Search Query Performance: find where you lose the click and where you lose the sale
The Search Query Performance dashboard in Brand Analytics shows your share of every step from impression to purchase, query by query. Read it as a funnel and it tells you whether a keyword needs a bigger bid, a better main image or a better price.
The Sadeeqy teamAmazon Ads strategists for UAE and Saudi brandsAI reads everything. A senior strategist decides everything.
A keyword report from the ad console tells you what your ads did. It can't tell you why a shopper who saw your product chose someone else. That question sits in a different report, and plenty of GCC brands have never opened it.
The Search Query Performance dashboard (SQP) lives in Seller Central under Brands, then Brand Analytics. It shows, for each search query, how many shoppers searched it and what share of the impressions, clicks, cart adds and purchases went to your brand. Read across a row and you can see exactly where you drop out.
What the dashboard shows
To open it you need a Professional selling account and the Brand Representative role on a brand enrolled in Amazon Brand Registry. Amazon introduced the dashboard to Amazon.ae brands through the Seller Forums; open Brand Analytics in each marketplace you sell in and switch between them at the top of the page.
For each query you get:
Search query volume: how many times shoppers searched it in the period.
Impressions, clicks, cart adds and purchases, each as a total for all products, your brand's count, and your brand share.
Median price for all products and for yours, and clicks split by shipping speed (same-day, 1-day, 2-day).
There's a brand view for the whole catalogue and an ASIN view for one product, with weekly, monthly and quarterly periods. Amazon's own page describes the purpose plainly: to "identify drop-off points" between discovery and purchase.
One point that matters in the Gulf: queries appear as shoppers typed them, so "شنطة ظهر" and "backpack" are separate rows with separate volumes. Don't add them together in your head. They're often different shoppers with different price expectations.
Reading it as a funnel
Take your brand share at each step and compare it with the step before. The step where your share drops is where you're losing. Here's a worked example for a backpack brand, with made-up numbers for illustration.
Query
Impression share
Click share
Cart add share
Purchase share
What it means
backpack for school
4%
4.5%
6%
7%
Strong page, too little visibility
laptop backpack
12%
6%
6.5%
6%
Losing the click
شنطة ظهر
9%
10%
9%
3%
Losing the sale late
Plotted step by step, the three patterns look like this:
Brand share at each funnel step
backpack for schoollaptop backpackشنطة ظهر
Example numbers from the table above
Each pattern points to a different fix.
Shares rise as you go down ("backpack for school"). Shoppers who reach you buy at a higher rate than the market. Your problem is reach. This is where more bid, top-of-search placement and a Sponsored Brands campaign usually pay, because each extra click converts better than average.
Share falls between impressions and clicks ("laptop backpack"). You're seen but not chosen. Bidding higher here buys more impressions that don't get clicked. Look at what the shopper sees in the results: main image, price against the median price column, star rating and review count, and whether a competitor shows a coupon or a faster delivery promise. The shipping-speed columns help here; if most clicks go to same-day or 1-day offers and yours isn't one, that's part of the answer.
Share holds through cart adds and falls at purchase ("شنطة ظهر"). The shopper liked you enough to add to cart, then bought elsewhere. Common causes are a price change, a lost Buy Box, a delivery date that slipped, or a competitor's deal. Check the Arabic version of the listing too. If the Arabic title or bullets are a weak machine translation, a shopper browsing in Arabic may add to cart on the image and then drop out after reading the page.
Here's how to turn the funnel into a weekly routine.
Export the last 3 months at brand level, sorted by search query volume, and keep the top 50 queries that match what you sell.
Add one column per step: your share divided by the previous step's share. Under 1 means you lose shoppers at that step.
Tag each query: reach, click, page or checkout.
Fix in that order of cost. Bids and placements for "reach" queries, main image, price and coupon for "click" queries, listing content for "page" queries, stock, Buy Box and delivery promise for "checkout" queries.
Re-read the same queries after four weeks. SQP gives you the before-and-after record that ad reports alone can't.
Applied to the example table, step 2 gives these ratios:
Step ratios (under 1 = losing shoppers at that step)
backpack for schoollaptop backpackشنطة ظهر
Example numbers from the table above
Where it hits profit
SQP stops you from paying to fix the wrong problem. On a "losing the click" query, every bid increase buys impressions your listing already fails to turn into clicks, and your cost per order rises with nothing to show for it. On a "reach" query, the same money does the opposite, because you're buying traffic that already converts better than the market.
There's a second gain that rarely gets counted. SQP purchases include organic orders, so it shows the queries where you already win without ads. Those are the first places to test lower bids, which is often the quickest TACoS improvement an account has. If you'd like help finding them, we published a free Search Query Performance analyser you can run yourself.
Why don't my SQP numbers match my ad reports? SQP counts all purchases from shoppers who searched a query, ad and organic together. Your ad reports count orders credited to your ads. They answer different questions: use SQP for share and the ad report for cost.
Can I see which competitors take the rest of the share? Not in SQP. It shows totals for all products and your brand's share, not the other brands. Look at the search results page yourself, or use the Top Search Terms report in Brand Analytics, which lists the most-clicked products for a term.
Should I read it weekly or monthly? Monthly for planning, weekly for checking a change you made. Smaller Gulf queries can swing a lot from one week to the next, so don't act on a single week.
Get the next brief by emailWhat changed on Amazon for sellers in the UAE and Saudi Arabia, in English or Arabic. Short, and only when something matters.