Kill your ads on day fourteen

Underperforming ad accounts tend to share one problem, and it is not a creative problem. It is a decision-making problem. Somebody spent four weeks making that ad, so nobody wants to be the one to turn it off.

The fix is boring: make the kill a rule that fires automatically, so no person has to own the decision emotionally.

The rule worth running

• Every creative gets a fixed 14-day window and a stated success metric
• At day 14, it either beat the account baseline or it did not
• Losers are turned off the same day, no meeting, no defence
• Winners get four versions built within 72 hours
• The hypothesis and the outcome both go in a written log

The written log is the part people skip and the part that matters most in month six. Without it you re-test the same losing angle twice a year and call it a fresh idea.

Why fourteen days

Short enough that a bad ad cannot quietly eat a quarter of your budget. Long enough to clear the learning phase and get past a weekend anomaly. Under roughly $15K monthly spend, stretch it to twenty-one — the data simply is not there yet at fourteen.

You are not killing the ad. You are ending an experiment that already gave you its answer.

Once the rule is in place, something else changes: people stop over-investing in single concepts. If you know a piece has fourteen days to prove itself, you stop spending six weeks polishing it, and the whole production cadence speeds up.