How Native Ads Are Priced: Auctions & Floors

Native ads are priced through a real-time auction: every time a placement is about to load, the network runs a bid, combined with a predicted CTR (and often a predicted conversion likelihood), through an internal ranking formula, and the winner's price is typically set just above whatever the next-highest-ranked bidder would have paid rather than the literal bid amount. Floor prices set a minimum, and from there your effective CPC moves constantly based on who else is competing for that exact impression at that exact moment.

The auction mechanics behind every native impression#

Most native networks run some variant of a second-price auction layered with a quality or relevance multiplier: your bid gets multiplied by a predicted-CTR score before it's ranked against competing bids, so a lower bid with a strong, well-targeted creative can beat a higher bid attached to a weak one. This is the same core logic that underpins OpenRTB-based programmatic auctions generally, adapted to the native placement's specific inventory.

What you pay isn't necessarily your max bid. In a second-price structure, you pay roughly what it took to beat the next competitor, plus a small increment, which is why your effective CPC often sits meaningfully below your set bid ceiling when competition is thin, and creeps toward it when competition tightens.

This is worth internalizing because it changes how you should think about setting a bid ceiling in the first place. A common mistake is setting the bid at exactly what you're willing to pay, treating it as if it were a fixed price. In a second-price auction, setting your bid at your true maximum willingness to pay is actually the mathematically correct move, since the mechanism itself protects you from overpaying beyond what competition requires. The risk isn't setting your ceiling too high; it's setting it too low and losing auctions you would have profitably won.

Floor prices and reserve bids#

Every network and, on some platforms, every individual publisher or placement sets a floor price below which no bid clears, regardless of predicted CTR. Floors exist to protect publisher revenue on premium placements and tend to be higher on well-trafficked, brand-safe sites than on long-tail inventory. This is one reason the same creative can show wildly different minimum viable bids depending on where the network chooses to place it.

You generally can't see a specific placement's floor directly; you infer it by testing bid levels and watching where delivery starts to appear. Bidding at or just above a network's suggested minimum is a reasonable starting point, but suggested minimums tend to be set conservatively in the network's own interest, so treat them as a floor, not a target.

Floors also aren't static across a placement's lifetime. A publisher renegotiating its deal with a network, a shift in that publisher's traffic quality, or a broader repricing pass by the network can all move a floor without any public announcement. If a campaign that used to deliver comfortably at a given bid suddenly stops clearing on a specific placement, a quiet floor increase is a more likely explanation than a sudden change in your own creative's quality score.

Why your CPC moves hour to hour#

Auction dynamics mean price is never static. A few forces move it constantly:

• Competing advertiser volume. More advertisers bidding in your vertical and geo at a given hour pushes the clearing price up; retail-heavy verticals visibly spike around major shopping periods.
• Time of day and day of week. Evening leisure-browsing hours often carry more competition on consumer verticals than midday, and weekday versus weekend patterns differ by vertical.
• Device mix. Mobile and desktop inventory clear at different price points because CTR and conversion behavior differ by device, and advertisers bid accordingly.
• Your own creative's predicted CTR. As your creative accumulates a performance history, the network's prediction of its CTR updates, which changes your effective rank in the auction even at an unchanged bid.
• Seasonal and vertical-specific demand. Insurance and finance advertisers, two of the largest verticals by creative volume in the OpenAdLibrary index, compete hardest around enrollment and tax-season windows.

Quality score and CTR multipliers#

The predicted-CTR multiplier is the single biggest lever you actually control, since your bid is a fixed number you set but your creative's predicted performance compounds over the campaign's life. A creative that consistently earns real clicks gets a rising CTR prediction, which lowers the effective bid needed to win the same auction over time. This is the mechanical reason why refreshing creative before it fatigues protects your CPC, not just your CTR: a fatigued creative's falling real-world CTR eventually drags its predicted-CTR multiplier down too, and your effective cost per click rises even though your bid hasn't changed.

Geo, device and vertical pricing differences#

Tier-1 geos (the US, UK, Canada, Australia) generally clear at higher CPCs than tier-2 or tier-3 markets, reflecting both higher advertiser demand and generally higher payouts on the affiliate and ecommerce offers running there. Within a single geo, mobile inventory is usually cheaper per click than desktop on Taboola and Outbrain, though conversion rate by device varies enough by vertical that a cheaper mobile click isn't automatically a better-value one. Reported CPC ranges by network vary widely enough across these dimensions that a single blended average tells you very little about what you'll actually pay in your specific geo, device and vertical combination, which is also why budgeting for a native campaign has to start from your own funnel math rather than a published rate card.

What you can and can't control#

You control your bid, your targeting scope, and your creative, and creative is the one that compounds. You don't control the floor price, the competing advertiser volume at any given hour, or exactly how the network's internal ranking formula weights CTR against bid, since none of the major networks publish that formula in full.

The most reliable lever available to a buyer is watching what's actually clearing in your vertical rather than guessing from a suggested bid range. OpenAdLibrary's ad intelligence tools surface which creatives are running longest in a given network and vertical, which is a workable proxy for figuring out where the real price-clearing activity is happening before you commit budget to finding out the hard way.

How this differs from search and social auctions#

Media buyers coming from search or social sometimes assume native pricing works the same way, and mostly it does at the mechanical level: an auction, a quality multiplier, a clearing price. The practical difference is transparency. Search platforms expose keyword-level bid estimates and impression share data in enough detail that you can reverse-engineer a rough sense of the competitive landscape. Native networks generally expose far less: no keyword-equivalent signal to bid against, thinner reporting on why delivery moved, and a predicted-CTR mechanism that's effectively a black box from the advertiser side.

That opacity is exactly why watching the actual creatives running in a vertical, rather than the network's own reporting dashboards, tends to be the more reliable read on where real competitive pricing sits. A creative that's been live and running for weeks in your target vertical has, by definition, cleared whatever the current floor and competitive price level is for that inventory. A dashboard's suggested bid range hasn't necessarily kept pace with that.

Practical takeaways for setting a starting bid#

• Set your ceiling near your true maximum willingness to pay, since the second-price mechanism protects you from overpaying beyond what competition demands.
• Expect floors and clearing prices to shift without notice, and don't assume a sudden delivery drop is automatically your creative's fault.
• Weight geo and device separately, since a single blended bid across a broad targeting scope will overpay in cheap segments and underbid in expensive ones.
• Refresh creative before predicted CTR decays, since that multiplier compounds into your effective price on every single auction, not just the ones where fatigue is visible in reporting.