Benchmark browse category ID is a dimension in Amazon Ads unified reporting that records which product category a campaign is being compared against. Its value is a numeric code from Amazon's retail catalogue, the same code the store uses for a "browse node", one shelf in the tree of departments and subcategories shoppers click through. The field exists because Amazon's benchmark metrics are relative. A click-through rate of 0.4% means little until it is set against the rates of comparable brands, and the category ID tells the report reader which group of brands that comparison was drawn from.

It is one of seven attributes in the Benchmark category of the unified reporting dimensions reference that Amazon updated on November 11, 2026. The others are benchmark brand ID, benchmark brand name, browse category path, browse category level, browse category marketplace and benchmark campaign goal. According to PPC Land's count of the reference, six of the seven are "derived from the products promoted in the campaign", while the campaign goal field comes from the campaign's goal and optimisation settings.

How the field is populated

The value is not typed in by the advertiser. It is derived in three steps.

First, Amazon looks at the products a campaign promotes. Each product, identified by its ASIN (Amazon Standard Identification Number), already sits in the retail catalogue's browse tree. Amazon's Creators API documentation, which replaced the deprecated Product Advertising API 5.0 for affiliates, describes how that tree is structured: an item returns the browse nodes it "falls under and the ancestry of that browse node", with a ladder of ancestors running up to a root node. A root node has no ancestor; a leaf node has no children. A single product can sit in several nodes at once. The documentation's sample book appears under three, including node 3060, displayed as "Orphans & Foster Homes" in the US store, whose root is node 283155, "Books".

Second, a category is selected from those nodes to define the peer group. The help page on benchmark metrics, also updated on November 11, 2026, says benchmark dimensions show "the peer group for each comparison" and names "Benchmark browse category path" as an example. Amazon has not published the rule it uses to pick one node when a campaign's products span several.

Third, the selected node's numeric identifier is written into the report row as Benchmark browse category ID. The sibling fields describe the same node in other ways. The path gives its ancestry in readable form, from department down to subcategory. The level, by its name, gives its depth in the tree. The marketplace records which national store the category belongs to, which matters because Amazon runs separate catalogues for each country and the documentation's examples are tied to a specific store.

Evidence that the category ID is a browse node ID comes from Amazon's own API definitions. An endpoint described in a third-party integration guide for the Amazon Ads API, POST /benchmarks/brandsAndCategories, returns a categoryId field defined as "the same as the browse node ID", alongside brandName and categoryName. That endpoint, last updated in February 2026 according to the guide, served Sponsored Brands only, with four metrics: advertising cost of sales (ACOS), click-through rate (CTR), impressions and return on ad spend (ROAS).

Where it appears

The field lives in the unified reporting builder in the Ads Console under Measurement and Reporting, and in the reporting API, where it can be added as a column next to benchmark metrics. The dimensions reference marks the Benchmark category as applying to all five of Amazon's ad products: Sponsored Products, Sponsored Brands, Sponsored Display, Sponsored TV and DVA+, the display, video and audio campaign type launched on September 29, 2026. Where a dimension does not apply to a product, its value appears as null.

A worked example shows how the columns read together. A kitchenware brand runs a Sponsored Products campaign for a set of chef's knives. A row in a benchmark report might carry the brand's ID, a browse category ID pointing to a knives subcategory, a path running from Home & Kitchen through Kitchen & Dining to Cutlery, a level indicating that depth, the US marketplace and a campaign goal. Beside those sit the brand's own CTR and cost per click (CPC), and the peer figures. Rate metrics are shown at the 75th percentile, cost metrics at the 25th, with the median for both, according to Amazon's help page. The category ID is what allows the brand to confirm that it is being measured against cutlery sellers rather than all of Home & Kitchen.

Eight metrics are benchmarked: percentage of purchases new to brand, new-to-brand purchase rate, cost per new-to-brand purchase, CTR, CPC, video completion rate, cost per completed view and cost per thousand impressions (CPM), according to Amazon's general availability notice. Reports can be cut at daily, weekly or monthly intervals and filtered by ad format and campaign goal.

Origin and evolution

Browse nodes are as old as Amazon's catalogue. Category benchmarks inside the ad console are recent. Amazon announced self-service benchmarks on November 19, 2025, showing "brand category performance across different ad formats and campaign goals" in Campaign Manager, the report centre and the API. PPC Land covered the launch a day earlier. The beta was limited to the United States and to brand owners in Amazon's Brand Registry plus endemic advertisers, those selling on Amazon. Two report types were added to the existing reporting API: dspBenchmarks for programmatic campaigns and crossProgramBenchmarks for sponsored ads.

Unified reporting, which carries the dimension today, opened in beta on November 11, 2025 and reached general availability on June 8, 2026. On March 27, 2026, Amazon published a white paper on benchmark methodology. Its summary page credits "hybrid K-Nearest Neighbor (KNN) and Large Language Model (LLM) approaches" with finding comparable competitors and promises a "privacy-first methodology". Benchmarks reached general availability in 18 markets on May 18, 2026, from the US to Egypt. The Benchmark dimensions, with browse category ID among them, were set out field by field in the November 11, 2026 reference. None has a former name, because no legacy report carried a peer category.

