Nobody wants to keep spending on something that does not work, and nobody wants to stop something that was about to work. The way through is to decide the stop rules before you spend, when you are calm. A rule written during a bad week is usually a bad rule.

Good reasons to stop

  1. Tracking is broken. You are spending blind. Pause until it is fixed, or you will make decisions on false numbers.
  2. The offer has ended. An event has sold out, a date has passed, a promotion is over.
  3. The page or form is failing. Sending paid traffic to something that does not work burns money at full speed.
  4. The cost is well beyond what you agreed, with enough data. Define both "well beyond" and "enough" beforehand.
  5. The audience is exhausted. Frequency is high, results are falling and fresh creative does not lift them.

Bad reasons to stop

  • One poor day. Results fluctuate. A day tells you very little unless you have a large volume.
  • Impatience during early delivery. The platform needs time and events to find its audience. Turning a campaign off in the first days often means you never see what it could do.
  • A bad week that has a known cause, such as a public holiday, a site outage or a weather event.

Write the rules down

Before launch, agree in a line or two: how long you will wait, how many results you need to see, what cost you will tolerate and what you will do if the number is worse. When the moment comes, the decision is already made.

How to stop well

  • Pause, do not delete. The history is useful.
  • Record why you stopped and what you think you learned.
  • Note what you would change if you tried again: the offer, the page, the audience or the creative.
  • Do not pour the released budget into something else without a reason. Sometimes the right decision is to spend nothing until the problem upstream is fixed.

Examples of stop rules

  • If, after the agreed test period and the agreed number of results, cost per enquiry is still more than the level we set at the start, we pause and diagnose.
  • If the conversion count is zero while spend continues for a set number of days, we check tracking and the page before anything else.
  • If frequency has risen for two weeks and results have fallen, we introduce new creative or pause.
  • If the offer has ended, the campaign ends the same day.

Telling the people who care

When you stop a campaign, tell the people who depend on it, such as sales, a partner or a client, and explain why. A stop with a reason and a plan for what comes next reads as good management. A silent stop reads as a problem.

The trap of money already spent

People often continue a campaign because they have already spent a lot on it. That money is gone whether you continue or not. The only useful question is whether the next rupee will do better here than elsewhere. Asking it that way makes stopping easier when it is right.

Restarting with a better plan

A stopped campaign is not a failed strategy. Often it is the end of a test. Before you restart, change something that the evidence points to: the offer, the page, the audience or the creative. Restarting the same thing usually gives the same result.

Stopping versus fixing

A campaign that fails because of a broken page should be fixed and restarted, not abandoned. A campaign that fails because nobody wants the offer should be stopped. The difference is in the diagnosis, and a short diagnosis is often worth more than another week of spend. To decide how long is long enough, read how much data before judging an ad.

Keep reading