Creative refresh: replace on decay, not on a calendar

Hook rate decaying against that creative's own first week. The opening stopped stopping people, which is the earliest signal available.

Cost per result rising while your other creatives hold steady. If everything rose together it is an auction or a seasonality problem.

Frequency climbing without a matching rise in results. The same people are seeing it again and not responding.

Any one of the three is worth investigating. Two together is a replacement decision.

A calendar rule replaces creatives that are still working and keeps ones that stopped. That is the worst of both outcomes.

Write your own thresholds down once, so the decision is a reading rather than an argument.

Put a number next to each signal. A twenty percent fall in hook rate is a threshold. A feeling that it seems tired is not.

A refresh schedule is meaningless without a queue. The queue is where most plans fail.

Published benchmarks put the winner share at roughly 5 to 8 percent, from Motion's analysis of 550,000+ Meta ads.

So each replacement you eventually run sits behind thirteen to twenty attempts that did not become one.

If you run four creatives and want to replace any of them within a fortnight, you need a steady production rate rather than an occasional burst.

One three minute take can hold four scripts recorded in one session. Each batches separately for 100 credits.

That habit turns the refresh schedule from a plan into something that happens on the week it is needed.

When hook rate is the signal that moved, the first two seconds died. Not the whole creative.

Re-cutting an existing project with a different opening is cheaper and faster than filming a new one.

Trim the opening. Write a different headline. Emphasise a different word so it pops in the captions.

Change the pace and the whole ad re-cuts. That alone produces a materially different creative from the same source take.

Switch the caption pack and the look changes without the words changing at all.

Each of those is a re-export at about 13 credits rather than a new shoot at a new fee.

Only when several openings have failed against the same body is the whole idea finished.

That order matters. Most teams replace the whole ad first and never find out the body was fine.

A filmed replacement is about 113 credits. That is 100 for the batch and roughly 13 to export forty seconds.

Re-exporting an existing project with a new opening is 20 credits per output minute, so about 13 credits.

Basic at $39.99 gives 2,500 credits a month. That is roughly twenty-one replacements, far more than most accounts need.

So the cost of refreshing is not the obstacle. The obstacle is somebody filming, and that cost does not appear on any invoice.

The cost of not refreshing is a rising cost per result on a creative that was working. It compounds quietly for weeks before anybody notices.

Set a weekly ten minute review of the three signals and the schedule becomes a habit rather than a project.

Put the review on the same day each week and give it to one person. Shared ownership of a recurring check reliably means nobody does it.

Ten minutes a week is cheaper than every alternative, including the one where a decaying creative runs for a month unnoticed.

Everything declined together. That is an auction, a season or a targeting change, and new creative will decline too.

The creative never worked. Refreshing a loser produces another loser. The fix is a different idea rather than a newer one.

The offer is the problem. No opening rescues a proposition the audience has already rejected.

It has not had enough impressions to read, in which case you are replacing on noise.

The placement needs a horizontal file, since every export here is 9:16.

Check those five first. More creative is the expensive answer to four of them.

Then build the habit that makes the sixth case cheap: a ten minute weekly look at the three signals, written into one person's week.

Most accounts discover a decayed creative weeks after it started decaying. Those weeks cost more than the whole production line does.

A schedule on a calendar and a review that happens every week are different objects. Only the second one saves money.