A federal complaint says Amazon secretly raised ad-auction prices. Amazon says the auction delivers better results and advertisers knew a bid was the most they could pay. Your dashboard is about to become an argument.
Do not let it.
On August 31, the Federal Trade Commission and 22 states sued Amazon over pricing in Sponsored Products, Sponsored Brands, and Sponsored Display auctions. The allegations are serious. They are still allegations, and Amazon published a detailed rebuttal the same day.
An advertiser has a narrower job this week: preserve the data, document how the account bids, and learn which cost changes can actually be explained.
A lawsuit is not a bid strategy. A platform rebuttal is not an audit.
What the two sides say
The 181-page complaint alleges that Amazon added undisclosed reserve prices after running auctions that advertisers understood as generalized second-price auctions. The complaint says those reserves could push the winning advertiser’s cost closer to its own maximum bid, while aggregate reports kept advertisers from inspecting individual auction prices.
Amazon says advertisers set a maximum, relevance helps select the winning ad, and a winner never pays above its bid. Amazon also says reserve prices are common, its current help materials explain them, inflation-adjusted average CPC remained flat from 2019 through 2024, and Sponsored Products advertisers received better conversion rates and return on ad spend as relevance models improved.
Those accounts disagree about disclosure, causation, and harm. The case will take time. Your next budget meeting probably will not.
Build one auction truth sheet
Start with a one-page record for each material campaign. Keep these fields together:
- What can set the price? Record the ad product, bid strategy, base bid, placement adjustments, dynamic bidding setting, budget rule, and any automation that can change them.
- What did the platform charge? Save spend, clicks, average CPC, and available keyword, target, placement, and hourly detail.
- What did the business receive? Track orders, revenue, margin, new customers, returns, cancellations, and another downstream outcome that matters.
- What changed nearby? Log price, promotion, inventory, Buy Box status, ratings, creative, competition, seasonality, and landing-page changes.
- What can you reproduce? Name the report, date range, time zone, attribution window, currency, filters, and person who pulled it.
- What remains unknowable? Write down any auction-level field you cannot inspect. Do not let an average pretend to answer it.
Amazon points advertisers to real-time CPC, return-on-ad-spend, purchase, keyword, placement, and hourly data, including Amazon Marketing Stream. Pull what your account can access. The FTC complaint says those reports still do not reveal the individual auction calculation. Both facts can be true at once.
Run the peak-day comparison carefully
The complaint alleges larger price increases around high-volume shopping days. That makes Prime Day and Black Friday obvious audit targets. It does not make a simple before-and-after chart honest.
For each peak period, compare matched windows:
- the same campaign, product group, keyword or target, placement, and bid strategy;
- base bids and placement adjustments before, during, and after the event;
- average CPC, click volume, conversion rate, revenue, margin, and return rate;
- price, coupon, inventory, Buy Box, rating, and competitor changes;
- hourly or daily patterns, not one blended event average.
Competition, conversion probability, inventory, promotions, and your own bid automation can all move during a peak event. Label them. A suspicious line is a reason to investigate. It is not a homemade damages calculation.
Do not let ROAS close the argument
Amazon’s response leans on results: better relevance, more sales, stronger conversion rates, and improved return on ad spend. Those numbers matter to a buyer. They do not answer every pricing-transparency question raised in the complaint.
The reverse is also true. A dispute over auction disclosure does not prove that every campaign performed poorly or that every CPC was too high.
Keep three questions separate:
- Performance: Did the campaign produce enough profitable business?
- Price formation: What rules and controls determined the charge?
- Disclosure: Did the buyer receive an accurate explanation and enough data to make an informed bid?
Our marketing causality check handles the first question. The truth sheet keeps the other two from getting buried under one cheerful ratio.
Ask your agency or media owner six direct questions
- Which Amazon ad products and bidding modes do we use?
- Where are maximum bids, placement adjustments, and automated changes recorded?
- How far back can we retrieve keyword, target, placement, and hourly performance?
- Which peak-event windows can we compare without mixing unrelated changes?
- What data can we export now and preserve in its original form?
- Which conclusion would require legal, forensic, or auction-level evidence we do not have?
If the answer is a slide with three blended averages, ask again.
Use a restrained decision ladder
- Preserve: Export account settings, reports, invoices, change history, and agency documentation before retention windows close.
- Explain: Reconcile CPC and performance changes with bids, placements, promotions, inventory, and automation.
- Test: Lower or isolate bids in a bounded campaign when the business can tolerate the learning cost. Record what changes.
- Escalate: Send unresolved pricing questions to the platform, agency, procurement owner, or counsel with dates and report evidence attached.
- Decide: Hold, narrow, expand, or move budget based on profitable outcomes, control quality, and confidence in the evidence.
Do not promise refunds, estimate legal claims, or tell leadership the lawsuit proved your favorite theory. For rights, deadlines, or potential recovery, talk to qualified counsel.
For the media team, the useful move is less dramatic. Save the account as it existed. Build the truth sheet. Separate what happened from what the dashboard merely suggests.
If automated settings are part of the mess, pair this with our paid-media field-test plan. If campaign results are arriving under ten different names, start with the UTM system. Clean evidence is dull right up until everybody needs it.
Can your media report explain the charge?
SigServe can help audit the settings, evidence, and business outcomes behind paid-media spend before the next budget decision gets made on a blended average.