Live slots over healthy span, over the share that survives

First, how long your creatives stay healthy. Plot performance against days since launch across your last dozen. Read off the point where the median one has decayed halfway towards your kill threshold.

Second, how many healthy creatives your spend needs live at once. For most accounts that is three to five per ad set, which lets the delivery system choose without starving each candidate.

The first is an afternoon with a spreadsheet. The second is a division you can do in your head.

Those two give you a maintenance rate. Four live, each lasting three weeks, means about four new creatives every three weeks to hold position.

That is maintenance before any growth or experimentation. Work out your own number before reading on. If it already exceeds what you shipped last month, the rest of this page is about your account.

Maintenance arithmetic assumes every creative you produce becomes a healthy live creative, which is not remotely true.

Published benchmarks put the winner share at 5 to 8 percent, per Motion's analysis of 550,000+ Meta ads. Use a looser bar than winner if you like.

Even on a looser bar, most teams find well under half of what they produce is still running after two weeks.

So take the maintenance number and divide by your own survival rate. Four replacements needed at one in three surviving means producing twelve.

Now picture twelve a month where each video needs a brief, a shoot slot, an editor and two revision rounds. That is not a plan. It is why the schedule quietly slips every quarter.

Frequency climbing past your historical tolerance while the three-second view rate slides below the creative's own baseline. That is exposure fatigue arriving with about a week of notice.

A budget increase. Spending more reaches further into the audience. Creative that worked on the easiest segment often does not travel. Plan new creative alongside a spend increase rather than after it disappoints.

A seasonal shift in the auction. When costs rise across the category, weak creative gets exposed first. A roster you were comfortable with becomes marginal.

The signal that should push a refresh back gets ignored constantly. A creative comfortably under your cost threshold should keep running, whatever the calendar says.

Pausing a performing ad to obey a schedule is throwing money away to feel organised.

Set the review as a question rather than a date. Is this creative still under threshold, and has frequency passed the level where your last several started sliding. Two answers, thirty seconds, no calendar involved.

Producing an entirely new concept is the expensive form of refresh. Exhaust the cheap form first. Recognition happens in the opening frames, and the opening is the cheapest part to change.

A new opening on the same body. A shorter cut. A different caption style. Music on where it was off. Different footage over the same phrases.

Each can reset attention on a proven body for a fraction of a new production. One of them usually buys another fortnight from a creative you already trust.

Work down that list in order. The opening is the cheapest lever, and the one that moves the most.

Log which refresh you used and what it bought. After two quarters you will know which lever works on your audience and stop guessing in the Monday meeting.

That workflow is what Cutroom is built around. One take, up to three minutes, becomes a finished 9:16 MP4. Every cheap refresh is an adjustment on the transcript.

Trim the opening. Delete a line and the cut rebuilds around it. Change the pace and the whole piece re-cuts. Swap the footage over a phrase. Turn music on for 20 credits.

Each of those exports at 20 credits per output minute. Nowhere else is the fifth refresh of a proven body priced like the first one.

The ceiling is worth knowing before you plan around it. Once several openings have run, the body itself has been consumed. Restyling then buys nothing, and the account needs a new recording.

The other facts to plan around. Somebody at your company records the take. Three minutes is the upload ceiling. Every export is a 9:16 MP4, and there is no timeline underneath.

So the sequence is short. Measure the span. Do the division. Exhaust the cheap refreshes. Then record again, and have the replacement live the same day.