How to rank for keywords on Amazon: one keyword, six steps, and how to measure it

Ranking for a keyword means appearing high in the results a shopper sees after typing that search, in the order Amazon calls Featured. Amazon does not publish how it orders those results. It does publish enough to work from: what it takes to be in the results at all, and some of the signals that move a product up.

To be in the results, the listing has to match. Amazon's search glossary says search does not do partial matching: a product appears for a query only when its catalog data holds every word of it.

To move up, the product has to perform. Amazon's KDP help names sales history and Best Sellers Rank alongside relevant keywords, and its seller blog says competitive pricing, listing quality and account health can contribute.

The keyword decides how hard the rest will be. Amazon's keyword research blog describes the trade: short tail keywords such as "shoes" are searched often but are competitive, while long tail keywords of three or more words are searched less, are less competitive, and so are easier to rank higher for.

The customers using them know what they want and are more likely to buy.

For a new listing, that points one way. Aim first at specific phrases the product matches exactly, win those, and let the sales history they build support the broader terms later. Amazon's KDP help says the same for books: sales history and Best Sellers Rank work alongside keywords.

A listing cannot rank for a keyword if Amazon will not show it for any keyword. Amazon's help page on unsearchable ASINs lists five causes: a parent ASIN, no browse node, no buyable offer, a future launch date, and listing quality suppression. Rule them out before touching the copy.

Then check that the product can match the specific search. Amazon's glossary explains that some queries carry constraints: a search for red dresses only returns products on a dress browse node with a red color refinement. If your keyword names a color, size, material or use, the matching attribute has to be filled in, not just mentioned in the title.

With the listing eligible, put the keyword where it counts. The title comes first: Amazon's seller blog calls it the place for primary keywords, and it is what shoppers see in results. For a new listing it must stay within 75 characters, and no word may appear more than twice, brand names included.

Every word of the target keyword must appear somewhere in the catalog data, because matching needs all of them. The words do not all have to sit together in the title, but putting the core phrase there keeps it where both shoppers and search see it first.

Before building a listing around a keyword, find out whether it converts. Sponsored Products manual keyword targeting lets you bid on the exact keyword and see what happens when shoppers who typed it see your product. Amazon Ads says exact match is the most restrictive match type and tends to produce the highest conversion rates.

Amazon Ads' own advice is to test all three match types with different bids: the highest bid on exact, lower on phrase, lowest on broad, for at least one to two weeks before pausing anything. It adds that you do not bid against yourself by adding the same keyword in several match types.

Once the product listing appears for the keyword, customers decide the rest. Amazon's Search Query Performance dashboard measures the stages for every query: impressions, clicks, cart adds and purchases. Each stage has its own levers, and a weak one caps the rank a keyword can reach however well it is placed.

Third party rank trackers estimate your position by checking search results from outside. Amazon's own data is better where you can reach it. Search Query Performance, in Brand Analytics for brand owners, shows impressions, clicks, cart adds and purchases for each query tied to your brand, and a Query Detail view that compares your ASIN with the top 10 for that search.

Much of the advice on how to rank for keywords on Amazon is built around the ranking algorithm, often called A9 or A10, with lists of exact factors and weights. Amazon's own seller documentation does not name the algorithm or publish its weights. Treat any precise list as a vendor's model of Amazon search, useful for ideas but not a rule.

What Amazon does say is consistent across its pages: products must be relevant to the customer search, and sales history, Best Sellers Rank, competitive pricing, listing quality and account health play a part. A keyword strategy built on those documented points holds up when a tool's model turns out to be wrong.

Keyword tools are the same story. Amazon's keyword research blog names Helium 10 and Jungle Scout among the tools sellers use for search volume, trends and competition, alongside Amazon's own Product Opportunity Explorer and Brand Analytics. Search volume from outside tools is an estimate; Amazon's dashboards show real customer searches.

Books follow their own keyword rules. KDP's help page says relevant keywords, along with sales history and Best Sellers Rank, can boost placement in search results. Authors get up to seven keywords or short phrases, entered under Keywords in Edit book details, and can change them as often as they like.

KDP's advice mirrors the product rules: combine words in the order readers search, test each keyword by searching Amazon before publishing, and think like a reader. Useful types include setting, character types and roles, plot themes and story tone.