Home · Blog

How to Read an Amazon Ads Search Term Report

The search term report shows what shoppers actually typed before they clicked your book ad. A step-by-step triage order for KDP authors: what to harvest, what to negate, and what to leave alone.

The BookAdsAI team · Reports · Published October 5, 2026.

Your keywords are what you bid on. Search terms are what shoppers actually typed. The gap between the two is where most book ad budgets leak, and the search term report is the only place Amazon shows it to you. If you read one report each week, make it this one.

Where to find it and what it contains

In the Amazon Ads console, open a Sponsored Products campaign, choose an ad group, and look for the Search terms tab. You can also download a Sponsored Products search term report from the Reports section for a date range across all campaigns, which is easier to sort in a spreadsheet.

Each row pairs one of your targets with one customer search term and shows impressions, clicks, spend, orders, sales, and, for KDP books, KENP pages read. The important columns are:

• Targeting: the keyword or product target you bid on, or the auto-targeting group (close match, loose match, substitutes, complements).
• Match type: broad, phrase, exact, or auto.
• Customer search term: the words the shopper typed. When your ad appeared on a product page instead of in search results, this column shows that product's ASIN, a ten-character code that usually starts with B0 for Kindle books.
• Clicks, spend, orders, and KENP: what that pairing cost and what it produced.

Pull a window long enough to mean something

Book ads produce small numbers. One sale more or less can swing a keyword from hero to failure. Use at least 30 days, and leave out the last three or four days because attributed sales keep arriving after the click. If a book gets very little traffic, 60 days is better. You are looking for patterns, not individual days.

A triage order that keeps you calm

Sort the report by spend, highest first, and work through it in this order.

1. Spenders with zero orders and zero page reads

These are the leaks. Look at the search term, not just the number. Ask whether a reader of your book would plausibly type it.

• If the term is clearly the wrong intent, such as a different genre, a nonfiction term on a novel, a free-book search, or an unrelated product, add it as a negative. Negative exact blocks that exact search. Negative phrase blocks any search containing the phrase.
• If the term looks right but simply has not converted yet, check the click count. Ten or fifteen clicks without a sale on a well-matched term is not proof of failure for a book that converts at 1 in 15. Lower the bid on the keyword that triggered it instead of blocking it.

2. Terms that converted but are not yet keywords

These are your harvest. When a broad, phrase, or auto target produces a sale or meaningful page reads through a specific search term, add that term as an exact-match keyword in a manual campaign. You then control its bid directly. Many authors also add it as a negative exact in the original auto or broad ad group, so the two campaigns do not compete for the same search.

3. ASINs in the search term column

Rows where the search term is an ASIN mean your ad showed on another book's product page. Look up the ASIN on Amazon. If it is a close comp in your subgenre and it converts, consider adding it to a product-targeting campaign. If it is a book that has nothing to do with yours, add it as a negative product target in that ad group.

4. High impressions, almost no clicks

A term with thousands of impressions and a click-through rate far below your other terms is telling you shoppers see the ad and do not care. That is rarely fixed by bidding more. The search is either a poor fit, or your cover and title do not look like what that shopper wants. Leave it or negate it, depending on how relevant it is.

5. Everything else

Small terms with a click or two are noise. Leave them for next month.

Reading the numbers correctly

• Orders count only attributed purchases. A Kindle Unlimited borrow is not an order. It shows up later as pages read in the KENP column, so a term with zero orders and healthy page reads is earning money.
• Sales are list price, not royalty. Convert before you call a term profitable. A sale of a $14.99 paperback with a high print cost earns less than a $4.99 ebook sale.
• Conversion rate needs volume. Two sales from three clicks is luck as much as signal. Treat anything under about ten clicks as unproven.

A worked example

Say a cozy mystery author runs an auto campaign at a $0.45 default bid. A 45-day search term report sorted by spend shows:

1. "cozy mystery books kindle unlimited" with 38 clicks, $15 spend, 3 orders, and 2,100 pages read. This is a winner. Add it as exact match in a manual campaign and negate it as exact in the auto ad group.
2. "true crime books" with 22 clicks, $9 spend, no orders, and no pages read. Wrong intent for a cozy. Add "true crime" as a negative phrase.
3. A competitor ASIN with 14 clicks, 1 order, and 600 pages read. Promising. Add it to a product-targeting campaign at a modest bid and watch it.
4. "free kindle books" with 9 clicks and nothing. Add "free" as a negative phrase.
5. "mystery books for women" with 6 clicks and nothing. Too early to judge. Leave it.

Twenty minutes of work removed $24 of waste from the next six weeks and gave the author two new targets with proven buyers behind them.

Habits that make the report easier

• Keep a running document of negatives by book, so you do not decide the same term twice.
• Name campaigns so you can tell auto, broad, exact, and product targeting apart at a glance.
• Write down why you negated something. In three months you will not remember.
• Change a few things per week, then wait. If you change twenty targets at once, you cannot tell which change helped.

Where BookAdsAI fits

BookAdsAI does not harvest search terms or add negatives for you. Autopilot works on bids and spend caps for campaigns you allowlist, and only uses metrics Amazon has reported. The search term report remains your job, and this weekly read is the most valuable fifteen minutes you can give your ads. For the rest of the routine, see a weekly book ads review ritual.