What Is Amazon's Soft Reserve Price? Explained
Amazon's soft reserve price quietly reshaped ad auction pricing after 2019. Here's how the mechanism works and what it means for your ad spend.

Amazon is facing a lawsuit from the FTC and 22 state attorneys general over a pricing mechanism most advertisers never knew existed. The agency claims it added up to more than $20 billion in extra ad costs. If you run Sponsored Products campaigns and your CPCs have crept toward your max bid over the past few years, this is the mechanism behind it.
Here is how to understand what changed, why it matters, and what to check in your own account.
Step 1: Understand How the Original Auction Worked
Amazon's ad auctions started as a version of the second-price auction model, the same general approach Google Search and other major ad platforms use. You set a maximum bid, the amount you're willing to pay for a click. But you typically didn't pay that full amount.
Instead, the system charged you just enough to beat the next-highest competing bid. If you bid $2.00 and the next closest advertiser bid $1.50, you'd pay something like $1.51, not your full $2.00.
Think of it like an eBay auction with a proxy bidder. Your maximum is your ceiling, not your expected payment. This structure rewarded advertisers for bidding generously since the auction, not their own number, usually set the price. That's the model advertisers expected, and for years, it held.
Step 2: Learn What Changed in 2019
In 2019, Amazon introduced something it calls a "soft reserve price." According to the FTC's complaint, Amazon set a minimum value for individual ad placements and used it as a floor beneath the traditional second-price calculation.
Here's the practical effect. Say your bid is $2.00 and the next-highest competing bid is $1.50; the classic second-price outcome would charge you $1.51. But if Amazon's soft reserve for that placement is set at $1.90, you'd instead pay $1.90, even though no competing advertiser bid anywhere near that amount.
The complaint describes Amazon internally referring to this reserve as an "invented auction participant" — essentially a phantom bidder Amazon inserted into the calculation. Amazon also used a "proxy 2nd price" to help set what you actually paid. Your bid still acted as a hard ceiling; you never paid more than $2.00 in this example. But the floor beneath your final price moved up without a real competitor pushing it there.
Step 3: See Why the Percentage Shift Matters
The most telling data point in the FTC's case isn't a dollar figure. It's a percentage.
According to the complaint, Sponsored Products advertisers paid their full bid roughly 30% to 40% of the time in 2021. By 2022, that jumped to about 70%. By 2024, it reached roughly 80%. We cover related ground in see also: why search volume screens out your best content wins.
That trend line shows the soft reserve mechanism doing a lot more work over time. When you pay your max bid only a third of the time, your bid functions as the ceiling most advertisers assume it to be. When you pay your max bid four out of five times, your bid has effectively become your price. That's a fundamentally different auction, even if Amazon never called it that publicly.
Step 4: Compare Amazon's Defense to the FTC's Argument
Amazon doesn't deny using soft reserve prices. It disputes how the FTC characterizes them and whether they hurt advertisers.
Amazon's explanation: its auctions weigh both your bid and predicted relevance for the placement. If your winning bid clears both the hard reserve and the soft reserve, you pay the soft reserve amount. If your bid clears the hard reserve but falls short of the soft reserve, you pay your own bid instead. Amazon says you never pay more than your stated maximum in any scenario.
Amazon also points to performance metrics as evidence advertisers came out ahead. It says average winning bids for Sponsored Products search ads fell 50% between 2019 and 2025, and average CPC stayed roughly flat from 2019 through 2024 once adjusted for inflation. Conversion rates, it says, rose 24% between 2021 and 2025. The company estimates its relevance-based model saved advertisers more than $8 billion between 2021 and 2025. For more on this, see the merchant center feed checklist for ai mode ads: the details.
The FTC isn't arguing advertisers paid above their stated maximums. Nobody disputes that ceiling held. The agency's argument is about expectations: if advertisers believed competition alone set their final price, they likely set that maximum bid differently than they would have if they knew a hidden floor could push their cost close to it far more often.
Step 5: Audit Your Own Bidding Strategy
Regardless of how the lawsuit resolves, you can act now. Pull your Sponsored Products search term reports for the last 12 to 24 months. Compare your average CPC to your max bid on your top campaigns.
Also read: also worth reading: how to rewrite old content so it ranks in ai search
If that gap has narrowed significantly — especially if your CPC now sits within 10% to 15% of your bid on a majority of placements — you're likely experiencing the soft reserve effect directly. This matters because it changes how you should think about bid strategy going forward.
Under the old assumption, bidding high felt low-risk since competition usually set your actual price below your ceiling. Under the current reality, where 70% to 80% of wins reportedly charge the full bid, setting an aggressive maximum is closer to setting your actual spend. Treat your max bid as a real cost commitment, not a safety margin.
Step 6: Adjust Campaigns With This Reality in Mind
Run smaller, more frequent bid tests instead of large upward adjustments. Since your bid now more directly determines your cost, incremental changes let you find efficient CPCs without overcommitting budget on a placement where the reserve, not competition, sets the price.
Watch relevance signals closely too. Amazon says its model factors in predicted relevance alongside bid amount, so improving your listing quality, keyword targeting, and click-through rate may do more to lower effective costs than raising your bid ever will.
Finally, keep historical performance data on hand. If your account shows a sharp jump in full-bid-rate starting around 2022 — matching the FTC's cited timeline — that's useful context for budget planning and for understanding why past campaign benchmarks may no longer apply.
What to Expect Next
The lawsuit is in its early stages, and Amazon has signaled it will contest both the FTC's characterization of the mechanism and its damage estimate. Expect continued back-and-forth over methodology, particularly around how to define "harm" when no advertiser paid above their own stated maximum. In the meantime, treat your bid strategy as if the soft reserve is fully in effect, because the data suggests it already is.
Related Articles

YouTube Music Tests AI Hosts: What This Means for Business
YouTube Music is testing AI hosts that will interrupt your tunes, raising questions about user engagement and the future of music streaming.
Sep 28, 2025

Motorola's Ultra-Thin Answer to Apple and Samsung: Battery Power Unleashed
Motorola's Edge 70 aims to disrupt the smartphone market with its ultra-thin design and superior battery life, challenging Apple and Samsung's dominance.
Oct 10, 2025

Meta Ray-Ban Display Glasses Leak: A Game Changer in Smart Tech
Meta's new Ray-Ban Display glasses are set to revolutionize the smart glasses market with an impressive heads-up display. Learn what this means for the future of tech.
Sep 16, 2025