The Search Query Performance dashboard in Amazon Brand Analytics is the closest thing Amazon gives you to a view of the whole search funnel. For every query that led shoppers to your brand, it shows the impressions, clicks, cart adds and purchases across all of Amazon, and how many of them went to you.
The prompts below turn an SQP download into decisions: which queries you lose at which step, whether the listing, the price or delivery speed is the problem, and where ads fill a gap or pay for sales you would win anyway.
How to use them: download the report as a CSV from Brand Analytics, paste it where a prompt shows a placeholder in square brackets, and run the prompts in order the first time. They work in ChatGPT and Claude alike.
What Brand Analytics and Search Query Performance show
Brand Analytics is a set of reports for brand owners enrolled in Amazon Brand Registry. In Seller Central you also need a user role that can see it, usually the Brand Representative role or permissions granted by the primary user; vendors find the same reports in Vendor Central. Access and the list of reports vary by marketplace and change over time, so check the current Brand Analytics help page in Seller Central if a report is missing.
Three reports matter here: Search Query Performance (the funnel per query), Search Catalog Performance (the funnel per ASIN across all searches) and Top Search Terms (the most searched terms and their top clicked ASINs). SQP comes in Brand View and ASIN View, by week, month or quarter.
In ASIN View the brand columns become ASIN Count and ASIN Share %. If a prompt names a column you do not have, tell the model what you have instead.
Prompt 1 · What Brand Analytics and Search Query Performance showChatGPT Claude
Act as an Amazon Brand Analytics specialist. Below is a Search Query Performance export, Brand View, for [period] in [marketplace]. First list the columns you found and flag any missing or renamed. Then build one table, a row per query: query, Search Query Volume, my brand share of impressions, clicks, cart adds and purchases, plus my click-through rate (brand clicks / brand impressions) and my purchase rate (brand purchases / brand clicks). Sort by Search Query Score, keep one decimal place, never invent rows, and say so when a calculation divides by zero. Data: [paste SQP Brand View CSV]
Tip: Keep this output. The later prompts work better on this clean table than on the raw download.
Find where you lose the shopper in the funnel
Every query has four share numbers that should roughly line up. When one drops off, it points at a specific part of your offer. The next prompts use this logic.
Pattern on a query
Likely cause
Check first
Low impression share, healthy click and purchase share
Not enough visibility
Keyword in title, bullets or backend; ad coverage
Impression share well above click share
Shoppers see you and pick someone else
Main image, title, price in results, reviews
Click share well above cart add share
The detail page does not convince
Price versus the median, images, bullets, A+, delivery date
Cart add share well above purchase share
Hesitation after the cart
Stock, delivery promise, competing coupons
Purchase share above click share
You convert better than the market
Whether to buy more visibility here
Prompt 2 · Find where you lose the shopper in the funnelChatGPT Claude
Act as an Amazon conversion analyst. Using the SQP table below, diagnose my top [30] queries by Search Query Score. Compare my brand share at impressions, clicks, cart adds and purchases. A drop of more than [3] percentage points between stages is a leak at that stage. Skip queries with volume below [500]. Output a table: query, volume, the four shares, the leak stage, the likely cause and one thing to check on my listing. Then rank the five biggest fixes by estimated lost purchases (total purchases for the query x the share points lost at the leak), marked as estimates. Data: [paste the clean SQP table]
Prompt 3 · Find where you lose the shopper in the funnelChatGPT Claude
Act as an Amazon category manager. Sort my SQP queries into Converters (purchase share at least [1.2] times click share) and Leakers (purchase share below [0.8] times click share). For each Converter, say whether impression share is low enough that more visibility is the obvious next step. For each Leaker, name the part of the detail page most likely at fault, citing the price and cart add columns as evidence. End with one paragraph on what each group has in common, such as a size, use case or price band. Data: [paste SQP Brand View CSV]
Prompt 4 · Find where you lose the shopper in the funnelChatGPT Claude
Act as an Amazon pricing and fulfillment analyst. The SQP export below has Price (Median) and Brand Price (Median), plus shipping speed counts, at the click, cart add and purchase stages. For queries with volume above [1,000], show the market median price, my median price and the gap in percent at the purchase stage, and the share of market purchases versus my purchases that went to same-day, 1-day or 2-day delivery. Flag queries where I am more than [10]% above the median or slower than the market and my purchase share trails my click share. Do not recommend a new price; list what to check before deciding. Data: [paste SQP Brand View CSV]
Tip: Treat any price conclusion as a lead, not a verdict, and check your margin before you change anything.
