Read SQP as shares, not sales
Each SQP row is one search query and the whole market's funnel for it: impressions → clicks → cart adds → purchases, with a total for all products and a count for your ASIN or brand. The useful number is never your count on its own: it's your count divided by the total, your share of each step, and how your rates compare with the market's.
| Metric | How to work it out | What it tells you |
|---|---|---|
| Impression share | Impressions: ASIN ÷ Impressions: total | How often you're seen for the query (rank and ads) |
| Click share | Clicks: ASIN ÷ Clicks: total | How much of the query's traffic you win |
| Purchase share | Purchases: ASIN ÷ Purchases: total | How much of the query's sales you win |
| Relative CTR | (Clicks: ASIN ÷ Impressions: ASIN) − (Clicks: total ÷ Impressions: total) | Above 0: shoppers who see you click more than average (main image, price, reviews, title work) |
| Relative CVR | (Purchases: ASIN ÷ Clicks: ASIN) − (Purchases: total ÷ Clicks: total) | Above 0: shoppers who click you buy more than average (the offer wins) |
| Relative price | Purchases: ASIN Price (median) vs Purchases: Price (median) | Whether you're cheaper or dearer than what shoppers actually buy for this query |
The four patterns and what to do
| Pattern | What it means | What to do |
|---|---|---|
| Purchase share > click share, low click share | When shoppers click you, you beat the market, but few of them get to you. This is the opportunity most sellers miss: they push keywords that already bring traffic and never try to win more clicks where they convert best. | Grow click share: exact-match PPC with top-of-search placement, the query in the title and main image message, a coupon or deal to lift CTR, then a ranking push. Every extra click converts above market. |
| Click share > purchase share | You win the click but lose the sale: price, reviews, images or the listing don't convince for this query, or the product doesn't fit the need. | Don't buy more clicks yet. Fix the offer for this need (price vs the query's median, images, bullets, A+), or treat the query as less relevant. |
| Impression share high, click share low | You're seen but not chosen in the results. | Work on CTR: main image, price, rating and review count, title start, badges, coupon. |
| Little or no impressions on a relevant query | Amazon doesn't show you for it: weak relevance, not indexed, or rank too low. | Check indexing, add the words to the title, bullets or backend, and test with exact PPC. |
"hiking backpack", one month: 21,000 market clicks and 1,650 purchases → market CVR 7.9%.
Your ASIN: 1,050 clicks (5.0% click share) and 99 purchases (6.0% purchase share) → your CVR 9.4%, relative CVR +1.6 pts.
You convert better than the market but get only 5% of the clicks. Reaching 8% click share = 630 more clicks × 9.4% ≈ 59 more SQP purchases a month on this one query, before counting the purchases SQP doesn't see.
SQP impressions, clicks and purchases (total and ASIN) to impression, click and purchase share, relative CTR and CVR, a verdict, and the purchases a higher click share would add.
Want these numbers for your own account? Get a free audit →To do it for every query at once, paste your SQP export into the SQP share & concept workbook ↓: the Query analysis tab works out every share, relative CTR and CVR, a conversion signal and a visibility signal per query, and the extra purchases if your click share caught up with your purchase share.
What SQP can't tell you
| Limit | What it means for you |
|---|---|
| Only page-1 search traffic | Clicks from product pages, Deals, category browsing, widgets and page 2+ aren't in it, nor are product-page and audience ads. |
| Short purchase window | A purchase counts only soon after the search (about 24 hours in practice). Shoppers who compare for days and buy later are missed, so higher-priced, considered purchases look weaker than they are. |
| One purchase per order, clicked ASIN only | 5 units in one order = 1 purchase. In the ASIN view, buying a different variation than the one clicked isn't credited to the clicked ASIN: check the Brand view too. |
| Not your sales or PPC numbers | SQP purchases won't match Business Reports, and SQP isn't comparable with the ads console, which credits sales over a longer window and from other placements. |
| Small numbers are noise | A query with 30 clicks a week can swing wildly. Use the monthly view (or 4+ weeks) and judge queries with enough market clicks. |
| Shares move with competitors | Your share can drop because a competitor ran a deal or out-of-stocks came back, not because you got worse. Check what changed in the market before acting. |
How each number is built: Search Query Performance: how Amazon computes the data.
Rank-ready: which factors to work on first
Organic rank for a query comes from many factors at once. A useful way to prioritize, from what SQP can measure:
| Priority | Factor | What to do |
|---|---|---|
| 1st | Search term relevance | The title and bullets cover the queries where your conversion rate is above the market's. |
| 1st | Relative CVR | Beat the market conversion rate for the query in SQP. |
| 1st | Relative CTR | Beat the market click-through rate for the query in SQP. |
| 2nd | Recent sales velocity | Grow units sold from the query week over week over about 4 weeks: sponsored ads, deals. |
| 2nd | Consistent sales history | Keep unit sales steady for 12+ weeks: no stockouts. |
| 3rd | Price competitiveness | Compare your price changes with your conversion rate, and your price with the query's median purchase price. |
| 3rd | Review rating and volume | Compare your rating with the products ranking near you for each target query. |
| 3rd | Availability and fulfillment | Inventory runway and fast delivery to the main regions. |
| 3rd | Content quality | Clear, varied wording: synonyms and benefits instead of repeating the same words. |
| 3rd | Engagement trends | Your monthly or quarterly share of impressions, clicks, cart adds and purchases. |
Think of rank as a stack. Each product's rank is the sum of its relative strengths: rating + price + conversion rate + bullets SEO + title SEO + sales velocity, and many more factors. You don't need to beat a competitor on every factor: being much stronger on one can lift the whole stack above theirs.
