The Amazon Search Term Report: How to Read It and What to Do Next

The Amazon search term report is the single most useful file in your advertising account. It tells you exactly what shoppers typed into the search bar before they clicked on your ad, and what happened next. Most sellers download it, sort by ACoS, pause a few things and close the spreadsheet. That leaves a lot of money on the table.

This guide goes further. We will cover where to find the report on amazon.com and other marketplaces, which columns deserve your attention, and a simple four-bucket method that turns every row into a clear decision. You will also see a worked example, a weekly routine and how the report fits with Amazon's other search data.

What is the Amazon search term report?

The Amazon search term report lists the shopping queries that triggered your ads and led to at least one click. Amazon describes it in its Sponsored Products best practices guide as a way to see the search terms that resulted in a click, so you can refine your keywords and targets.

It is one of six Sponsored Products reports Amazon lists in its guide to measuring and improving campaigns. The others are the targeting, advertised product, placement, performance over time and purchased product reports. The search term report answers the most practical one: what are people actually searching for when they find me?

Keywords vs customer search terms

This is the distinction that trips people up. A keyword is what you bid on. A customer search term is what the shopper typed.

Say you bid on the broad match keyword "bamboo cutting board". Your ad might show for "large bamboo chopping board", "cutting board set for kitchen" or even "bamboo plates". The report puts these side by side: the targeting column shows your keyword, and the customer search term column shows the shopper's actual words.

With exact match the two columns look almost identical. With broad, phrase and automatic targeting they can differ a lot, and that gap is where most of your waste and new opportunities hide.

What the report includes (and leaves out)

Each row is a combination of campaign, ad group, targeting, match type and customer search term for the date range you chose. Against each row you get impressions, clicks, click-through rate, cost per click, spend, orders, sales, conversion rate, ACoS and ROAS.

Some rows will show a product code such as B0XXXXXXXX instead of words. Those are ASINs. They appear when your ad was shown on, and clicked from, another product's detail page, usually through automatic targeting or product targeting. They are worth as much attention as the keyword rows.

Here is what it does not tell you:

• Which product sold. The report works at ad group level. If an ad group holds five products, you cannot see which one the shopper bought from this report alone.
• Searches with no clicks. Terms that earned impressions but never a click do not appear, so you cannot judge everything you were shown for.
• Your share of the market. It shows your results, not how much of the search you captured. That is what the impression share report is for, which we cover later.

The first gap has a simple fix. Amazon's own advice is to use separate campaigns or ad groups for each product group and to keep ASINs from the same category together. If you keep one product, or a tight group of variations, per ad group, every search term maps cleanly to the product it sold.

How do you download the search term report?

Amazon renames menus from time to time, so treat the labels below as a guide rather than gospel. The path has been broadly the same for years.

Sponsored Products

1. Open the Amazon Ads console. From Seller Central you can reach it through the Advertising menu and Campaign Manager.
2. Go to the reporting area (usually Measurement & Reporting, then Sponsored ads reports).
3. Create a report. Choose Sponsored Products as the ad product and Search term as the report type.
4. Pick the time unit. "Summary" gives one row per term for the whole period. "Daily" splits each term by day, which is useful for trend checks but makes the file much larger.
5. Set the date range, name the report, then run it. Download the file when its status shows as completed.

Sponsored Brands

Yes, there is a Sponsored Brands search term report, and you create it the same way: pick Sponsored Brands as the ad product, then the search term report type. If you run more than one Sponsored Brands format, check which formats your report covers before you compare results.

Two differences matter. Sponsored Brands ads often feature several products, so results sit at campaign or ad level rather than one ASIN. And they are often judged on brand goals as well as direct sales, so a slightly higher ACoS can be acceptable if a term brings new shoppers to your brand.

Choosing a date range

Too short, and you will react to noise. Too long, and old behaviour hides recent changes. For most accounts we work with a rolling 30-day view for decisions and a 60- to 90-day view for low-traffic terms that need more data.