Turn SQP into keyword and listing decisions
SQP only shows queries where your brand already appeared, which makes it a poor discovery tool and an excellent listing tool: these are real queries with real purchases where you are already in the game. For terms outside that set, use our prompts for discovering high-impact Amazon keywords.
Prompt 5 · Turn SQP into keyword and listing decisionsChatGPT Claude
Act as an Amazon listing SEO specialist. Below are my SQP queries for [ASIN] and my current title, five bullets and backend search terms. For every query with volume above [500] and my purchase share above [2]%, check whether its main words appear in the title, the bullets or the backend terms. Return three lists: in the title, only in bullets or backend, and nowhere. For the last list, suggest where each term could go without stuffing, keeping the title readable and within my category's length limit. Never add brand names I do not own. SQP: [paste rows]. Listing: [paste title, bullets, backend terms]
Why: Queries where you already convert but never mention the term are the cheapest wins in the account. The backend keyword prompts cover what does not fit on the page.
Prompt 6 · Turn SQP into keyword and listing decisionsChatGPT Claude
Act as an Amazon brand strategist. Put every query in the SQP data below into one bucket. Defend: high volume and one of my highest purchase shares. Grow: healthy click and purchase share, low impression share. Fix: fine impression share, but a leak at clicks, cart adds or purchases. Ignore: not relevant to what I sell. Give each query its bucket, a one-line reason from the numbers, and the owner of the next step: listing, ads, pricing or nobody. Output a table sorted by bucket, then volume. Data: [paste the clean SQP table]
Prompt 7 · Turn SQP into keyword and listing decisionsChatGPT Claude
Act as an Amazon catalog manager. Below are SQP exports in ASIN View for [number] of my ASINs. For every query that appears for two or more of them, show each ASIN's impression, click and purchase share. Tell me which ASIN wins the query on purchases, whether another ASIN takes impressions without converting, and whether to differentiate the listings or focus the query on one ASIN. Data: [paste ASIN View exports, labeled by ASIN]
Line SQP up against your ads
SQP counts every impression, click and purchase on a query, organic and sponsored together. The Sponsored Products search term report counts only what ads drove, under Amazon's attribution rules. Side by side, they show where ads carry a query and where organic does the work.
Act as an Amazon PPC analyst. File 1 is my SQP Brand View. File 2 is my Sponsored Products search term report for the same period, with Customer Search Term, Impressions, Clicks, Spend, 7 Day Total Sales and 7 Day Total Orders (#). Match rows where the search term equals the query, ignoring case. For each match, show SQP brand purchases next to ad-attributed orders and classify the query as Ad-dependent (ad orders are most of my purchases), Mixed or Organic-led. Then list SQP queries with my purchase share above [2]% that I do not advertise on. Note that attribution differs, so ratios are directional. File 1: [paste SQP]. File 2: [paste search term report]
Tip: This is where most sellers stop, because it means downloading both reports every week and lining them up by hand. DataDoe connects Seller Central, Vendor Central and Amazon Ads to Claude, ChatGPT and other AI clients through MCP, with SQP, ads and sales joined in one schema and refreshed daily, so you can ask this question without the copy-paste.
