Guide
Not on a calendar. Swap it when frequency and cost climb together
Change it when frequency climbs and cost per result climbs with it. Not on a schedule somebody wrote in an article. For most small accounts that lands between two and six weeks per creative. This page covers the two numbers that trigger a swap, why your cadence is set by supply, and what a replacement costs. By the end you will know how often your own account can actually refresh.
- Numbers that trigger a change
- 2
- Where most small accounts land
- 2-6 wks
- Replacement that should already exist
- 1
A calendar cadence throws away winners and keeps losers on schedule
Swapping creative every two weeks regardless of performance makes two expensive mistakes at once.
It retires ads that were still working, because a calendar cannot see cost per result. It also keeps ads that were never working, because they have not hit the date yet.
The signal is available and it is two numbers. Frequency, which is impressions per person. And cost per result. Read together they say whether the creative is the cause.
Rising frequency with rising cost means the pool of people seeing it for the first time is shrinking. That is when a replacement is due.
Flat frequency with rising cost means something else moved. A new ad will cost a fortnight and fix nothing.
Calendar cadence against signal cadence
The right-hand column costs nothing extra to run. It requires only that somebody reads two numbers once a week.
| Change on a schedule | Change on the signal | |
|---|---|---|
| An ad still winning at week two | Retired anyway | Kept until frequency bites |
| An ad losing on day three | Runs another eleven days | Off, and logged |
| What it demands of you | A reminder | Two numbers, once a week |
| What it demands of production | A fixed count, regardless | A replacement ready when it triggers |
Your cadence is set by supply, and that number is the one to work on
Published benchmarks put the winner share at 5 to 8 percent, per Motion's analysis of 550,000+ Meta ads. That is roughly fourteen tested creatives per winner.
So a change of creative is not one video. It is one draw from a pool where most draws lose. The replacement has to come out of that same pool.
A route producing three tested creatives a month finds a winner every four or five months. Fatigue arrives faster than that in most accounts.
That mismatch is the actual cause of the panic that gets called a creative problem. The ad was always going to decline. The replacement was not ready.
Compare production routes on turnaround, revision rounds, cost per edit and consistency. Those four decide the count you can sustain rather than the count you can plan.
Fourteen tested creatives, one expected winner
At the middle of the published 5 to 8 percent range. Your replacement interval cannot be shorter than the time it takes to draw from this grid.
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
Refresh the entrance first, because it is a new ad for a fraction of the cost
When hook rate falls and hold rate stays healthy, the opening went stale before the argument did. Most of the audience only ever saw the opening.
A different first three seconds against the same body reaches those people as a different ad. It costs an export rather than a shoot.
This is why recording several openings in one session is worth more than it looks. One recording, several entrances, and a replacement already sitting there when the signal triggers.
In Cutroom the change is a transcript action rather than a re-edit. Trim the opening. Delete a line and the cut rebuilds. Change the pace and it re-cuts. Then export at 20 credits per output minute.
A batch is 100 credits per take. Further versions of an existing take cost 10 credits for a thirty second cut. That is the difference between refreshing weekly and arguing about it monthly.
Six shapes, not six rewordings. A contradiction, a named audience, a specific number, a demonstration, a confession, a question the viewer already asked.
What a replacement costs, by how much you change
Only the first line needs a camera. The other two are why a weekly refresh is a production question rather than a budget one.
- New take, batch and export110 CR
- New opening from an existing take10 CR
- New pace or length, same take10 CR
Keep the winner running while the replacement is tested, always
The most common operational mistake is switching creative in one move. The old ad goes off, the new one goes on, and the account loses its reference point.
Run the replacement alongside for a week in the same ad set. Same audience, same offer, one variable. Let the account decide rather than the calendar.
Retire the old creative when the new one beats it at the volume you named in advance, and not before.
Log every swap in one line. What changed, what won, by how much, at what volume. Six months of that log is worth more than any published cadence.
Write the verdict rule before launch. Without one, the winner is chosen by whoever opens the dashboard first and whichever metric flatters their preference.
Build your own interval from six dates, then refresh on the signal
One habit costs nothing and pays for years. Note the date each creative went live. A replacement interval you can quote is worth more than any published cadence.
Six months of those dates is your own benchmark. It will not match anybody else's, and that is the point.
Read a rolling week against the previous one. On a small budget, daily numbers are mostly noise, and swapping creative on noise is worse than leaving it alone.
The route needs one input from you. Somebody willing to record for twenty minutes. That session produces several entrances, which is a month of replacements.
Everything after that session is priced rather than scheduled. A batch is 100 credits per take. Each further version is a 10 credit export for a thirty second cut.
So the cadence stops being a production question. When the two numbers turn, the replacement already exists and going live takes minutes.
Questions people ask
- Is there a frequency number that means swap it now?
- No universal one, because tolerance varies by audience size, offer and category. Watch the pair instead: when frequency climbs and cost per result climbs with it, the creative is done whatever the frequency figure reads.
- Should I change everything at once or one ad at a time?
- One at a time, in the same ad set as the incumbent. Changing everything at once removes your reference point, so a bad week afterwards cannot be attributed to anything and the whole exercise teaches you nothing.
- Does the whole ad need replacing, or only the hook?
- Check hold rate. If hold is healthy and hook rate fell, the entrance is stale and a new opening against the same body is enough. If both fell, the argument is tired and the replacement needs new material.
- Is a refresh cadence right for every account?
- It needs a week of spend that is readable, and somebody willing to record. Below that, the cadence exists on paper only. Spend the time on the offer and start when the constraint has moved.
Let the two numbers decide the date. Cutroom makes sure the replacement already exists, because a new entrance is a 10 credit export rather than a production booking.