How Amazon search actually works
A buyer types a phrase; Amazon shows listings it judges both relevant to that phrase and likely to sell. Two things have to be true for you to benefit. First you must be indexed — the term appears somewhere Amazon reads (title, bullets, description, or the backend search-terms field). Then you have to rank, which is earned through relevance, clicks and sales on that term. Keyword research is about choosing which terms are worth earning.
Three sources of keywords
Discovery
Start broad: pull the keywords in your category with their search volume and relevance. This maps the territory — the terms buyers actually type when they want your kind of product.
Reverse-ASIN
Point a lookup at a competitor's listing and see every keyword it ranks for. This is the richest source there is: if a competitor is selling, the market has already validated those terms for you.
Gap analysis
Line your keyword ranks up against a competitor's. The terms where they rank and you don't are your open opportunities — demand you could capture that you're currently invisible for.
Relevance and winnability beat raw volume
The instinct is to chase the biggest search numbers. Resist it. Head terms carry the most volume and the most competition, so a fresh listing burns money fighting for them. Specific long-tail phrases — “bamboo cutting board with juice groove” rather than “cutting board” — bring fewer searches each but far less competition and buyers who know exactly what they want. Win a cluster of those, build velocity, and the head terms come within reach.
Put the words where Amazon reads them
Once you've chosen your terms, they have to land in the listing — the title and bullets for the important ones, the backend search-terms fieldfor the rest. That field is byte-limited and rewards no repetition, so dedupe at the word level and don't waste bytes. Then confirm Amazon actually indexed you and watch your rank move; putting a word in your copy is the start of ranking, not the end. The guides library covers the listing-side mechanics in more depth.
Do it systematically
Discovery, reverse-ASIN, gap analysis, index checking and daily rank tracking are one connected workflow: find candidate terms, see what competitors already rank for, spot the gaps, place the words, confirm you're indexed, and watch the positions. A research suite runs the whole loop so you're targeting terms you can actually win instead of guessing.
FAQ
How does Amazon decide which listings to show for a search?+
When a buyer types a phrase, Amazon returns listings it considers both relevant to the term and likely to convert. To appear at all, your listing has to be indexed for the keyword (the term is in your title, bullets, description or backend search terms); to rank well, it also has to earn sales and clicks on that term.
Should I target the highest-volume keywords?+
Usually not first. The biggest 'head' terms are also the most competitive, so a new listing rarely wins them. More specific long-tail phrases have less volume each but far less competition and higher purchase intent — they're easier to rank and convert better. Build on those, then push toward the head terms as you gain velocity.
What is reverse-ASIN research?+
Looking up the keywords a specific competitor listing ranks for, by its ASIN. It's the fastest way to build a keyword list grounded in reality rather than guesswork — you're borrowing the research a proven seller already did.
I added a keyword but I'm still not showing up. Why?+
Adding a term to your listing doesn't guarantee Amazon has indexed you for it, and being indexed doesn't mean you rank on page one. Verify indexing with an index checker, then track your rank on the term over time — ranking is earned through relevance and sales, not just by putting the word in your copy.