Act as an Amazon advertising strategist. Using the matched table below, give me three lists. Harvest: queries where I win organic purchases but run no exact-match ads. Review: queries with heavy spend where my SQP purchase share is flat or falling across [two] periods. Hold: Organic-led queries with high purchase share, where ads may buy sales I would win anyway. For each, give the evidence and one test, such as a bid change or a two-week pause, with the metric that decides it. These are suggestions for me to review, not changes to make. Data: [paste matched SQP and ads table]
Why: The Hold list is where ads quietly inflate cost, and measuring it takes TACoS rather than ACoS. Our TACoS vs ACoS prompts walk through it.
Track SQP week over week
One SQP week is a snapshot. When search volume moves, the market moved; when your share moves, something about you changed.
Act as an Amazon performance analyst. Compare two SQP Brand View exports, Period A [dates] and Period B [dates]. For queries in both, show the change in volume and the change in my share at each stage, in percentage points, and separate market change (volume moved, share stable) from share change. Flag any query where my purchase share moved by more than [2] points and match each flag to my change log: [paste price, listing, stock and ad changes]. Do not claim a cause the log does not support. Period A: [paste]. Period B: [paste]
Act as an Amazon brand manager. From the last [8] weekly SQP exports below, pick the [25] queries that matter most, weighted by my brand purchases and Search Query Score. Return a table I can paste into a spreadsheet: one row per query, one column per week for purchase share and impression share. Then list every query with three falling weeks in a row. Data: [paste weekly exports, labeled by week]
Pair SQP with the other Brand Analytics reports
SQP shows the queries you already touch. Top Search Terms shows who wins a term, and Search Catalog Performance shows the funnel per ASIN.
Prompt 12 · Pair SQP with the other Brand Analytics reportsChatGPT Claude
Act as an Amazon competitive analyst. Below is a filtered Top Search Terms report: Search Term, Search Frequency Rank, and the top three clicked ASINs with their click share and conversion share. For each term, give the combined click share of the top three and whether one ASIN dominates conversion share. Mark terms where the top three hold a low combined share as open. Cross-check with my SQP data and flag terms where I appear in SQP but not in the top three. Top Search Terms: [paste rows]. My SQP: [paste]
Prompt 13 · Pair SQP with the other Brand Analytics reportsChatGPT Claude
Act as an Amazon catalog analyst. Below is my Search Catalog Performance report with impressions, clicks, cart adds and purchases per ASIN for [period]. Calculate each ASIN's click, cart add and purchase rate between stages, flag the ASIN with the weakest step against my catalog median, and say which part of its detail page to review first. Data: [paste Search Catalog Performance CSV]
Two habits keep these numbers honest: never treat cart adds as sales, and never read a share without its volume, because a big share of a tiny query is noise.
FAQ
What is Amazon Search Query Performance?
It is a Brand Analytics dashboard that shows, for each search query that led to your brand, the total impressions, clicks, cart adds and purchases on Amazon and how many went to your brand or ASIN. It comes in Brand View and ASIN View, by week, month or quarter.
Who can access Amazon Brand Analytics?
Brand owners enrolled in Amazon Brand Registry, on a Seller Central or Vendor Central account with the right permissions, such as the Brand Representative role. Eligibility can differ by marketplace, so check the current help page if you cannot see it.
What does brand share mean in SQP?
Brand share is your brand's count divided by the total count for the query at the same funnel stage. A 10% purchase share means one in ten purchases from that query went to your brand in the period.
What is the difference between Search Query Performance and Search Catalog Performance?
Search Query Performance is organized by query and shows your share of each query's funnel. Search Catalog Performance is organized by ASIN and shows each product's funnel across all the searches it appeared in.
Can I use ChatGPT or Claude to analyze SQP data?
Yes. Paste the CSV into a prompt that names the columns, the decision rules and the output format. Check the model's math on a few rows before you act on any listing, price or bid change it suggests.
SB
Sophie Brooks
Prompt engineer and AI strategist helping marketplace sellers save time and grow. Specializes in building high-impact prompts for listings, ads, and support workflows.
Free download
27 prompts top sellers actually use.
Listings, ads, reviews, analytics and support: the Amazon Seller Prompt Pack as one PDF.
