Pre-mortem a campaign before you spend

Here's a campaign I'm about to launch: {{describe channel, offer, budget, timeline, target}}.

Run a pre-mortem. It's 60 days later and the campaign clearly failed.

1. Write the three most likely failure stories in detail — what happened, in order, and what we saw in the dashboard while it was happening.
2. For each, name the assumption that turned out false.
3. Rank all assumptions by (damage if wrong) × (probability wrong) and mark which are testable before we spend real money.
4. For the top three, design a cheap test: what to run, budget, duration, and the number that would make us stop.
5. State the tripwires — specific metric thresholds and dates at which we kill or double down.
6. Name the one failure mode we could not detect with our current tracking, and what to instrument now.

Be blunt about which parts of my plan are hope.