How often should you refresh Meta ad creative?

The short answer

There is no fixed number of days. Refresh a Meta ad when its results start to slip, not when the calendar says so. In practice, check cost per result every week, add genuinely new ads as soon as Meta flags an ad as tiring, and always keep two or three fresh concepts ready so you are never starting from zero.

Why "every two weeks" is the wrong question

Ads don't expire on a date. They tire when the people seeing them stop responding. How fast that happens depends on your budget, how big your audience is and how many different ads you are running. A small local audience with one ad will tire fast. A broad audience with ten very different ads can run for months.

So a fixed schedule either throws away ads that still work, or keeps running ads that stopped working weeks ago.

The signals Meta gives you

Meta shows two warnings in the Delivery column of Ads Manager:

• Creative Limited: the ad's cost per result is higher than it has been in the past, but less than double.
• Creative Fatigue: the ad's cost per result is at least twice as high as it has been in the past.

Meta does not explain exactly how it decides when a drop is caused by people seeing the ad too often, so treat these as early warnings, not the whole story.

Two other signals worth watching:

1. Cost per result rising week on week while your offer, landing page and budget have not changed.
2. Click-through rate falling for the same ad over time.

High frequency on its own is not a problem. If people keep seeing your ad and keep buying, leave it alone.

Why the number of different ads matters more than the date

In April 2025 Meta wrote that, with its newer ad delivery system, "the focus has shifted from niche targeting to creative diversification as the best lever." In plain English: Meta finds the right people for you based on your ads, so the ads themselves need to be different enough to reach different people.

That means a refresh is not the same ad in a new colour. It is a new angle, a new hook, a new format or a new type of customer in the ad.

A simple refresh routine

Step 1: Check weekly. Once a week, look at cost per result and click-through rate for each ad over the last 7 days compared with the previous 7.

Step 2: Act on the warnings. When an ad shows Creative Limited or Creative Fatigue, or its cost per result has clearly climbed, add new ads to that ad set rather than tweaking the old one.

Step 3: Make the new ads genuinely different. Change at least one of: the problem you lead with, the opening line or first three seconds, the format (video, image, carousel) or the person shown.

Step 4: Batch your changes. Meta's delivery system needs data to learn; its documentation describes a learning phase of about 50 optimisation events within seven days. Make changes in one go each week rather than editing every day.

Step 5: Keep a bench. Have two or three ready-to-go concepts at all times. The worst time to make new ads is when results have already dropped and you are under pressure.

Step 6: Retire, don't delete. Pause ads that have run their course and keep a note of what each one tested. Old winners often work again later with a fresh opening.

How many new ads per month?

It depends on spend and audience size, so there is no honest single number. A useful rule of thumb: if you are regularly seeing fatigue warnings, you need more new concepts each month; if your best ads run for months without slipping, you can refresh less often.

If keeping up with new ads is the bottleneck, that is what a fresh ads every month setup is for.

Source

• Jon Loomer Digital, Creative Fatigue: What It Is and How to Prevent It (Creative Limited and Creative Fatigue definitions in the Delivery column; frequency alone is not the problem).
• Jon Loomer Digital, Meta Andromeda: What It Means for Your Ad Strategy (quotes Meta's April 2025 post on creative diversification).
• Jon Loomer Digital, Is the Learning Phase 10 or 50 Events? (Meta's documentation: about 50 events in seven days).