Resources

Ads.txt is an initiative launched by the IAB (Interactive Advertising Bureau) in 2017 to stop the unauthorized reselling of publishers’ inventory and clean up the ad tech ecosystem by increasing transparency in the supply chain.

At the heart of the initiative lies the assumption that publishers can and need to keep track of all their authorized sellers and report to the buy-side about the type of relationship they have with each seller – either Direct or Reseller. This tracking is done through a file called “ads.txt” (which stands for Authorized Digital Sellers text file) in a predetermined path that everyone can access.

The initiative was a success, with the majority of publishers adopting it. Some criticism that surrounds ads.txt is that many publishers do not fully understand the logic of how to work with the file correctly. This confusion can lead to errors that allow many unauthorized sellers into the supply chain, files not being updated frequently, non-existing partners not being removed, and other problems.

To read more about ads.txt, go to this article in the IAB Tech Lab.

App-ads.txt was released by the IAB Tech Lab as an extension of ads.txt that supports both mobile app and CTV environments. It aims to improve transparency in app-specific inventory by providing the same information found in a regular ads.txt file, listing all the app's authorized sellers and more specific data for CTV, such as inventory partner.

Instead of attaching the file to an online domain, app developers provide a website URL within their app store's metadata and publish an ads.txt file on that domain. Buyers can crawl developer sites and gather app-ads.txt data to match the data they receive with a bid request to help them validate the legitimacy of each seller.

There are two ways a publisher can sell their ad inventory through programmatic bidding systems: Reseller or Direct. Reseller means a publisher authorizes another company to sell its inventory on its behalf.

There are two ways a publisher can sell their ad inventory through programmatic bidding systems: Reseller or Direct. Direct means a publisher works with an ad tech vendor or ad exchange to sell its ad inventory directly to the buy-side.

The primary or exclusive monetization partner of the publishers’ inventory.

This means the owner of the site has authorized a third-party company to manage the monetization either globally or locally.

Seat ID is the same as “seller network ID” or “publisher ID.” It is an ID associated with a publisher’s or seller’s account on an ad tech vendor platform or ad exchange platform. The “Seat ID” is a unique number mentioned in an ads.txt file that represents an ad tech company working with the publisher directly or indirectly. The Seat ID is transmitted as part of the OpenRTB protocol as the publisher ID and the publisher’s domain in the publisher’s object.

Domain spoofing is a type of ad fraud where low-quality inventory is passed off as high quality or a premium site. Domain spoofing can be done in several ways, such as URL substitution, where fraudsters deceive advertisers at bid time by substituting a fake URL through an ad exchange or network, so the ad goes to a different site instead of the one that was bid on.

Tricky domain spoofing tactics come in many forms. For instance, custom browsers where bots can make the URL of any site appear as a different premium site, so the ad reports back the "spoofed" higher value URL. Another example is fake human browsers, where malware injects an ad inside a premium site, but the premium site is not paid for the ad. Instead, the fraudsters collect the revenue. Another method is cross-domain embedding, where two sites are paired together – one with high traffic and low-quality content and one with low traffic and safe content.

As noted by the IAB, dark pool sales houses aggregate thousands of ads.txt lines across thousands of domains. Misstating RESELLER traffic as DIRECT puts the traffic at risk because it is misrepresented to demand partners. Partnering with these entities or facilitating such activities might result in reputational damage or even blocked traffic.

Sellers.json is the mirror image of ads.txt. In ads.txt files, publishers report to buyers about their relationship with every seller representing them. In sellers.json files, sellers report to the buyers about the type of relationship they have with each publisher. Both methods help buyers build a fully transparent picture of the supply path to target the most efficient route of each available impression.

Like ads.txt, sellers.json is a file posted in a predetermined place where sellers (mostly exchanges, SSPs, and other ad tech vendors that represent supply) update their relationship status with every publisher. By scanning those files, DSPs can build a full map, including multiple hops/nodes that sometimes occur between the buyer and publisher to help them choose the most efficient path to the publishers and ultimately cut costs.

For example, you can find Primis’ sellers.json file here: Primis.tech/sellers.json

Seller ID is the same as “seller network ID” or “publisher ID”. It is an ID associated with a publisher’s or seller’s account on an ad tech vendor platform or ad exchange platform. The “Seller ID” in a sellers.json file needs to be the equivalent to the same seller’s “Seat ID” in the ads.txt file. This ID is transmitted as part of the SupplyChain object, allowing buyers to validate the inventory source, including the ability to identify direct inventory versus reseller inventory.

Seller type “PUBLISHER” indicates that inventory sold through this entity is on a site, app, or other medium owned by the listed entity. The advertiser pays the publisher directly.

Seller type “INTERMEDIARY” indicates that the listed entity sells inventory on behalf of the inventory owner or a third-party vendor. The Intermediary seller does not own the inventory it sells.

Seller type “BOTH” indicates that both types of inventory are sold through this entity, including sites or apps owned by them, as well as sites or apps owned by others.

In a sellers.json file, the “Confidential” field indicates whether the seller is confidential or not. There are two possible indications; 0 = not confidential, and 1 = is confidential (meaning the publisher’s name and the domain would not appear in the sellers.json file). There might be several reasons for defining the seller as confidential, and it is up to the entity who published the sellers.json file.

The OpenRTB SupplyChain object (aka “schain” in bid requests) enables buyers to see all parties selling or reselling a given bid request. The SupplyChain object is composed mainly of a set of nodes, where each node represents a specific entity that is participating in the transaction of the inventory. The entire chain of nodes, therefore, represents all entities involved in the direct flow of payment of inventory. This lets the buyer see not only which exchange or SSP the inventory came from but also the entire payment trail back to the original content creators.

Buyers.json is an initiative from the IAB Tech Lab. It is designed to bring transparency to the buy-side, in the same way, sellers.json did for the sell-side. The buyers.json file is published by DSPs to identify the names and ID of the buyers they represent. This helps validate buyers and rapidly flag up any suspicious activity to prevent fraud.

The DemandChain Object is a new buy-side transparency standard introduced by the IAB Tech Lab in June 2021. It mirrors the Supply Chain Object, enabling sellers to see all parties involved in buying the creative embedded in a bid response used in OpenRTB.

Programmatic advertising is the process of automating the buying and selling of digital advertising space in real-time through an automated bidding system. Programmatic advertising enables buyers to decide which ad call to buy and how much to pay for it within milliseconds through a sophisticated ecosystem.

Automating the media buying process enabled many small and medium advertisers and publishers to step into the online advertising ecosystem, building internet economics that allowed the world wide web’s exponential growth.

The joining of so many buyers and sellers to the online advertising ecosystem also introduced new threats. Those threats come in the form of unauthorized reselling of traffic, fraud botnets, spoofing domains, and other methods of fraudulent activities that aim to take a share of the $120B yearly ad spend.