How Trinity Mirror is using server-side ad trading to tap brand budgets

Six months after British newspaper group Trinity Mirror integrated a server-to-server solution — the sequel to header bidding — the publisher is getting advertisers to increase their programmatic ad spending. In one test, it reduced an advertiser’s cost per acquisitions by 75 percent and got that advertiser to increase its programmatic spending by 175-fold, according to Trinity Mirror.

The parent of titles including the Daily Mirror and the Daily Record make 60 percent of its display ad inventory available programmatically. For its test, it shifted one-fifth of that to server-side ad auctions and all but dropped header bidding entirely, keeping just Amazon’s header bidding tool. Trinity Mirror improved its ad yields by 17 percent with header bidding.

But it sees server-side’s benefits going beyond higher CPMs, to regaining control over the auction process, reducing the number of ad tech middlemen and getting brand advertisers to spend more on programmatic, which historically has been the domain of direct response.

“Publishers have drawn the short straw for the last 36 months, with the advent of programmatic as well as the duopoly of Google and Facebook,” said Amir Malik, Trinity Mirror’s programmatic chief. “But if we can drive a more intelligent and research-led sale of our media, then using server-to-server and influencing market dynamics will mean we can claw back more autonomy and control of our inventory.”

Server-to-server integration is fast becoming the obsession du jour in the same way header bidding was over the last year. Amazon has rolled out its own cloud-based solution with U.S. publishers, while AppNexus and IndexExchange have paired up to offer a solution.

Trinity Mirror has gone a different route, integrating with ad tech platform Switch, which Malik stressed would act as an independent auditing layer and won’t compete with the publisher on bids.

Using Switch, Trinity Mirror’s team can see all third-party exchanges with a single dashboard. That offers a variety of benefits. For starters, it frees up resources by consolidating five hours of reporting into one. The publisher also gets a better view of advertisers’ buying patterns. That means it can tweak the sell-side algorithms to ensure that as soon as a rival buyer to any Trinity Mirror advertiser pops up in a bid window for an impression, the publisher can manipulate the algorithm to ensure their partner’s bid always wins.

“We believe this could be very disruptive for real-time bidding,” said Malik.

The publisher tried out this method in the past three months with two advertisers, from the telco and travel sectors, across its 80 websites spanning 30 million monthly unique users (comScore).

Before the trial, one of the advertiser’s buyers spent £13 ($16) a day on Trinity Mirror inventory. The publisher then applied the “conditional algorithm” so that every time a core rival to that buyer (such as a DSP or agency trading desk) tried to bid on an impression on behalf of the same advertiser, Trinity Mirror’s own buyer partner won the impression. Based on those results, the buyer increased its daily spend for that advertiser to £2,000 ($2,500) a day and its cost per acquisition dropped 75 percent. Malik plans to expand the server-to-server effort, pitching it to advertisers as a way to reduce their cost of acquisition in exchange for higher spending and guaranteed volume

“The commercials around this are still evolving, but the immediate upside is that the volume of [advertiser] spend grows astronomically,” said Malik.

Server-to-server is touted as being able to provide the same auction possibilities as header bidding, minus the risk of page latencies. But the method has its drawbacks. The main one is user matching being poor on the server-side.

“As soon as you introduce header bidding or server-to-server, you end up with multiple bid requests for every impression, which didn’t exist with the waterfall,” said Ed Lyon, programmatic director for Group M agency Essence. “And every bid request requires work on the DSP side, which pushes up costs to the advertiser, or whoever sits between the demand-side platform and advertiser — the agency.”

But Malik stressed that won’t be an issue, saying that centralizing the view of exchanges will decrease the chances of miscommunication. “Reducing the number of hops between supply-side platforms limits the friction between our buyers and our audience, so the user matching is actually improved,” he said. “Ultimately I have decluttered the sale of my audience, and that is very threatening to the SSPs.”

On paper, the concept is appealing to performance agencies, particularly when it comes to the potential to reduce CPAs. If the test client wasn’t spending much to begin with, you’d expect to see an increase in their spending if their approach got smarter, said James Diba, display and audience planning director at Dentsu Aegis’ performance agency iProspect.

“But to increase it by that much and more than halve the CPA is very interesting, and is something we would want to get on board with,” he said.

https://digiday.com/?p=220727

More in Media

Meta AI rolls out several enhancements across apps and websites with its newest Llama 3

Meta AI, which first debuted in September, also got a number of updates including ways to search for real-time information through integrations with Google and Bing.

Walmart rolls out a self-serve, supplier-driven insights connector

The retail giant paired its insights unit Luminate with Walmart Connect to help suppliers optimize for customer consumption, just in time for the holidays, explained the company’s CRO Seth Dallaire.

Research Briefing: BuzzFeed pivots business to AI media and tech as publishers increase use of AI

In this week’s Digiday+ Research Briefing, we examine BuzzFeed’s plans to pivot the business to an AI-driven tech and media company, how marketers’ use of X and ad spending has dropped dramatically, and how agency executives are fed up with Meta’s ad platform bugs and overcharges, as seen in recent data from Digiday+ Research.