Product 3 ranks below Product 1, which sells 40 units a day for the query.
A strong promotion takes Product 3 from 24 to 52 units a day. Its rating, price, CVR and SEO are unchanged, but velocity is now ahead of Product 1, and so can its rank be.
The lift lasts only while the stack stays ahead: plan how CVR, CTR and sales hold up after the promotion ends.
From keywords to shopping concepts
Queries are the words shoppers type; behind them are a smaller set of shopping concepts: the needs, attributes, uses and decision criteria that make them pick a product. "Lightweight backpack for hiking", "waterproof hiking backpack" and "comfortable backpack for long hikes" are three queries but four concepts: hiking, lightweight, waterproof, comfort. One query can carry several concepts, and one concept appears in many differently worded queries.
Backpack queries grouped into shopping concepts
Use case
hiking backpack for womenhiking
travel backpack carry ontravel
daypack for hikinghiking
Capability
lightweight backpack 30Llightweight
waterproof hiking backpackwaterproof
durable hiking backpackdurable
Specification
lightweight backpack 30L30L capacity
travel backpack carry oncarry-on size
Compatibility
backpacking pack with hydration bladderhydration bladder
backpack with rain coverrain cover
Environment / performance
backpack with rain coverrain
best backpack for long hikescomfort
| Type | Examples |
|---|---|
| Use case | hiking, travel, everyday use |
| Capability | lightweight, expandable, waterproof |
| Specification | capacity (20L / 30L / 50L), dimensions, weight |
| Compatibility | hydration bladder, laptop, rain cover, trekking poles |
| Features / design | multiple compartments, attachment points |
| Environment | rain, outdoor use, extreme conditions |
| Performance | durability, load support, comfort |
| Material / appearance | nylon, recycled materials, color, style |
Amazon increasingly organizes shopping this way: Rufus questions about compatibility, use and capability, needs-based shopping ideas, and Narrow your search filters by attribute and use case. Amazon's COSMO research describes learning these intent links from shopper behavior and using them to improve query-to-product relevance. Keywords still matter: they measure the demand and prove which needs convert. Concepts add the question of which needs your product should be found for.
Find the needs you're not capturing
- Export SQP for the ASIN
ASIN view, monthly, as many queries as possible. Paste it into the workbook ↓.
- List the needs your product genuinely meets
In the Concepts tab, one row per need with a word that signals it in a query (waterproof, rain cover, women, 30L…). Use the concept types as a checklist so you don't miss a whole dimension.
- Read each need's shares
The tab adds up every query containing the word: market clicks and purchases, your click and purchase share, relative CVR, and what share of all clicks the need represents.
- Act on the gaps
See the table below.
| What you see for a need | Likely reason | What to do |
|---|---|---|
| Real demand, almost no impressions or clicks for you | Amazon doesn't link the product to the need, or the listing never says it | Say it clearly in the title, bullets, images, A+ and backend attributes; answer the Rufus questions about it; test with exact PPC |
| Clicks but purchase share below click share | Shoppers aren't convinced you meet the need | Prove it: images, specs, comparison chart, reviews that mention it; check price against the need's median |
| Purchase share above click share | You win this need when seen | Grow click share on its queries: PPC top of search, then a ranking push |
| You don't really meet the need | The product isn't competitive for it | Don't chase it with copy: improve the product, or drop the need |
- Demand: which queries express it and how much traffic they bring
- Behavioral fit: do shoppers already click and buy you for it?
- Product truth: does the product genuinely meet the need?
- Evidence: does the detail page clearly show it (text, images, attributes)?
- Amazon's understanding: does Rufus or Narrow your search link the product to it?
Related guides
- Search Query Performance: how Amazon computes the dataWhy SQP purchases never match Business Reports: SQP counts only clicks from page 1 of search results (organic and some ads), purchases…
- Keyword research from Search Query PerformanceBuild a keyword sheet from Search Query Performance, organic rank and your search term report, and mark each keyword good ranking, target…
- Master keyword list and keyword groups for a launchFrom the SQP and PPC tabs, keep the keywords where your CVR beats the average, group them under 1–3 word root keywords with their long…
- From keywords to intent: optimizing for COSMO and RufusHow Amazon's AI (COSMO behind the scenes, Rufus in front of shoppers) matches intent instead of keywords, how to run a prompt-based Rufus…
