How long should I let an ad run before killing it?
Do not decide how long an ad should run from the calendar alone. Judge it against spend, impressions, conversion volume and the decision you are trying to make.
- 01A campaign that has barely spent anything has not had the same opportunity to prove itself as one that has spent…
- 02At the same time, you do not need to keep funding an obviously broken ad just because it has not reached a platform…
- 03If the failure is already clear high in the funnel, more time may simply buy more evidence of the same problem
- 04Platform “learning phases” can matter, but they are not a reason to switch your brain off
A campaign that has barely spent anything has not had the same opportunity to prove itself as one that has spent several times your acceptable customer-acquisition cost.
At the same time, you do not need to keep funding an obviously broken ad just because it has not reached a platform benchmark.
Watch the chain. Is the ad getting shown at a reasonable cost? Are people stopping and clicking? Are the right people arriving? Are they taking the next meaningful action?
If the failure is already clear high in the funnel, more time may simply buy more evidence of the same problem.
Platform “learning phases” can matter, but they are not a reason to switch your brain off. Modern ad systems change their optimization rules, and there is no timeless 50-conversion rule that should override your economics.
Let the ad run long enough to collect useful evidence, not long enough to satisfy a ritual.