No sales after two weeks: should you pause a new Amazon campaign?

You launch a new Sponsored Products campaign, it gets clicks, and two weeks later it still has no order. The usual advice goes two ways. One says Amazon needs a learning period, so wait. The other says it is wasting money, so pause it.

We checked what happened to new campaigns on real accounts. The answer is neither "wait" nor "pause". It is: count the clicks, not the days.

The measured numbers below come from the Sponsored Products campaign report of 10 advertisers in Europe, the UK and the US, day by day. One advertiser runs ads in two regions, so below we count 11 accounts, one per advertiser and region. We took the 295 campaigns that first appeared in the report between April 11 and August 28, 2026 and showed at least one impression, and followed each one for 42 days from its first impression. Day 0 is the day of the first impression. The last of these 42-day windows ended more than a week before the last available day, so the ad-attributed orders are close to final. All comparisons are made inside one account, so currencies never mix.

How long the first sale took

• 251 of the 295 campaigns got at least one click in the 42 days.
• 122 got at least one order. The other 173 got none in six weeks.

Among the 122 campaigns that sold, the first order came quickly in half of them, and late in a large part:

• 61 had their first order between day 0 and day 4.
• 75 within the first week, 91 within the first two weeks.
• 31 made their first sale on day 14 or later. That is one in four of the campaigns that sold at all.
• 15 on day 21 or later, 9 on day 28 or later.

So a "no sale in two weeks, pause it" rule would have stopped 31 of these 122 campaigns before their first order.

Most campaigns that never sold never got enough clicks

Of the 173 campaigns without an order, 44 never got a single click. The other 129 got clicks, but few: the median campaign among them had 5 clicks in six weeks.

Five clicks say almost nothing. If your product converts 5% of its clicks into orders, 5 clicks without an order is the most likely outcome: it happens 77% of the time even when the campaign is fine. These campaigns did not prove that they do not work. They did not get enough traffic to prove anything.

10 clicks and no sale in two weeks: what came next

37 campaigns had at least 10 clicks and no order in their first 14 days. These are the ones most people would pause.

• 19 kept running: they got at least 10 more clicks between day 14 and day 41.
• 11 of these 19 made at least one sale later. 8 did not.
• The other 18 got fewer than 10 more clicks. We cannot tell whether they were paused or simply stopped getting traffic.

Of the 11 late sellers, 4 brought at least 3 orders between day 14 and day 41. In 3 of them, the ACOS of that period was below the account's own Sponsored Products ACOS (11.5% against 25.9%, 6.3% against 25.9%, 35.7% against 38.1%). In one it was above (55.6% against 29.5%). The other 7 had one or two orders, too few for a reliable ACOS.

The pattern holds with other limits. With at least 5 clicks and no order in two weeks: 65 campaigns, 32 kept running (10 or more later clicks), 19 of those sold later. With at least 20 clicks: 23 campaigns, 10 kept running (20 or more later clicks), 7 of those sold later.

Waiting does not make a campaign cheaper

The "learning period" idea says a new campaign gets better after a couple of weeks. We compared days 0 to 13 with days 14 to 41 on campaigns that got clicks in both periods, account by account. Eleven accounts had such campaigns, and nine had orders in both periods.

• ACOS was lower in the later period in 3 of the 9, higher in 6.
• Cost per click was higher in the later period in 8 of the 11.
• Conversion rate was lower in the later period in 8 of the 11.

So in these accounts, time alone did not bring the cost down. A campaign that is expensive in its first two weeks, with enough clicks to judge, was usually not cheaper later. What time does bring is more clicks, and clicks are what you need to judge.

Count clicks, not days

The question is not "how many days without a sale" but "how many clicks without a sale". The answer depends on your conversion rate (orders divided by clicks). If a campaign converts like the rest of your account, this is the chance of seeing zero orders anyway:

A simple rule of thumb: divide 3 by your conversion rate. At 5% that is 60 clicks, at 3% it is 100, at 10% it is 30. A campaign that converted like the rest of your account would reach that many clicks with no order only about 5% of the time or less. So if yours does, that is strong evidence that it converts worse. Before that point, a pause is a guess.

This only tells you whether the campaign converts like your account. Whether that is good enough is a separate question: a campaign can sell and still lose money. For that, look at ACOS against your margin once it has orders.

What we could not see

We do not have the campaign status history, only the current status. So for the 18 campaigns that got few clicks after day 14, we cannot tell a pause from a lack of traffic. We took the first day in the report as the start of the campaign, which can miss a campaign that existed earlier without any report row. The report counts orders for 7 days after the click, and an order can be for another product of yours than the one in the ad. 9 of the 11 accounts advertise in Europe or the UK, 2 in the US. The 10, 5 and 20 click limits and the 14-day split are our choice, not a measurement.

Check it in your account in 15 minutes

1. Find your account's conversion rate for Sponsored Products over the last 60 days: orders divided by clicks.
2. Divide 3 by it. That is your click limit (5% gives 60).
3. List your campaigns that started in the last two months, with clicks, spend and orders since launch.
4. A campaign with orders: judge it by ACOS against your margin, not by this rule.
5. A campaign with no orders and fewer clicks than your limit: do not pause it yet. If it gets almost no clicks, the problem is traffic (bids, targets or relevance), not conversion.
6. A campaign with no orders that has reached your limit: now a pause or a change is a decision, not a guess. Before pausing, open its search term report and see which searches the clicks came from.

Two weeks is a date in the calendar, not a measurement. The number of clicks is.