Where an Amazon Listing's Visibility Comes From: Search, Ads and Related-Product Placements

When a competitor outsells you, the usual explanation is "they rank better". That is half of it.

A product is seen in two places: on search results pages, and on other products' detail pages. The second is easy to forget because you never see it in a keyword tool. For some products it is most of their exposure.

Two endpoints count both. One limit first: they count placements, not visitors. They tell you where a product is shown, not how many people clicked.

Part one: search results

The traffic sources endpoint takes an ASIN and counts the keywords it appears for, split by the kind of placement:

curl --request POST \
  --url https://ecommercedataapi.com/v1/amazon/traffic/source \
  --header "Content-Type: application/json" \
  --header "X-API-Key: ${ECOMMERCE_DATA_API_KEY}" \
  --data '{
    "marketplace": "US",
    "q": "B0D5B6RLYY",
    "page": 1,
    "size": 20
  }'

We ran it on 12 October 2026 for a 7-litre dog water fountain selling at $49.99 on the US marketplace. For the main variation:

One keyword can be counted in several rows, so the rows do not add up to the total.

The shape is what you read. This product is organic on almost every keyword it appears for (1,237 of 1,306) and advertises on about one in seven. That is a listing that ranks on its own and uses ads selectively. A product with 300 organic keywords and 900 ad keywords is buying its visibility, and would lose most of it if the budget stopped.

The response returns one row per variation. In this test a second variation appeared for 516 keywords and a set of replacement filters for 371. Read the variation you are actually comparing against, not whichever comes first.

To see which keywords those are, use the reverse lookup: ASIN reverse keyword lookup.

Part two: other products' pages

The related-product statistics endpoint counts the placements a product holds on other detail pages:

curl --request POST \
  --url https://ecommercedataapi.com/v1/amazon/traffic/listing/stat \
  --header "Content-Type: application/json" \
  --header "X-API-Key: ${ECOMMERCE_DATA_API_KEY}" \
  --data '{
    "marketplace": "US",
    "asin": "B0D5B6RLYY"
  }'

For the same product it returned 1,264 placements, 1,243 free and 21 paid:

The counts add up to 1,264, and the two paid types (sp and fsa) add up to the 21 in paidRelations.

Three readings:

Almost all of it is free. 98% of these placements are ones Amazon assigns from shopper behaviour. This product is recommended on other pages because people who look at those pages also look at it.

"Viewed" far outweighs "bought together". 1,207 placements are view-based and 32 are frequently-bought-together. View-based placements put you next to substitutes: shoppers comparing alternatives. Bought-together placements put you next to complements.

Paid placement on detail pages is small here. 21 of 1,264. Sponsored placement on competitors' pages is not how this product gets seen.

Finding the actual products

To see which products those placements connect to, call the related products endpoint with the relation types you care about:

curl --request POST \
  --url https://ecommercedataapi.com/v1/amazon/traffic/listing/page \
  --header "Content-Type: application/json" \
  --header "X-API-Key: ${ECOMMERCE_DATA_API_KEY}" \
  --data '{
    "marketplace": "US",
    "asinList": ["B0D5B6RLYY"],
    "relations": ["fbt"],
    "page": 1,
    "size": 20
  }'

Each item is a related product with its price, BSR, estimated units, ratings and seller. The related product traffic reference lists all fourteen relation codes.

Frequently-bought-together is the short, high-value list. Those are the products shoppers add to the same order. They are candidates for a bundle, for a product you could add to your own range, or for a sponsored placement of your own.

What to do with the two halves

Putting this beside a competitor's price, ratings and sales is covered in Amazon competitor analysis. Where a product's price sits within its category is in reading a category's price bands.

What this cannot tell you

Traffic volume. A count of 787 placements is not 787 visits, or 787,000. It is the number of places the product was observed.

Conversion. Being shown is not being bought.

Ad spend. adKeywords counts keywords with an ad, not what was paid for them.

Anything permanent. Placements are recalculated continuously; this is one day's picture.

Questions

Can I see how much traffic a competitor gets? No. Amazon does not publish visits for other sellers' listings. You can see where a listing is shown and estimate its sales.

What is the difference between "also viewed" and "frequently bought together"? The first links products shoppers compare. The second links products shoppers buy in the same order.

How many calls does this take? One for search placements, one for the related-product counts, and one per relation type you open.

Can I run this on my own product? Yes, and you should, to compare the shape with a competitor's.