But, how will that work, and are there any concerns? The answers are below.
How did the industry come to ads.txt?
With the rise of the Internet, digital ads became the driving force for many brands. Increasingly efficient the technologies have become with the evolution of programmatic ad buying when volumes of big data based on precise targeting extended measurement metrics are enabled. Nearly universally acknowledged as the nowadays direction in efficient media buying it guarantees lots of gains to publishers and advertisers’ parties.
When this is added to the unprecedented scale and streamlined process of media optimization available in programmatic, using an RTB-based open exchange is a very attractive prospect for brands looking to gain the maximum return on their marketing budgets. Ad fraud, on the other hand, also follows the revenue, including the well-known nightmare for all the programmatic buying chain participants – domain spoofing when a genuine domain name is hijacked so advertisers think they are buying an impression on a premium site. And the more valuable ad placements are the bigger the risks are, as with the recent Russia-based botnet that generated an estimated $3 to 5 million per day – targeted the premium video advertising ecosystem. The operation spoofed around 6,000 premium domains and used bots to watch around 300 million video ads per day on these falsified sites. This is probably the biggest case of a massive proliferation of misrepresented inventory.
Here is where ads.txt comes into play.
With the aim to eliminate the ability to profit from counterfeit inventory in the open digital advertising ecosystem IAB Tech Lab has proposed the new open standard for improving transparency in the digital programmatic supply chain.
Called ads.txt (Authorized Digital Sellers), the solution represents an easy to implement and secure process of declaring who is authorized to sell publisher inventory by simply listing Publisher IDs in in the ads.txt file.
How does it work?
A publisher must place an ads.txt file, as illustrated by Business Insider, that lists all of his authorized sellers in the highest-level directory of the website (or the root of the domain). Programmatic buyers can then scan ads.txt files using a re-coded according to their preferences reference crawler written by the IAB OpenRTB Working Group to match the ads.txt lists against the data available in the OpenRTB protocols.
Is it generally applicable?
Not yet, the spec meets only the most common scenarios and there are numerous open items:
- The solution hasn’t been adjusted to non-web environments like mobile apps, OTT solutions, etc.
- Ads.txt isn’t ready for those who are in Content Syndication.
- When working with content syndicators, at least two pieces of information have to be verified. First, it’s important to validate the domain that the content is syndicated through, second, the domain that the content is produced from.
- Ads.txt is ready for the four most common formats (Banner, Audio, Video, Native), leaving aside many of the new and widely adopted formats. Along with that, many publishers partner with different exchanges for different ad formats. That means publishers have to reach out many platforms to gather their seller accounts’ IDs in a single place.
- Some publishers block the identity of their web sites (so-called blind/anonymous inventory) in order to protect their direct sold models. The only thing the parties can do here is build a blind trust between themselves.
The other major opportunity for fraudsters to make profits is traffic blending, which is perfectly illustrated by the IAB OpenRTB Working Group with the example of a brand coffee blend.
Let’s say you go to a premium coffee store you probably will get an authentic blend of coffee beans for $25 per pound of a certain brand, if you go to a grocery store and you buy the same brand for $7 it’s probably 5-10% of the declared brand and the rest is something else. To market you a cheaper product, they’ve blended authentic beans with unauthentic ones. It works in the same manner with media buying. Imagine, you as a DSP have several ad exchanges where you purchase your audience reach, but one of the exchanges is selling you misrepresented domains blended with original domains in order to meet your criteria. You’d definitely like to know which of the domains are fake and cause you all those issues, but that’s highly complicated to identify considering how huge are the volumes of domains especially when a single domain can come from several exchanges. See the chart below that gives you a detailed description of the traffic blending process.
However, these issues are minor compared to the problem the standard is about to solve. The industry has been remaining too long safe and comfortable for bad actors to make their “business”. Now, it’s high time for buyers to weigh in on the topic of audience quality and that’s them who have to insist on the industry-wide adoption of the solution. As much as the buyer side, publishers will gain from the adoption, as now, they have the chance to receive fair yields, which usually have been absorbed by shady entities and fraudsters.
Considering the fact that the solution is simple and doesn’t require additional development and resources we all hope to see it implemented in the nearest time, what is very promising.