Ads.txt lists and set up filters to make sure they’re buying only from authorized sellers. The IAB Tech Lab has also released a web crawler that can accelerate the process of sifting through and verifying IDs in publishers’ Ads.txt files.
The catch (of course there’s going to be one of those!) is that in order for Ads.txt to work for buyers, it needs broad adoption by premium publishers, and the industry doesn’t really have that yet. Ad Ops Insider did some research and found, as of September 2017, 12.8% of publishers had posted Ads.txt files. Why not? To some pubs, their dev teams are already overloaded, and implementing Ads.txt just isn’t a priority. Some others might benefit from arbitrage as it is, and may be hesitant to get out of that game. Among publishers who have adopted it, though, there’s a wide variety of content verticals and levels of “premium-ness.”
Without a scalable effort from publishers, media buying agencies have been hesitant to insist on buying from only Ads.txt-verified inventory. But major players like Google, Digitas and most recently MediaMath have announced their intentions to move toward Ads.txt-only transactions. If the past of digital media is any indication, that buy-side pressure will push more publishers to make Ads.txt a higher priority.
We’ll be discussing this in greater depth at the upcoming AdMonsters Publisher Forum in Nashville, Nov. 5-8.