Pause an ad too early and you may kill a winner. Leave a weak ad running too long and you waste budget.
Pause an ad when you have enough real leads or sales to make a fair comparison and it's still clearly performing worse than other ads doing the same job.
There's no set amount an ad needs to spend before you can judge it. What matters is having enough actual leads or sales to know whether the difference you're seeing is real. Two or three conversions per ad usually aren't enough to reach statistical significance, which means you don't yet have enough data to rule out normal ups and downs. Keep the test running until the pattern is much clearer. It's also worth remembering that some ads help move people toward a later conversion without getting credit for it, so direct conversions shouldn't be the only reason an ad stays on or gets paused.
It's easy to look at two ads, see that one has cheaper clicks, and assume it's the better ad. But cheaper clicks don't necessarily mean cheaper leads or customers.
Cost per click (CPC) only tells you what you paid to get someone to the next step. It doesn't tell you whether they filled out a form, booked a call, or bought anything. For example:
If you only look at CPC, Ad A looks better. But Ad B is actually producing leads for less. If you pause Ad B because its clicks cost more, you have turned off the better-performing ad.
That's why leads, booked calls, and sales matter more than clicks when you're deciding which ad is actually working. Metrics like click-through rate and cost per click can help explain what may be happening, but they shouldn't decide the winner on their own. Our guide on which KPIs to measure for digital advertising covers which numbers matter for different types of campaigns.
There's no magic number of days or dollars an ad needs before you can judge it. You need enough actual leads or sales to tell whether one ad is genuinely performing better than another.
If your target cost per lead is $40, one ad might generate three leads while another generates two. That's not enough data to confidently say one is better. Keep the test running until you have enough conversions to reach statistical significance, meaning there's enough evidence to believe the difference is real and not just normal variation.
Google Ads' experiment tools use the same basic idea. They look at how much data you have and how large the difference is, then tell you when the result is statistically significant. In plain English, that means there's enough evidence that the difference probably isn't just random chance. Google also recommends giving the test enough time for delayed conversions to come in before making a decision.
Meta results can also be less stable early in a campaign, when ad sets are still in learning mode, so avoid making decisions based on small swings before the ads have had time to gather meaningful data.
An ad doesn't need to be perfect to stay on. The question is whether another ad doing the same job is clearly performing better. Once you have enough leads or sales to make a fair comparison, look at:
One thing that can make Meta reports misleading is how the platform distributes budget. Meta may spend more on the ad it believes can produce the most total results, even if that ad's average cost looks a little higher. As an ad gets more spend, it often becomes harder to maintain the same cost, so your strongest ad can actually look more expensive simply because Meta is asking it to do more of the work. This is sometimes called the breakdown effect.
That's why moving all of the budget to the ad that looks cheapest in the report can backfire. Look at how the whole campaign is performing, not just which individual ad has the lowest average cost.
Once you have enough data to make a fair comparison, there are a few good reasons to pause an ad:
Don't pause an ad just because it had a bad day, has a higher cost per click, or isn't getting many direct conversions (unless it's spending a disproportionate percentage of your ad spend). Make sure you have enough data to support the decision and consider the role the ad plays in the broader customer journey.
Don't eliminate an ad just because another one is ahead for a few days. Give the test enough time to collect a meaningful number of conversions. If the results are still close after that, there may not be a clear winner.
Cost per lead matters when an ad's job is to generate leads directly, but not every useful ad is the one that gets credit for the final conversion.
The Rule of 7 is an old marketing principle that says people often need to see or hear from a business several times before they take action. The number seven isn't magic, and the real number will vary. The point is that most customers don't see one ad and immediately buy. They usually need a few different touchpoints before they're ready.
That doesn't mean showing someone the exact same ad seven times. Those touches should come from different ads and different channels. One video may introduce the business, while another builds trust with a testimonial. Later, that person might search for the company on Google, see a retargeting ad, visit the website again, or open an email before finally converting.
Because of that, an ad can still be valuable even if it doesn't generate many conversions on its own. If you pause every awareness or educational ad just because its direct cost per lead is higher, you may remove the very ads that are helping people become comfortable enough to convert later.
Judge an ad based on the job it's supposed to do. If the goal is awareness or education, look at whether people are seeing the ad, watching it, engaging with it, and then taking other actions such as visiting the site or searching for the brand later. More importantly, look at what happens to the campaign as a whole. If your overall cost per customer improves or your return on ad spend (ROAS) goes up when those ads are running, they may be contributing even if they don't get credit for many conversions themselves.
For ads that are supposed to generate leads or sales directly, follow those leads all the way through the sales process. Look at which ones became appointments and customers, then compare the actual cost to acquire a customer. An ad that looks expensive based on cost per lead may turn out to be your most profitable once you see what happens after the lead comes in.
Before blaming the creative, check the rest of the process. If leads from every ad are going cold, the problem may be slow follow-up or the sales process rather than the ads themselves. Pausing creative won't fix that.
Once you have a clear winner, the next question is how to scale it. Our guide on how to scale ads without losing money covers how to raise budgets without moving too fast.
There's no universal number of days or dollars. Run the test until you have enough leads or sales to make a reliable comparison. If you're using Google Ads' built-in experiment tools, let the test run until the results are statistically significant and give delayed conversions time to come in. That can take several weeks or even months at low ad spend levels.
Not based on CPC alone. An ad with more expensive clicks can still produce cheaper leads if a higher percentage of those clicks convert.
It can. Meta considers adding a new ad a "significant edit," meaning a large enough change that the system may need to relearn how to deliver the ad set effectively.
Not automatically. Meta often puts more budget behind the ads it expects to perform best, so an ad with very little spend may simply not have enough data to judge yet. It may also be doing a different job, such as building awareness or helping warm people up before they convert through another ad.
Not necessarily. First ask what the ad is supposed to do. An awareness or educational ad may help introduce the business, while a retargeting ad may keep you in front of someone who has already shown interest. Either one can help a person convert later through another ad, a Google search, an email, or a sales conversation. Judge the ad by the role it plays and what happens to overall campaign performance, not just the conversions credited directly to it.
At Power Couch Media, we manage Meta and Google Ads and build creative specifically for testing. If you aren't sure which ads to keep, pause, or replace, we can look at the campaign data and help separate real performance differences from normal noise. If you want to talk through your account, get in touch with a strategy specialist.