Why do advertisers blacklist zones even when the traffic is good?

Because the blacklist is the only tool most platforms give them for a decision that isn't always about quality. When a zone brings leads but the cost per lead doesn't fit the campaign economics, the advertiser has two options: keep buying at a loss, or stop buying entirely. There is no middle option built into the platform. So they blacklist the zone, not because the traffic is worthless, but because at the current campaign bid it isn't profitable enough to keep. The zone is good. The price is wrong. And without a way to change the price at the zone level, the blacklist becomes the default answer to a problem it wasn't really designed to solve.

What is the difference between blacklisting a zone and setting a lower bid for it?

Blacklisting removes the zone from the campaign entirely — no impressions, no spend, no data going forward. Setting a dedicated lower bid keeps the zone active at a price the advertiser can make work. The traffic keeps flowing, the advertiser gets the ROI they need, and the inventory keeps earning for the network and the publisher. The practical difference is between a permanent stop and a pricing adjustment. One closes the door on a zone that might have been profitable under different conditions. The other gives it a chance to keep running on terms that actually work.

How does per-zone bid control affect ad network revenue?

In two ways, and the more significant one is not the obvious one. The direct effect is that zones which would have been blacklisted stay in auction at a lower bid — the network earns less per impression on those zones, but earns something rather than nothing. The less obvious effect is on campaign lifetime. When advertisers have a way to tune individual zones rather than block them, campaigns stop hitting dead ends. They keep running because the advertiser keeps finding configurations that work. Longer campaigns generate more revenue over time than campaigns that stall and churn. In the data from this case, the lifetime effect accounted for roughly 4% incremental revenue impact against 0.7% from the direct bid efficiency gains — the compounding effect of campaigns staying alive was the dominant driver, not the margin on individual impressions.

How does custom bid for a zone affect advertiser retention?

It removes one of the most common reasons campaigns quietly stop running. When an advertiser can only blacklist a zone or keep buying it at the campaign bid, they eventually narrow their targeting down to the zones that work at that price and abandon the rest. The campaign gets smaller, performance becomes harder to maintain at scale, and at some point the advertiser pauses and moves budget elsewhere. A custom bid for a zone breaks that pattern. Instead of progressively blacklisting their way into a dead end, the advertiser adjusts the price and keeps the campaign moving. The result is campaigns that run longer, advertisers that stay active on the platform, and a retention dynamic that compounds over time rather than eroding it.

How Custom Bids per Zone Reduce Blacklisting and Grow Ad Network Revenue

The Problem Account Managers Couldn't Solve

Account managers at ad networks spend a meaningful part of their day watching campaign performance at the zone level. When a zone underperforms, the advertiser blacklists it. That's the standard move, and most of the time it's the right one — low quality traffic deserves to be cut. The AM accepts it, the campaign tightens, and everyone moves on.

But not every blacklisted zone is bad traffic. Some zones bring leads. The leads are just expensive — the zone converts, but at the campaign bid it doesn't convert enough to be profitable. The advertiser's only available action is the same one they'd take for genuinely bad traffic: block it. There's no middle option. So they block it, the AM loses that slice of inventory, and the zone goes dark.

The AM knows something the advertiser's analytics also shows: this zone works. If you go back and say "this zone is good, you should try it again," the advertiser doesn't disagree. They look at their numbers and say: "I know it converts. At this price, it doesn't convert enough." And they're right. There's nothing to argue with. The platform just didn't give them a way to act on that judgment.

What Advertisers Were Actually Doing Instead

Advertisers who wanted to keep buying from a zone at a lower price had a workaround, and it wasn't elegant. They'd create a separate campaign — a whitelisted campaign — place the zones they wanted to buy cheaper into it, and set a lower bid there. This gave them the price flexibility they needed, but it meant running two campaigns where one should have been enough.

It got more complicated from there. Because the zones in the whitelist campaign still performed differently from each other, some advertisers created several whitelist campaigns, ranging them by price. Each new campaign meant going through moderation again. Each update to the zone list meant touching multiple campaigns. The operational overhead compounded, and all of it existed for one reason: to move a bid a few cents lower on a specific zone.

All of this just to do something that should take thirty seconds. The workaround worked, technically, but it added a layer of campaign management overhead that grew with every zone and every price adjustment — and it was entirely an artifact of the platform not having the right tool, not of the problem being genuinely complex.

Meanwhile, advertisers were thinking at the zone level the entire time. They watch analytics per zone, they make decisions per zone, and they know exactly which zones they'd keep buying if the price were right. The campaign was just the container the platform forced them to work inside. The zone was always the unit that mattered.

