How to Read Your Amazon Search Term Report (2026 Step-by-Step Guide)

The Amazon search term report is the single highest-ROI document in your Seller Central account. Every rupee of ad waste you currently pay for is sitting in this report, waiting to be negated. Every winning keyword you have not yet promoted is in here too. Yet most Indian sellers either never download it or open it, feel overwhelmed by 40,000+ rows, and close it. This guide fixes that. In the next 10 minutes you will know exactly what to promote, what to negate, and which columns to ignore.

1. What is the Search Term Report?

The search term report shows the actual words customers typed into Amazon before your ad was shown or clicked. This is different from your keyword report, which shows the keywords you targeted. If you bid on the keyword “yoga mat,” Amazon might show your ad to someone who typed “yoga mat for beginners pink,” “best yoga mat india,” or “eco friendly yoga mat.” Each of those is a customer search term — and you only see them in this specific report.

Three match types produce search terms: broad (loose matches), phrase (contains your keyword), and exact (identical to your keyword). Auto campaigns also generate search terms because Amazon picks the match. For your weekly audit, 80% of the insights come from broad and auto campaigns — phrase gives you mid-value signals, and exact rarely surfaces new information.

2. How to Download the Report from Seller Central

From Amazon Seller Central India: go to Advertising → Reports → Sponsored Products → Search term report. Choose the time range — we recommend 30 days for weekly audits and 90 days for monthly deep dives. Filter by Campaign Group if you want to audit a specific brand or SKU line. Click Run report and download as CSV or Excel.

Two common gotchas for Indian sellers: (1) The report takes 5-20 minutes to generate for mid-sized accounts; download on a 5-minute timer in a separate browser tab. (2) Dates are in UTC, not IST — a 30-day window ending on 18 April IST actually captures 18 March 00:00 UTC to 18 April 00:00 UTC, which is 5:30 AM IST boundary.

3. The 7 Columns You Must Look At

The report has 20+ columns. Ignore most of them. These seven carry 95% of the decision-making weight:

4. “Customer Search Term” vs “Keyword” — the Distinction

This confuses most sellers. The keyword is what you bid on. The customer search term is what was actually typed. In a broad-match campaign on “running shoes,” a customer search term could be “running shoes for flat feet men,” “gym shoes,” or even “trainers for jogging.” Amazon decides which typed queries are semantically close enough to match your broad-match bid.

The mental model: keywords are your fishing rods, search terms are the fish that actually came up. Your job auditing the report is to keep the fish you want (promote), throw back the ones you do not (negate), and sometimes build a better rod (exact-match the winners).

5. The Promote / Negate Decision Framework

Run every search term through this three-step filter:

A fourth case — converted but ACoS above break-even — is the hardest judgment call. If the term is strategically important (new-to-brand, defensive, category-awareness), keep bidding at break-even. Otherwise treat it like a zero-order term and negate. Use the free ACoS calculator to compute your break-even ACoS in rupees.

6. Using Search Term Impression Share

Amazon added this column in 2023 and it is criminally underused. It shows what percent of total impressions for a given search term your ad captured. A converting term with 15% impression share means you are winning the sale only 15% of the time someone searches for it — there is 85% of upside sitting right there.

The move: for any high-converting term with impression share below 40%, raise your exact-match bid by 20-30% and monitor for 7 days. If ACoS stays below break-even, raise again. If ACoS rises, you have found the ceiling for that term. This is the single fastest way to scale a winning campaign without building new campaigns.

7. Weekly Cadence — What Good Looks Like

Spend 45-60 minutes on this every Tuesday morning. Over 90 days the compounding effect on ACoS is dramatic — sellers who do this weekly typically run 8-12 percentage points lower ACoS than sellers who do it monthly.

1. Download the 30-day search term report (Sponsored Products, then Brands, then Display).
2. Sort by spend descending. Focus on the top 50 rows — these drive the decisions that matter.
3. Apply the three-step filter (data → conversion → profitability).
4. Batch negatives and promotions. Do not add them one-by-one in Seller Central — paste them into the bulk operations sheet and upload once.
5. Log what you did in a simple tracker. Next week you will know whether last week's changes moved ACoS.

8. Common Mistakes Indian Sellers Make

• Negating too aggressively in the first 30 days — sample sizes are small, and you may kill terms that just needed more data.
• Negating at campaign level instead of ad group level — blocks the term across all SKUs in the campaign, which is rarely what you want.
• Ignoring match type on negatives — “cheap yoga mat” as a negative phrase blocks anything containing those words; as a negative exact match only blocks that exact string. Pick deliberately.
• Only auditing Sponsored Products — Sponsored Brands and Display also produce search terms and waste. Do all three.
• Treating the report as a chore — this is the single highest-leverage activity in Amazon Ads. Block the calendar.

9. Automating This with eVanik

Doing this weekly across 200+ campaigns is where most seller teams break. eVanik's search term harvester ingests your search term reports from the Amazon Ads API every 24 hours, applies your custom promote/negate rules, and either auto-executes the changes or queues them for your approval. Our median customer eliminates 60+ hours of PPC operations per month and drops ACoS by 25-40% in the first 90 days.

See eVanik's Amazon PPC platform for a full tour — or if you want to learn more about the ACoS math first, read our companion guide on the Amazon ACoS formula with worked examples.