When to Pause a Keyword in Amazon Ads: How Many Clicks Is Enough

Pause a keyword too early and you throw away one that would have sold. Pause it too late and you keep paying for clicks that never will. Here is how many clicks book keywords actually took to make their first sale, from 43,904 clicks in our study of 2,846 keywords, and a rule you can use on your own.

How many clicks the first sale took

Of the keywords that sold at all, half had made their first sale by the third click, three in four by the seventh and nine in ten by the thirteenth. These were exact-match keywords made from searches that had already sold, so they converted well — about 16–20% of clicks became orders. A broad keyword or a product-page target usually converts less often, and needs more clicks before it can be judged.

A rule for any keyword

The rule comes from simple probability. If each click has a 10% chance of becoming an order, a keyword with no sales after 10 clicks has been unlucky about a third of the time:

Use your own conversion rate if you know it (orders ÷ clicks, over at least a hundred clicks for the book). If you don’t, use the money version instead:

If it sells, but above break-even

A keyword that sells at an ACOS above your break-even is a different decision. The usual advice is to lower its bid. In our data that rarely worked: of 197 keywords above target whose bids were cut, two got under target, and the ones that kept selling improved only from 64% to 60%. A lower bid mostly bought fewer clicks.

So decide instead of trimming. Keep it if the sales are worth more to you than the margin — a launch, a series opener, page reads — or pause it. What does lower ACOS.

If it gets no clicks at all

A keyword with impressions and no clicks costs nothing. In our study, 82 of 93 live keywords that had never been clicked were already bidding more than the search had paid per click when it converted. The search is simply rare. Leave it running, and don’t keep raising the bid to force it.

Pause keywords, not campaigns of untested keywords

The most expensive pause is the one nobody sees: a whole campaign switched off two weeks after launch, with most of its keywords still untested. In our data, 369 keywords that had sold as search terms never got a single click because their campaign was paused after a median of 14 days. If you need to cut spend, pause the keywords that have had their test and failed it.

A pause checklist

1. Ignore the last three days. Sales are credited to clicks from up to 14 days earlier, and recent days are still filling in.
2. Has it had its test? About 3 ÷ conversion rate clicks, or one royalty of spend.
3. No sale after the test: pause it, or add it as a negative where it came from.
4. Sold, but above break-even: keep or pause on purpose; don’t expect a lower bid to fix it.
5. No clicks: leave it alone.

Common questions

How many clicks should I wait before pausing an Amazon Ads keyword?

About three divided by your conversion rate: 18 clicks at 17% conversion, 30 at 10%, 60 at 5%. In a study of 2,846 book keywords, 71% of those reaching 10 clicks had sold by then, and 90% of those reaching 20.

My Amazon ad has clicks but no sales. What should I do?

Check whether it has had a fair test first. At a 10% conversion rate, 10 clicks without a sale happens by luck about a third of the time. Past about 3 ÷ conversion rate clicks, or one royalty’s worth of spend, pause it or add it as a negative.

Should I lower the bid instead of pausing?

For a keyword with no sales after a fair test, lowering the bid just loses more slowly. For one that sells above break-even, a lower bid usually cuts its sales more than its ACOS. Make a keep-or-pause decision.

Should I pause a keyword with impressions but no clicks?

There is no need. It costs nothing until it is clicked. A keyword with no clicks is usually a rare search, not a low bid.