The Fix and How It Works

The solution was straightforward to describe and not complicated to build. Instead of only being able to blacklist a zone within a campaign, advertisers can now set a dedicated bid for it. You add the zone, you set the price you want to pay, and from that point on, whenever that zone comes up in auction within that campaign, the zone-level bid takes priority over the campaign-level bid. Everything else about the campaign stays the same.

For multi-geo campaigns, you can go one level further and attach a specific geo to the zone setting, so the custom price applies only when the zone is serving users from that geography. The same zone can perform very differently across geos, so having that granularity closes the last gap in the logic.

The result is that the main campaign does what it was always doing, and the zones that needed a different price now have one — without a second campaign, without moderation, without an expanding list of whitelist variants to maintain. The advertiser acts at the level they were already thinking at.

What the Numbers Showed

We tracked results over five weeks after launch. Adoption was low — roughly 2% of campaigns used dedicated zone bids during the measurement period. That number is important context for everything that follows, because it means the impact described here came from minimal penetration. The feature was barely used, and it already moved the platform.

Those 2% of campaigns generated 3 to 5% of total platform revenue. The contribution was disproportionate from the start, which is a signal that the advertisers using the feature were running it effectively rather than experimenting with it.

EPC on campaigns using dedicated zone bids was 23% higher than on campaigns without. That result wasn't what we expected. The feature was built to solve a loss-reduction problem — give advertisers a way to lower bids on zones they'd otherwise block, reduce churn, keep inventory selling. We anticipated longer campaign lifetimes and roughly flat volumes. EPC improving wasn't part of the hypothesis.

What happened was that a subset of advertisers used the feature in the opposite direction: they identified zones that were performing well and set higher dedicated bids on them to secure more of that traffic. The feature worked symmetrically — lower bids to rescue borderline zones, higher bids to double down on strong ones — and the EPC lift came largely from that second behaviour. It contributed about 0.7% to total platform revenue, which is the smaller part of the story, but it's worth noting because it revealed a use case that wasn't in the original design thinking.

The dominant driver was campaign lifetime. Campaigns with dedicated zone bids ran significantly longer than campaigns without, and the incremental revenue from that extended lifetime was the largest single component of the overall effect — roughly 4% added to total platform revenue against the 0.7% from EPC gains. The total incremental impact came to approximately 4.7% of platform revenue.

The feature was built to reduce losses. What it also did was give advertisers a new way to double down on what was already working — and that's the part nobody anticipated.

The logic behind the lifetime effect is straightforward. When advertisers can tune individual zones rather than blacklist them, they stop hitting dead ends. They find a configuration that works and they keep the campaign running. The platform doesn't lose the campaign to a block list — it keeps it active and generating.

The Win That Goes Three Ways

Most product decisions in ad networks involve a trade-off somewhere. Something the advertiser gains usually costs the network something in margin or volume, or the other way around. This one didn't work that way.

The advertiser gets zones back into rotation that they'd previously had no good option for — zones that convert but didn't convert enough at the campaign price. Now they can buy that traffic at a price where the ROI works. Their campaigns run longer and perform better because the configuration actually fits how they think about their traffic.

The network stops losing inventory to the blacklist. Zones that had gone dark because they'd been blocked start selling again. The bid is lower, but the alternative was zero — and zero doesn't help anyone. Getting a zone back into auction at a reduced price is strictly better for the network than having it blocked.

The publisher earns from placements that had stopped delivering entirely. From their perspective, nothing changed except the placement started running again. That's the cleanest part of the outcome: inventory that was generating nothing starts generating something, and nobody had to give anything up to make it happen.

What This Means If You're Running a Network

When advertisers are blacklisting zones at scale, the instinct is to look at traffic quality — and sometimes that's the right place to look. But scale blacklisting is also a signal worth examining more carefully, because not every block is a quality decision. Some of them are pricing decisions made with the only tool available.

The question worth asking is what the advertiser is actually deciding when they hit blacklist, and whether the platform gives them a better option for the cases where price is the real issue. In most setups, it doesn't. The blacklist handles everything: bad traffic, expensive traffic, traffic that needs a different bid. They all go to the same place.

Giving advertisers bid granularity at the zone level doesn't complicate campaign management. It gives them a lever that already exists in how they think about their traffic, and lets them use it directly instead of routing around it with workaround campaigns and manual overhead.

The adoption numbers tell you what to expect at launch: a small percentage of advertisers will find it and use it immediately, and their results will be disproportionately strong. The larger opportunity is in making the feature discoverable and accessible enough that adoption grows from 2% toward something that reflects the actual size of the problem. If 2% penetration produces 4.7% revenue impact, the scaling potential is significant even under conservative assumptions.

When "This Zone Is Bad" Is Actually "This Price Is Wrong"