Two practical rules help:

• Leave out the last few days. Sales attributed to recent clicks can still be coming in, so recent days often look worse than they are. Judge them next week.
• Keep an archive. The report only covers a limited window of history, so download it on a schedule and save each file. Six months from now you will want to compare seasons.

Running it for several marketplaces

If you sell on amazon.com and amazon.co.uk, or on .de and .ca too, each marketplace has its own campaigns and its own report. Run them separately. Search behaviour differs even between the US and the UK: "cutting board" and "chopping board" perform very differently on each site, so never copy a negative list across without checking it.

Which columns matter most?

The file has a lot of columns. You need to understand about ten of them. We keep definitions short here, because we cover the ratio metrics in depth in our guides to what ACoS means and ACoS and TACoS benchmarks.

Impressions, clicks, CTR and CPC

Impressions are how often your ad was shown for that term. Clicks are how often a shopper clicked. Click-through rate (CTR) is clicks divided by impressions. A low CTR on a term that is clearly relevant often points to your main image, price or review count rather than the keyword itself.

Spend is what you paid in total. Cost per click (CPC) is spend divided by clicks. A steady rise on your best terms usually means more competition for that search.

Orders, sales and conversion rate

Orders and sales are attributed to the ad click within Amazon's attribution window, and the column name tells you which window applies to that report (for example "7 Day Total Sales"). Conversion rate is orders divided by clicks. This is the column that tells you whether the term brings buyers or browsers.

ACoS and ROAS

ACoS is spend divided by sales, shown as a percentage. ROAS is the inverse: sales divided by spend. They tell the same story, so pick the one your team uses and stick with it.

The figure that matters is not ACoS on its own. It is ACoS compared with your target, which comes from your margin. If a product makes $9.00 profit before advertising on a $29.99 sale, your break-even ACoS is about 30%. A term at 35% ACoS is losing money on that product. On a product with a 45% margin, the same term is profitable.

From columns to decisions: the four buckets

Every row in the report needs one of four actions, or none. Here is the model we use, and the rest of this guide takes each bucket in turn.

The 20-click trigger comes from Amazon's own Sponsored Products targeting guide. The other triggers depend on your margin and category, so treat them as starting points to tune.

Bucket 1: harvesting winning search terms

Keyword harvesting is the process of taking search terms that sell from your discovery campaigns (automatic, broad and phrase) and giving them their own exact match keyword. It is how most well-run accounts grow.

Amazon's advice points the same way. Its article on common sponsored ads mistakes recommends migrating successful search terms to tighter match types, and its best practices guide suggests combining automatic and manual targeting rather than relying on one.

What counts as a winner

A winner in your Amazon search term report has three things: it converts, it does so at or under your target ACoS, and it has enough data behind it to be trusted. One order from three clicks is a hint, not a winner.

As an example rule of thumb, many sellers start with "at least two orders and ACoS at or below target over 30 days". Adjust it for your price point. A $200 product may never get two orders from one long-tail term in a month, while a $10 consumable might need five or more before you can trust it.

Do not forget the ASIN rows. If a competitor's product page keeps sending you profitable clicks, add that ASIN as a product target in a manual campaign. That is ASIN harvesting, and it is often overlooked.

Moving winners to exact match

Add the winning term as an exact match keyword in a manual campaign, ideally in an ad group that holds only the product it sold. Start the bid close to the CPC the term was already paying, or slightly above if you want more impressions.

Negating the harvested term in the source campaign

This step is the one people skip. Once a term has its own exact keyword, add it as a negative exact in the auto or broad campaign it came from. Otherwise both campaigns can bid on the same search, your data gets split across them and you lose the control you just created.

Use negative exact here, not negative phrase. Negative phrase would also block longer searches that contain the term, and those longer searches are exactly what you want your discovery campaigns to keep finding.

