How to find wasted ad spend on Amazon
Most Amazon ad accounts leak money in the same few places. Here's how to find the leaks in about 15 minutes, with math you can check.
Short answer
Wasted ad spend is money spent on clicks that won't turn into profitable orders. To find it, work out your break-even ACoS, pull the Sponsored Products search term report for the last 30 to 60 days, flag terms with enough clicks and zero orders, flag terms whose ACoS is above break-even, then negate, lower bids or move each term to its own exact-match keyword.
On this page
What counts as wasted spend
Wasted spend isn't “any click that didn't buy.” Some clicks never buy; that's advertising. Wasted spend is money on clicks that can't turn into profitable orders:
- Irrelevant search terms. Someone searching “dog bed pattern” clicking your finished dog bed.
- Terms with plenty of clicks and no orders. The shopper wants something you don't sell.
- Terms that convert, but at a loss. Orders come in, but each one costs more in ads than it earns.
The first two are easy to see once you look. The third hides behind a nice-looking sales number, so you need one number before you start.
Step 1: work out your break-even ACoS
ACoS (advertising cost of sales) is ad spend divided by ad sales. Your break-even ACoS is your profit margin before ad spend. Above it, every ad-driven sale loses money.
| Line | Per unit |
|---|---|
| Price | $30.00 |
| Amazon fees (referral + FBA) | −$10.00 |
| Cost of goods and shipping in | −$8.00 |
| Profit before ads | $12.00 |
| Break-even ACoS | $12 ÷ $30 = 40% |
That's an example. Use your own numbers per product, because a 25% ACoS is great on one ASIN and a loss on another. Profit per ASIN shows how to get the inputs.
Step 2: pull the search term report
In the ads console, create a Sponsored Products Search term report under sponsored ads reports, for the last 30 to 60 days. Shorter windows don't have enough clicks to judge; longer ones mix in old seasons. Amazon only keeps search term data for a limited window, so make this a habit, not an annual event.
Or skip the export:
Ask your AI
Pull my Sponsored Products search term report for the last 60 days. List every search term with spend over $20 and zero orders, sorted by spend, with clicks and campaign.
Step 3: flag terms with clicks and no orders
How many clicks with no order is “enough”? Use your conversion rate. If 1 in 10 clicks usually buys, a term with 20 clicks and no orders is a likely dud: if it truly converted at 10%, you'd see zero orders in 20 clicks only about 12% of the time.
A simple rule:
- Threshold clicks = 2 ÷ your conversion rate. At 10%, that's 20 clicks. At 5%, it's 40.
- Below the threshold, wait. Above it with zero orders, act.
| Search term (example) | Clicks | Spend | Orders | Verdict |
|---|---|---|---|---|
| dog bed pattern | 34 | $41.80 | 0 | Negate: wrong intent |
| dog bed large washable | 52 | $63.40 | 9 | Keep: converting |
| cat tree | 23 | $27.10 | 0 | Negate: wrong product |
| orthopedic dog bed xl | 12 | $15.90 | 0 | Wait: not enough clicks |
Step 4: flag terms above break-even ACoS
Now sort the terms that *do* convert by ACoS. Anything above your break-even is losing money on every order it drives. Don't negate these by reflex. They're relevant, they're just too expensive:
- Lower the bid on the keyword or target that matched it.
- Move it to exact match in its own campaign so you control its bid separately.
- Keep it on purpose if it's a launch or ranking push and you know what you're paying for.
Step 5: check auto campaigns and product targets
Auto campaigns and ASIN targeting leak quietly because nobody chose the terms. Look at the targets report for product targets with clicks and no orders, and at the auto-targeting groups (close match, loose match, substitutes, complements). Loose match and complements tend to be where relevance gets thin.
Step 6: decide, then repeat weekly
| What you found | What to do |
|---|---|
| Irrelevant term, any clicks | Add as a negative (exact or phrase) |
| Relevant term, clicks above threshold, no orders | Lower the bid, or negate if it never converts |
| Converting term above break-even ACoS | Lower the bid or move to its own exact keyword |
| Converting term below break-even | Harvest it: add as an exact keyword, negate it in the source campaign |
Negative keywords covers which match type to use for each.
What it's worth
Put your own numbers in. The value is the waste you find plus the hours you stop spending on spreadsheets:
At your numbers
25×
Downpour pays for itself 25 times over, every month.
Wasted spend to find
$3,000/moHours back
26 h/moTotal value
$4,950/moValue is the wasted spend WavyCloud can surface plus your hours back. Cutting the waste is still your call, and the 30-day guarantee covers the rest.
Frequently asked questions
What is a good ACoS on Amazon?
There's no universal good ACoS. A good ACoS is one below your break-even ACoS for that product, which is your profit margin before ad spend. Many sellers also watch TACoS, ad spend divided by total sales, to see whether ads are lifting organic sales over time.
How often should I check for wasted ad spend?
Weekly for active campaigns. Search terms change with seasons and competitors, and a term can burn a lot of budget in a few weeks. A weekly 15-minute review beats a quarterly cleanup.
Should I negate a search term after one or two clicks?
Only if it's clearly irrelevant, like the wrong product. For relevant terms, wait until clicks pass about 2 divided by your conversion rate. Negating too early cuts terms that would have converted.
Can AI find wasted ad spend for me?
Yes, if it can read your live ads data. Connected through an MCP server like WavyCloud, an AI app can pull your search term report, apply your thresholds and propose negatives. You still approve every change before it reaches Amazon.