Why it matters

Benchmarks change how a campaign is judged. An agency reporting a 15% rise in CTR now faces the question of whether the category rose 25%. The category ID decides who the peers are, and so decides that answer. A brand placed in a broad parent node is compared with a mixed group of sellers; one placed in a narrow leaf node is compared with close rivals, but a smaller set.

Money follows the field. Amazon's help page pairs each benchmark result with a recommended action, and none of its 16 recommendations suggests cutting a budget. A brand above the cutlery median on CTR is told to scale; one below it is told to adjust bids or creative. Which node defines the median therefore shapes where spend goes next. Sandy Welsch, executive director of global commerce at WPP Media, wrote in November 2025 that benchmarks let marketers compare results "in relation to their peers across key categories (retail, CPG, electronics, and beauty)".

Limitations and disputes

The selection rule is undisclosed. Amazon documents what the field contains, not how one category is chosen from the several nodes a product, or a campaign of many products, can occupy. A campaign promoting both knives and cutting boards could be benchmarked against either shelf.

The peer calculation is contested too. The help page describes the median as the level where "half of peer brands perform above and half perform below", yet also says the method uses "ad-spend weighted calculations". PPC Land noted that these descriptions do not match: in a spend-weighted median, a handful of large advertisers can move the line. According to PPC Land's reading of the white paper, peer sets require at least five brands and cap any one brand's influence at 49%, a cap that still allows one advertiser to supply almost half the weight.

Product coverage is inconsistent. The dimensions reference marks the Benchmark category for Sponsored Display, but the help page lists benchmark metrics only for Sponsored Products, Sponsored Brands, Streaming TV and DVA+. The same reference shows Sponsored TV markers on attributes while a footnote says Sponsored TV is not yet supported.

There is also the question of who grades the test. Amazon sets the categories, builds the peer groups, computes the percentiles and writes the recommendations, and no outside auditor verifies any of it.

Benchmark browse category ID is not...

Advertised product category. A separate dimension in the Advertised product group of unified reporting, it describes the catalogue category of the product that was advertised. It describes the product. The benchmark field describes the comparison group.

Category targeting. In Sponsored Products and Sponsored Display, advertisers can target shoppers browsing particular categories. That uses the same catalogue tree, but as an input to delivery; the benchmark field is an output of reporting.

Benchmark brand ID. Its sibling in the Benchmark group identifies the advertiser's own brand within the comparison, not the category it is compared against.

Recent developments

The field's documentation is days old. On November 11, 2026, Amazon published both the 156-attribute dimensions reference and the updated benchmark help page, and PPC Land analysed both the next day. The reporting system that carries it is about to become mandatory: legacy report pages become read-only on December 17, 2026 and are scheduled for permanent shutdown on January 31, 2027. After that, this field is the only place in Amazon's reports that names a campaign's peer category.

Timeline

  • November 11, 2025 - Amazon opens unified reporting in beta.
  • November 19, 2025 - Amazon announces self-service benchmarks in the US, with dspBenchmarks and crossProgramBenchmarks API report types.
  • November 20, 2025 - WPP Media describes its early access to Amazon benchmarks.
  • February 2026 - Benchmarks brands-and-categories endpoint, returning a category ID equal to the browse node ID, documented for Sponsored Brands.
  • March 27, 2026 - Amazon publishes its benchmarks white paper describing KNN and LLM peer matching.
  • May 18, 2026 - Benchmarks reach general availability in 18 markets with eight metrics.
  • June 8, 2026 - Unified reporting reaches general availability.
  • September 29, 2026 - Amazon DSP renamed DVA; DVA+ campaign type announced.
  • November 11, 2026 - Amazon publishes the dimensions reference listing seven Benchmark attributes, including Benchmark browse category ID, and updates its benchmark help page.
  • December 17, 2026 - Legacy report pages scheduled to become read-only.
  • January 31, 2027 - Legacy report centres scheduled for permanent shutdown.

Summary

Who: Amazon Ads populates the field for advertisers running Sponsored Products, Sponsored Brands, Sponsored Display, Sponsored TV and DVA+ campaigns; brand owners in Brand Registry, endemic advertisers and their agencies read it.

What: A unified reporting dimension holding the browse node ID of the product category used to define a campaign's benchmark peer group, derived from the products the campaign promotes.

When: Benchmarks launched in the US on November 19, 2025 and went global in 18 markets on May 18, 2026; the field was set out in Amazon's dimensions reference on November 11, 2026.

Where: In the Benchmark category of unified reporting in the Ads Console and the reporting API, alongside browse category path, level and marketplace.

Why: Benchmark metrics are only meaningful against a defined peer group, and the category ID tells the reader which group that is, although Amazon has not disclosed how the category is chosen.