Bucket 2: cutting bleeding search terms

Bleeders are terms that spend money and do not sell. In a large account they can quietly eat a big share of the budget, which is why this bucket usually delivers the fastest savings.

Spotting clicks with no orders

Filter your Amazon search term report for rows with zero orders, then sort by spend from highest to lowest. The top of that list is where to start.

Amazon's targeting guide gives a clear benchmark: before adding a keyword as a negative target, evaluate its performance after it has received at least 20 clicks. That protects you from cutting a term that just had an unlucky week.

There is one sensible exception. If a term is plainly irrelevant, such as "plastic cutting board" when you sell bamboo, you do not need 20 clicks to know it will not convert. Relevance is your call. Performance needs data.

Negative exact or negative phrase?

Sponsored Products offers two negative keyword match types: negative exact and negative phrase.

• Negative exact blocks that precise search only. Use it when the term is close to relevant but does not convert, for example "bamboo cutting board for cheese" on a large chopping board.
• Negative phrase blocks any search that contains the phrase. Use it for words that will never suit your product, such as "plastic", "glass" or a size you do not make.

For ASIN rows that bleed, add the ASIN as a negative product target. Amazon also lets you exclude whole brands where their product pages do not meet your goals. Our full guide to Amazon negative keywords covers campaign-level versus ad group-level negatives and how to keep a shared list tidy.

Bucket 3: re-bidding terms that are close to target

Not every term is a clear yes or no. A lot of rows sit in the middle: they sell, but either not often enough or at too high a cost. These need a bid change, not a negative.

Raising bids on efficient, low-impression terms

Look for terms with a good conversion rate and a low ACoS but only a few hundred impressions. Amazon's measurement guide is direct about this: where impressions are low but CTR is high, consider raising bids or budgets, and check whether a low bid is the reason the ad is rarely shown.

Raise the bid in small steps (as an example, 10–20% at a time), then give it a week or two before the next change.

Lowering bids on expensive converters

The opposite case: terms that sell steadily but above your target ACoS. Pausing them throws away real sales; a lower bid usually keeps some volume at a cost you can live with.

A simple way to set the new bid is to work from your target cost per click. Multiply your conversion rate by your target cost per order. If the term converts at 8% and you can afford $7.50 per order, your target CPC is about $0.60. If you are paying $0.95, that tells you how far to come down.

For more on bid rules, dayparting and placement modifiers, see our guide to Amazon PPC bidding strategies.

Bucket 4: when the problem is the listing, not the keyword

Most guides leave this bucket out, yet it often holds the biggest gains. Sometimes the keyword is right, the shopper clicks, lands on your page and leaves. Cutting the keyword will not fix that.

High CTR, low conversion

Amazon's article on common campaign mistakes calls this detail page misalignment: clicks without sales usually mean a gap between what the ad promised and what the product page delivers. Shoppers liked what they saw in the results, then found something they did not like.

When a relevant, high-traffic term has a good CTR and poor conversion, check the page before the bid:

1. Does the title and first image clearly show the feature the shopper searched for? Someone searching "with juice groove" needs to see the groove.
2. Is the price in line with the products around you on that search?
3. How do your rating and review count compare with the top results?
4. Do the bullets and A+ content answer the obvious questions (size, material, care)?
5. Is the right variation being shown, and is it in stock?

If the product genuinely lacks what the term describes, negate it. If the product has it but the page hides it, fix the page. Our guide on how to optimise an Amazon listing walks through each element.

Low CTR on relevant terms

Amazon's mistakes article also suggests reviewing search term reports for terms with plenty of impressions but poor CTR, as a sign the keyword and product are misaligned. Sometimes that is true and the answer is a negative.

But if the term is clearly relevant and your CTR is low, the issue is usually in the search results themselves: your main image, title, price or star rating compared with the products next to yours. Look at the actual results page for that search.

How to analyse your Amazon search term report in a spreadsheet

You do not need software to do good Amazon search term report analysis. Excel or Google Sheets and a pivot table will cover most accounts. Here is a workflow you can rebuild as your own template.

