“The Sell Sider” is a column written by the sell side of the digital media community.
Today’s column is by Paul Bannister, co-founder and executive vice president at CafeMedia.
The buzz in programmatic media these days is about “exciting” topics like supply-path optimization, first-price auctions, video header bidding and blockchain. One theme that could use more analysis is Ads.txt.
Ads.txt certainly has sparked an industry discussion, but much of the commentary has focused on driving adoption (both on the sell and buy sides). It’s worth exploring in a little more depth the benefits and side effects it will bring to the digital media industry.
A Thoughtful Approach For All Publishers
While many are pushing for publishers to adopt Ads.txt, few are giving recommendations about how to implement it. After talking to many publishers, it’s become clear to me that most are going to be stingy with who they allow into their Ads.txt files. That’s a good thing, and the right approach for all to follow.
Of course, "stingy" means different things to different publishers. If a given publisher was directly or indirectly working with 40-plus ad tech partners previously, reducing the number listed in an Ads.txt file to 20 represents a big cleanup. Someone only working with three partners might keep all three of them in their Ads.txt.
The first benefit is the primary purpose of Ads.txt: a reduction in inventory spoofing. It’s well documented that most large, premium publishers’ inventory is heavily spoofed on exchanges. By forcing DSPs to buy only from named selling partners, the exchanges that were falsely pushing mislabeled inventory (knowingly or not) will be blocked out.
An interesting side effect of this change is that domain spoofers may be forced to move their efforts down to the mid-market. Blocked from spoofing the inventory of sites with over 5 million unique viewers (because most will have adopted Ads.txt), these fraudulent players will start focusing on the next tier of sites. These sites are large enough to end up on advertiser whitelists, but not always savvy enough to set up Ads.txt. It’s critical that this group of sites (and their representatives) adopts Ads.txt and blocks criminals from moving their shell game here.
Deeper Consolidation Within Ad Tech
The second benefit will be a reduction in the number of “innovative” ad tech companies spawning.
Many new companies are nothing but a slightly modified creative format, a small direct sales team and a lot of exchange-based backfill. Their primary purpose is to extract dollars from agencies based on their “new product,” minimize payment to publishers (and therefore working media dollars) and maximize their own revenue. When these companies are blocked from reselling a publisher’s inventory on exchanges, a major source of their revenue will dry up, hopefully stopping many of these companies from starting at all.
A third benefit will be continued industry consolidation. Marginal exchanges will lose a source of revenue from reselling inventory (whether fraudulent or just daisy-chained from other exchanges), leading some to fold or be acquired. This will reduce the number of exchanges in the marketplace and simplify things for publishers and buyers alike.
Ads.txt also has the potential to solve some of the buy side’s challenges around supply-path optimization. The winnowing of marginal vendors from the marketplace will greatly reduce the number of redundant impressions appearing to demand-side platforms. These remaining impressions will represent the optimal paths to those publishers, doing much of DSPs’ work for them.
A final change, building on all of the above, is that low-price buyers will be forced to move their purchasing into more transparent exchanges that have greater auction density.
A number of advertisers that focus on CPM above all else have been able to snipe impressions at very low prices in second- and third-tier exchanges. They have partially been buying fraud (despite their best efforts), but partially able to get impressions from premium publishers at rock-bottom prices. The consolidation of exchanges and reduction in fraudulent impressions will force them to buy on larger exchanges with more bid density, which will drive up their prices significantly.
Most of these changes won’t start accruing until 2018 when many publishers have posted their files, and when most DSPs are not only parsing sites’ Ads.txt files but actually basing their bid activity upon them. As the year wears on, the benefits and side effects will build upon each other, and by 2018's end we will view digital media very differently from how we do today.