Amazon Seller Benchmarks: CPC, ACOS, Fees, and How to Read Them
A seller benchmark is a reference value (an average CPC, a typical ACOS, a published fee rate) that tells you whether your own numbers are normal, exceptional, or a problem. Benchmarks are only useful when they are fresh, sourced, and matched to your segment; a stale or blended figure compared against your specific data produces confident wrong conclusions. This page explains the discipline behind the briefing's Key Metrics Snapshot (only data points fresh within 7 days, always with source and context, never rolling averages presented as news) and how to use any benchmark without misreading it.
Why the snapshot only carries 7-day-fresh data
Advertising auctions reprice continuously and fee schedules change on hard dates, so a benchmark's value decays fast. A CPC average from three months ago describes a different auction: different competitor budgets, different seasonality, possibly a different event calendar. The Key Metrics Snapshot therefore applies three filters: the data point must be fresh within 7 days, it must carry a named source, and it must carry context (what segment it covers and what changed). A rolling 90-day average republished weekly fails the test: it is not news, and treating it as news is how sellers "react" to movements that happened a quarter ago.
The benchmarks that matter for daily decisions
Four families of numbers cover most day-to-day decisions:
- Average CPC by category: the price of traffic. It sets what a click costs before you have sold anything, and its trend tells you whether rising ad spend is your problem or everyone's.
- Typical ACOS by category: the efficiency yardstick. Compared against your break-even ACOS, it tells you whether your advertising is subsidizing growth or consuming margin.
- FBA fee baseline: the per-unit fulfillment and referral structure. It anchors unit economics, and knowing the baseline is what lets you spot a fee change's impact instantly.
- Monthly storage rates: the cost of holding inventory, with the off-peak versus October-December split driving Q4 stocking decisions more than any other single number.
| Metric | Illustrative range | Varies with |
|---|---|---|
| Average CPC | $0.80-$1.60 | Category, keyword intent, match type, season |
| Typical ACOS | 20-35% | Category, product maturity, campaign goal |
| Referral fee | 8-15% of sale price | Category |
| Monthly storage, off-peak | $0.78 per cubic foot | Size tier, month |
| Monthly storage, October-December | $2.40+ per cubic foot | Size tier, peak surcharge |
How to use a benchmark without misreading it
Every benchmark misuse follows the same pattern: comparing two numbers that measure different things. The method is four checks, in order:
- Match the segment. Confirm the benchmark covers your category, marketplace, and campaign type. A "beauty category CPC" is meaningless to a garden-tools seller, and a US figure is meaningless in the German marketplace.
- Check the date. Find when the data was collected, not when the article was published. Anything older than a few weeks, or spanning an event like Prime Day, describes a different market.
- Compare like for like. Your exact-match CPC on high-intent keywords versus a blended category average is apples-to-oranges: the blend includes cheap broad-match and discovery traffic that pulls the average down. Decompose your own number to the same definition the benchmark uses before concluding anything.
- Act only on sustained deltas. A single week's gap is noise: auctions are volatile and your own conversion data is lumpy. A gap that persists 3-4 weeks in the same direction, after the first three checks pass, is a real signal worth a diagnosis and a response.
Worked example: diagnosing a 42% ACOS against a 25% benchmark
Illustrative numbers. A seller's campaign runs a 42% ACOS while the category-typical figure is 25%, and they want to know whether bids are the problem:
- Decompose ACOS: ACOS = CPC ÷ (conversion rate × average order value). The seller's numbers: CPC $1.18, conversion rate 10%, AOV $28; cost per order $11.80, and $11.80 ÷ $28 = 42% ACOS.
- Compare the CPC first: $1.18 sits inside the illustrative $0.80-$1.60 category range, so traffic is not overpriced. Cutting bids would shrink the campaign without fixing the ratio.
- Back out the conversion rate the benchmark implies: at the same CPC and AOV, a 25% ACOS requires a cost per order of $7.00, which at $1.18 per click means converting about 17% of clicks, versus the seller's 10%.
- Diagnosis: the gap is conversion, not click price. The fixes are listing-side (images, price position, reviews, keyword relevance), not bid-side.
- The misread this method prevents: a seller who saw "42% vs 25%" and slashed bids 40% would have kept the same weak conversion rate on less traffic, lowering sales while the ACOS barely moved.
Frequently asked questions
What is a good ACOS on Amazon?
There is no universal number: a good ACOS is one below your break-even ACOS, which equals your pre-advertising profit margin. A product with a 30% margin breaks even at 30% ACOS, so 20-25% is profitable growth spend while the same 25% would lose money on a 20%-margin product. Category-typical ranges of 20-35% are context, not targets.
Why is my Amazon CPC different from the benchmark?
Published CPC benchmarks are blends across categories, match types, placements, and keyword intent, while your CPC reflects your specific keywords and targeting. Exact-match bids on high-intent purchase keywords routinely run well above a blended category average, and broad-match discovery traffic runs below it. A gap versus the benchmark usually describes your keyword mix, not a pricing problem.
Where does fresh Amazon benchmark data come from?
The most reliable sources are your own account reports, Amazon's published fee schedules in Seller Central, and aggregated datasets released by large advertising and analytics platforms that see many accounts. Anything without a stated source, date, and segment definition is an anecdote. The briefing's Key Metrics Snapshot only carries figures fresh within 7 days, with the source named.
How often do Amazon seller benchmarks update?
Fee schedules change on announced effective dates, typically annually with mid-year additions, while advertising metrics such as CPC and ACOS drift continuously and swing hard around events like Prime Day and Q4. A CPC figure more than a few weeks old describes a different auction than today's. Check the publication date before comparing anything against your own numbers.
Should I change my bids based on a benchmark?
Not directly. Benchmarks are for diagnosis, telling you whether a change in your numbers is specific to you or market-wide, while bid decisions should come from your own conversion data and unit economics. Act on sustained deltas in your own metrics, using the benchmark only to decide whether the cause is internal or external.
Sources
- Amazon Seller Central: published fee schedules, storage rates, and your own advertising and business reports (primary source for every figure that affects your account).
- Aggregated advertising benchmark reports from major Amazon PPC platforms (segment definitions and collection dates vary by publisher; read the methodology before comparing).
- Seller Signal Key Metrics Snapshot methodology; all ranges and worked-example figures on this page are illustrative, not current market data.