Aggregating by search term

1. Download a summary Amazon search term report for your chosen range, leaving out the most recent few days.
2. Paste it into a raw data tab. Do not edit this tab, so you can always check back against the original.
3. Build a pivot table with customer search term as rows and impressions, clicks, spend, orders and sales as summed values.
4. Add calculated columns: CTR (clicks ÷ impressions), CPC (spend ÷ clicks), conversion rate (orders ÷ clicks) and ACoS (spend ÷ sales).

Aggregating matters because the same term can appear many times, once for each campaign, ad group and keyword that matched it. A term that looks small in each row can be one of your biggest spenders in total.

Before you act, filter back to campaign level, because negatives and new keywords are added in specific campaigns.

Adding target cost per order

Add one more column for each product or ad group: your target ACoS, based on margin. If you keep one product per ad group, a simple lookup against a margin table does the job. This is what stops you from applying a single "30% is bad" rule to products with very different margins.

Flagging actions with simple formulas

Then add an "Action" column that uses IF statements to label each row. Below is an example for a $29.99 bamboo cutting board with a break-even ACoS of 30% and a target of 25%. All figures are illustrative.

Look at the last row. Ninety-six clicks with a healthy CTR means shoppers want this. Two orders means the page is not convincing them. If the board has a juice groove but the images do not show it, that is a listing fix worth far more than a negative keyword.

Once the labels are in, sort by action and make the changes in the Ads console or through a bulk file.

Keep a simple change log: date, term, action and reason. When you review the report next month, you will be able to see whether your changes worked.

Search term report vs impression share report vs Search Query Performance

The search term report only shows searches you paid for. Two other Amazon tools fill in the rest of the picture.

Search term impression share report

Amazon launched the search term impression share report for Sponsored Products in March 2021. It sits in the Report Center and shows, for each search term, your account-wide impression share and impression rank compared with other advertisers. It is available in the US, the UK and other marketplaces.

Put it next to your Amazon search term report and look for terms that convert well but where your impression share is low. Those are the clearest places to raise bids or budgets, because you already know the term sells and you can see how much of the search you are missing.

Brand Analytics Search Query Performance

If you are brand registered, Amazon Brand Analytics includes the Search Query Performance dashboard. It lists the searches customers use to find your brand, with query volume, impressions, clicks, cart adds and purchases. It requires a Professional selling account and the Brand Representative role in Brand Registry.

The key difference is that Search Query Performance covers organic and paid activity together. Use it to find searches where your brand converts well organically but you do not advertise, and to check whether a term you are paying for also has strong organic results. Our Amazon keyword research guide and free keyword mapper help you organise what you find.

How often should you review your Amazon search term report?

There is no single right answer, but most active accounts need a short weekly check and a deeper monthly review.

Weekly routine

Once a week, download the last 30 days of your Amazon search term report (minus the most recent few) and work through two buckets:

• Bleeders that have crossed 20 clicks with no orders, plus anything clearly irrelevant.
• Obvious winners from your auto and broad campaigns that are ready for exact match.

Monthly deep dive

Once a month, look at a longer window and work through the re-bid and listing buckets. Pull the impression share report and, if you have it, Search Query Performance. Check your campaign structure too: Amazon's mistakes article flags inefficient structure without clear naming conventions as a common problem, and a messy structure makes every report harder to read.

Steady, small, regular changes tend to work better than big monthly overhauls.

When to get help

If your report runs to thousands of rows across several marketplaces, the weekly routine becomes a part-time job. That is usually when sellers bring in an Amazon PPC management agency. If you just want a second opinion, a free Amazon audit is an easy place to start.

Launching a new product with no search term data yet? Our Amazon keyword research service can give you a starting list to test.

Frequently asked questions

Last reviewed 19 October 2026. We update articles in place rather than republishing at a new URL, so the link you have keeps working.