Field Notes · Measurement

Amazon wrote 16 benchmark tips. None says spend less.

Amazon Ads explains how its benchmarks work and what to do about them. On PPC Land's count, not one of the 16 recommended actions tells you to spend less.

0 of 16Amazon benchmark actions that suggest cutting a budget or pausing a campaign
A single brass control knob on a small block, its pointer turned hard to the right
Every setting is a louder one. Art direction: Fiction Tribe, produced with AI.
· 3 min read

The short version

  • Amazon Ads publishes 8 benchmark metrics across 18 countries, each with one action for doing well and one for doing badly. That is 16 actions.
  • On PPC Land's count, none of the 16 suggests cutting a budget or pausing a campaign. Low CPC and CPM prompt a suggestion to test higher bids.
  • The benchmarks are weighted by spend. In Amazon's own example, a 10% median CTR becomes 3.00% once the big spenders count for more.
  • The benchmark is the seller's. Build a second one that is yours.

A benchmark is meant to tell you where you stand. This one also tells you what to buy next, and the answer is always more.

On 11 November Amazon Ads updated a support page called "Understanding benchmark metrics and recommended actions". PPC Land's Luis Rijo went through it the next day. The page covers 18 countries and 8 metrics, from video completion rate to cost per new-to-brand purchase. Each metric carries one suggested action when you beat the benchmark and one when you miss it.

What does Amazon tell you to do with a benchmark?

Spend, or spend differently. Beat the completion rate benchmark and the page says "Increase video ad investment." Beat the CTR benchmark and it says "Scale successful campaigns, test increasing bids." Come in below benchmark on CPC, which is good news for you, and it says "Test increasing bids to capture more premium placements." CPM gets the same treatment.

0 of 16
recommended actions that suggest cutting a budget or pausing a campaign. PPC Land, reading Amazon Ads' benchmark page, 2026

To be fair to the page, some below-benchmark advice moves money around: "Reallocate budget to better-performing products/campaigns", "Balance budget between prospecting and retargeting campaigns". That is still advice about where the money goes inside Amazon. None of it asks whether it should go there at all.

A benchmark that only ever recommends more is a sales tool with a measurement badge.

Whose median is it?

The biggest spenders'. The help page says benchmarks use "ad-spend weighted calculations" with controls so no single brand skews the result. Amazon's white paper, which PPC Land also cites, sets the controls: at least five peer brands or no benchmark shows, and no brand can supply more than 49% of the weight.

That cap still lets one brand carry nearly half the weight. The white paper's own example shows what weighting does. The unweighted median CTR is 10%. The spend-weighted median is 3.00%. Meanwhile the help page describes the median as the level where half of peer brands perform above and half below. PPC Land argues those two descriptions do not match, and on the white paper's numbers it is hard to disagree.

How far should you trust this reading?

Hold the limits in plain view. This is one publication's analysis of a support page. Amazon did not comment in the piece. Nobody has tested whether following the advice wins or loses money, so there are no results either way. Amazon's own caveat is on the page: the suggestions "serve as general guidance to help inform your strategy, but are not exhaustive."

None of that needs bad intent to matter. A platform writes the advice it is in a position to write. The problem is that these numbers flow straight into the reviews that set next year's allocation, and the review rarely says who wrote the yardstick.

What should a marketing team do about it?

Three things, before the 2027 plan is signed.

Build your own baseline. For every platform that grades itself, keep a second number from your own analytics: cost per sale, new customers, margin after media. Compare against your past and your other channels, not only against the platform's peers.

Ask how the benchmark is weighted. A spend-weighted median tells you how the biggest advertisers perform. If you are not one of them, it may describe a business that is not yours. Write the weighting method next to any benchmark that appears in a board deck.

Put "spend less here" on the page. When the data says a campaign has stopped earning, say so in the review. The platform's guidance will not. That is the job a media partner is paid to do, and it is the recommendation most likely to be missing.

Amazon is entitled to tell you how to spend on Amazon. You are entitled to a second opinion, and the only one that counts is the one you wrote.

Questions people ask

Do Amazon Ads benchmarks ever recommend spending less?

Not on PPC Land's reading. Amazon's support page, updated 11 November 2026, pairs 8 benchmark metrics with 16 recommended actions, and none suggests cutting a budget or pausing a campaign. Below-benchmark CPC and CPM prompt a suggestion to test higher bids.

How are Amazon Ads benchmarks calculated?

They are weighted by ad spend. Amazon's white paper says at least five peer brands are needed for a benchmark to show and no single brand can contribute more than 49% of the weight. In its own example, an unweighted median CTR of 10% becomes 3.00% once weighted by spend.

Should you use platform benchmarks in a performance review?

Only beside a baseline you built yourself. Amazon calls its suggestions general guidance that is not exhaustive, and PPC Land's analysis is one publication's reading with no test results either way. Check how a benchmark is weighted before it decides next year's budget.

Sources

  1. Amazon's 16 benchmark recommendations never suggest cutting budgets, PPC Land, . Luis Rijo's reading of the Amazon Ads support page updated 11 November 2026 and Amazon's benchmark white paper. Amazon did not comment.

Finished work in two weeks, at a fixed price.

A sprint ships a messaging platform, a brand system or a campaign in fourteen days. You see the scope and the number before we start.

See how sprints run

Keep reading

Positioning

Only 14% of buyers will pay for the difference you think you have.

Research by CEB, Google and Motista found buyers barely tell leading brands apart on business value, and only 14% would pay a premium for the difference. A launch built on features joins the 86%.

· 3 min read
14%of B2B decision makers will pay a premium for a brand's differentiation
Measurement

A ChatGPT ad click returns 89 cents. Search returns $2.45.

Two academics measured what a paid ChatGPT visit returns. It came to roughly a third of a search visit, at a price about 6.6 times what the return supports.

· 4 min read
$0.89returned per paid ChatGPT visit, against $2.45 for paid search
Every note