Most brand owners open the search term report, scroll for thirty seconds, feel overwhelmed by the rows, and close it. That report is the single most honest document in your advertising account. It tells you exactly what shoppers typed before they clicked your ad, what they bought, and what they cost you. Treated as a habit instead of a chore, it becomes the engine that lowers your ACoS and grows your sales at the same time.

The problem is not the data. The problem is reading it without a decision in mind. A strategist does not "look at" the report. A strategist runs every term through four buckets: harvest, negate, promote, and watch. Once each row has a home, the work is obvious.

First, set up the report so it can actually talk

Pull the Sponsored Products search term report for a window long enough to mean something. Seven days is noise. Use the last 30 to 60 days so a term has had a fair chance to convert. Anything shorter and you will negate a winner because it had not gotten its second sale yet.

Add one column the raw export does not give you: your target ACoS. If you do not have a number per product, derive it before you touch a single bid. We walk through that math in the right way to set a target ACoS for each product, and it changes every other decision on this page. A 40 percent ACoS is a disaster on a thin-margin product and a bargain on a fat-margin one. Without the target, the report is just numbers.

Sort by spend, highest first. The terms burning the most money, good or bad, deserve your attention first. You will never read all 4,000 rows. The top 50 by spend usually decide 80 percent of your wasted budget and most of your hidden winners.

Harvest: pull the winners into their own home

A harvest term is a customer search that converted well inside an automatic or broad campaign and deserves its own exact-match keyword. You are promoting it from "discovered by accident" to "managed on purpose."

Look for search terms with sales at or below your target ACoS and at least a couple of orders. Two orders is a signal. One could be luck. When you find one, add it as an exact-match keyword in a dedicated campaign where you control the bid directly, then add that same term as a negative exact in the campaign that discovered it. That last step matters. If you skip it, your auto campaign and your new exact campaign bid against each other for the same shopper and you pay more for the click you already won.

The search term report is not a performance review of your ads. It is a list of instructions your customers already wrote for you.

Harvesting is how a smart automatic campaign feeds a tight, profitable exact-match structure over time. The auto campaign is your scout. The exact campaign is where you press the advantage.

Negate: stop paying for clicks that never become sales

This is the cheapest profit in the account, and most brands leave it on the table. A negate candidate is a search term with real spend and zero sales, or an ACoS so far above target that no bid adjustment saves it.

Set a simple rule so you are not negotiating with yourself every week. A common starting line: if a term has spent more than two to three times your target cost per acquisition with no orders, it goes on the negative list. So if you would happily pay 12 dollars for a sale, a term that has spent 30 dollars and sold nothing has answered the question.

Watch for two kinds of waste in particular. The first is irrelevant intent: someone searching for a competitor's brand name, a use case your product does not serve, or a "free" or "cheap" modifier that signals the wrong buyer. The second is internal cannibalization, where a broad keyword keeps matching to a term you already manage better elsewhere. Negate it from the broad campaign so the traffic flows to the right place. The full system for doing this without cutting a winner by mistake is in our piece on negative keywords, the cheapest profit on Amazon.

A note on match types

Negate at the level that matches the problem. If one specific phrase is the waste, use negative exact so you do not accidentally block a profitable cousin. If an entire theme is wrong for your product, negative phrase casts the wider net you actually want. Precision here is the difference between trimming fat and cutting muscle.

Promote: feed more budget to what already works

Harvest creates the keyword. Promote decides how hard to push it. A promote term is one already converting under target that is held back by a low bid, a small budget, or a weak placement, not by demand.

Three signals tell you a term wants more:

Raise the bid in steps, not leaps. A 15 to 20 percent increase, then a week to read the result, beats doubling a bid and panicking. As the product matures and you push harder, the discipline that keeps this from blowing up is covered in scaling PPC without letting ACoS run away. Promotion is where growth comes from, but only when the underlying term has already proven it converts.

Watch: the bucket that saves you from yourself

The watch bucket is for terms that have not earned a decision yet. Some spend, no orders, but not enough spend to call it dead. A term with three clicks and no sale has told you nothing. Mark it, leave the bid alone, and check it next week.

This bucket is where patience pays. The biggest mistake in search term work is acting on too little data. The second biggest is never acting at all. Watch is the honest middle: you have seen the term, you have a threshold, and you will decide when it crosses one.

Keep a short list of watch terms from week to week. When one finally gets its second or third order, it graduates to harvest. When it crosses your negate threshold with nothing to show, it graduates to negate. Nothing stays in watch forever.

What to do this week

Block 45 minutes on the calendar. Pull the Sponsored Products search term report for the last 60 days, sort by spend, and work the top 50 rows only. Put every one into harvest, negate, promote, or watch. Make the obvious moves, the clear winners and the clear losers, before you agonize over the gray middle.

Then make it a standing weekly appointment. The strategists who win on Amazon are not smarter readers of the report. They read it on a schedule, with the four buckets in mind, and let small disciplined moves compound. If your ACoS is creeping while sales stay flat, the answer is almost always sitting in rows you have not read. The companion read on lowering ACoS without killing your sales shows how these weekly decisions add up to a number your accountant will notice.