US sports publisher × Adverge · head-to-head vs. the previous partner
+22.36%
more from the same inventory, once the auction knew who it was selling to.
A US sports publisher ran Adverge against the incumbent platform it had been monetizing with — an established partner operating at industry scale.
Adverge finished 22.36% ahead.
No new placements, no redesign, no additional ad slots. The inventory was the same on both sides of the test.
The default is not the ceiling.
A large incumbent has to be right across thousands of publishers at once, so it runs on defaults: a general floor, a fixed refresh cadence, a session treated as anonymous unless proven otherwise. Those defaults are safe. They are also the reason there is room left on the table.
The previous partner asked
Who is this?
An impression on a sports page. Priced the way sports pages are generally priced.
When to refresh?
On a timer, at the same cadence for every reader and every moment of the day.
What does it know?
What it can observe from the outside, shared across every publisher on the platform.
Adverge asked
Who is this?
A known reader — logged in or returning — described to demand partners before the auction opened.
When to refresh?
When the publisher’s own engagement signals say the reader is still there and still reading.
What does it know?
What this publisher knows about this audience, which no platform-wide default can approximate.
What we actually changed.
Two changes, both drawing on data the publisher already held.
01
Better use of first-party signals
The publisher’s own audience data was synced to demand partners ahead of the auction instead of staying inside the property. The bidders did not change. What changed is that they could see who they were bidding on, and priced accordingly.
02
Refresh optimization driven by that data
Refresh was set from first-party engagement signals rather than a clock. That yields impressions from readers who are genuinely still on the page, and stops the stack manufacturing them from readers who have already gone.
The inventory was never the constraint. The information was.
Both stacks were selling the same pages to the same market in the same weeks. The 22.36% is the difference between an auction running on platform defaults and one running on what this publisher knows about its own readers.
Why this keeps happening in sports.
Sports audiences are among the most identifiable on the open web — people return for a team, a league, a beat writer, and they come back on a schedule the fixture list sets. That makes first-party signal unusually rich here, and unusually wasteful to leave unused. A platform default cannot tell a returning subscriber mid-game from a one-off search visitor at midnight. The publisher can, and that distinction is worth real money on every impression.
Figure is the publisher’s own reported uplift for Adverge against its previous monetization partner in a head-to-head test. Neither the publisher nor the incumbent platform is named here. This account describes the changes Adverge made; the daily test data sits with the publisher.
Run the same test on your traffic.
Adverge tests head-to-head against your incumbent on live inventory, measured daily. If your first-party data is sitting in the property instead of reaching the auction, a month of side-by-side data will show you what it is worth.