From buyer’s market to publisher leverage: what publishers actually want from bidstream classification

BY HAZEL BROADLEY, SENIOR CONTENT EDITOR AT BEELER.TECH

Every Pub.Call, we get a room of publishers together to pull apart a problem the sell-side is genuinely wrestling with. In July, that meant handing the mic to Jounce Media founder Chris Kane, who walked us through how his classifications shape what buyers see, then stuck around for the harder questions from the publishers those classifications actually affect.

Programmatic still runs as a buyer’s market: buyers hold the classification power, buyers make the go or no-go call on your site, and often you’re being judged without knowing it’s happening. Tilting that back starts with seeing what buyers see, so that’s where the session went, along with what publishers can actually do about their scores.

See your inventory from the buyer’s end of the pipe

Once a bid request leaves your page, it can end up looking nothing like the tidy setup you configured. Jounce stitches together transaction data and public disclosures (ads.txt, sellers.json and the like) to rebuild what actually turns up at the DSP: how many requests, and roughly how much spend they pull. Those labels then feed real buying decisions, so it’s worth knowing what yours say

The logic is simpler than the jargon makes it sound. If an exchange pays the primary seller (you, or whoever holds your exclusive sales rights), that’s a direct path. Add an extra party in the middle and it’s indirect, and indirect isn’t automatically bad. Plenty of middlemen hold genuine sales rights and earn their keep. What Jounce watches for is non-exclusive rebroadcasting, or in other words, the same impression bounced through extra hops, which can muddy the signals buyers rely on.

And the ‘Bellwether’ tag, for the record, isn’t a gold star for quality. It just means Jounce has manually reviewed that seller and is more confident the label is right.

Each site then rolls up to one of two verdicts buyers see: ‘buy with confidence’ or ‘buy with caution.’ A caution flag is narrower than it sounds. Rather than writing off your whole site, it points buyers toward a curated deal or a publisher-sold PMP when risk or rebroadcasting looks unusually high.

And the reason behind a flag can be frustratingly specific. One well-known site got caution mainly because more than half its requests were being rebroadcast. Another premium site got caution over a single low-viewability ad slot from a third-party vendor, sitting on the same page as its otherwise solid inventory. So the lesson for publishers here is don’t take a label at face value – chase down what’s actually driving it.

Turn buyer-side judgment into publisher-side leverage

The uncomfortable part is that most brands and agencies don’t have the tooling to cherry-pick a clean path through your site, so they make a blunt call: buy the whole thing, or block the whole thing. Some go further and treat ‘buy with confidence’ as a hard inclusion list. Jounce’s own numbers show confidence sites do monetize better, though nobody can say whether that’s cause or coincidence. The thresholds aren’t fixed either – they drift daily, somewhere in the region of 10% media risk or 23% rebroadcasting before caution kicks in.

One solution is granular buying, and a few buyers are already trying it. One major DSP lets buyers switch off rebroadcasting or cheap-reach auctions while still buying the good supply on the same site, so the risk melts away without punishing the publisher. Chris also nudges agencies to build inclusion lists of around 100 publisher brands they trust, using the Bellwether list as a starting point rather than gospel.

The pitch to buyers is straightforward: move budget out of the walled gardens and into quality open-web publishers, through the paths that leave the most money in the publisher’s pocket. It’s hard to argue with the ambition, but the question publishers kept coming back to was whether the execution helps them or hurts them.

This is where the room got lively. The worry is big publishers can become Bellwethers, and tiny ones can borrow credibility through a Bellwether sales house, but the mid-sized independents stuck in between get stranded. Chris acknowledged those publishers could potentially see demand suffer, but bolting on yet another intermediary would be unlikely to fix that.

One publisher argued that if you don’t hold a direct DSP connection, routing through a sales house or exchange is often the only practical way to activate demand at all, duplicative or not, and buyers actively avoid some of the ‘cleaner’ paths for reasons that have nothing to do with efficiency.

Stop being a price-taker: challenge your Property Score

So what can you actually do? Start by getting your Property Score, free through the Brand Safety Institute or by emailing Jounce support, and challenge anything that looks off. Keep an eye on each reseller’s share of your bidstream, too. If a partner you’d expect to sit at 1% or 2% of requests is suddenly pumping out 30%, go find out why.

You can authorize or drop sellers, but you can’t easily cap how loud any one of them gets, which is exactly the problem one publisher raised on the call: open the door to a reseller and they’ll kick it clean off the hinges. Worth saying plainly, though: ‘buy with confidence’ doesn’t mean killing all reselling. Some resold demand pays the bills. The fix is sharper buyer-side filters, not a scorched-earth ads.txt.

Two blind spots are worth holding in mind:

The Demand Score. Jounce doesn’t see Google Ads spend, so if that’s a big slice of your revenue, your score can make you look worse than you are. And with more than 80% of the spend Jounce tracks landing in the US, non-US publishers can get short-changed too.
The Property Score. Because a lot of the incoming data has no geography attached, risk from one market can bleed into your score in another, a real headache raised on the call by publishers whose rebroadcasting problems sit almost entirely outside the US. Acknowledging this limitation, Chris suggested that publisher-supplied data could help, if enough publishers played along.

The bigger worry is the stuff you never see coming, and Chris was refreshingly open about a genuine gap here. One SSP had a product that crammed three rotating impressions into the 30-second window where a standard slot shows one. Jounce was right to flag those auctions as cheap reach, but the knock-on nudged up the measured media risk for a run of publishers and tipped at least one from ‘buy with confidence’ to ‘buy with caution’ through no fault of their own.

To his credit, Chris’s team worked hand in hand with the vendor to confirm which domains had dropped the unit, then unwound the labels as fast as they could validate the changes. The call itself wasn’t the problem. The problem was the lag before affected publishers even heard about it. That’s where a community like ours earns its keep as it’s a place where changes like this can be flagged to publishers faster, before a score shifts under their feet.

Strip away the jargon and the session came down to 3 things

Publishers want:

to see their inventory the way buyers see it

a real say when a label gets it wrong

warning before a classification moves the market against them.

Tools like Jounce’s can help that along, as long as buyers pick up the controls that separate a good publisher from a bad path and stop treating one label as the whole story. That’s the exchange worth building, confidence that tracks quality supply rather than getting flattened into a single blunt label.

If you’re a publisher who wants a hand getting your own scores in front of you, or you just want to be in the room for the next Pub.Call, come find us at Beeler.Tech. And bring your tough questions. 👉Want to stay in the loop with what’s happening in the Beeler.Tech community? Subscribe to our newsletter. If you’re interested in attending or sponsoring a future event, you can explore our upcoming events here.