The flaw in Ads.txt
Fraudulent inventory is estimated to cost publishers and advertisers billions of dollars a year and their are countless companies focused specifically on solving this problem. It’s a massive issue that is being tackled industrially by increasing visibility through Ads.txt and encryption through blockchain. While blockchain has the current problem of scalability, Ads.txt might be flawed fundamentally.
Ads.txt helps publishers to makes sure that their sites are not emulated by fraudsters aiming to siphon off ad revenues by highlighting their ‘certified’ suppliers and relevant ID’s. If someone claims to be a publisher without Ads.txt they will not receive ad traffic and if they duplicate the Ads.txt file then they end up paying the legitimate publisher for their bootleg traffic. It solves a problem, but it might also make competition between certified supply partners impossible.
The issue is that if you let everyone know where they can buy your certified inventory from then they are likely to shop around for the best deal. If all of the re-seller inventory is of the same quality then buyers will just use the supplier that hits the floor rate first. Once everyone hits the floor rate then it’s the platform that can absorb the largest trade deal who wins.
We review all of the unique ads that are served on applications and websites. Our primary goal is to find inappropriate and ineffective ads, but we also help our publishers to understand trends in their advertising and advertisers. The elimination of insights from the reseller market actually works to our favour. However, that loss, along with additional margin doesn’t seem to be in the best interest of publishers.
Overall, Ads.txt is easy to implement and if removing the fraudsters is worth more than additional margin and the market insights from certified ad suppliers then it may prove to be the best solution. It may well be a necessary cost to pay while we wait for the next evolution in blockchain technology but I believe it does have a rather large flaw.