Prompt of the day

Channel plan red-team: find the failure before the spend does

The channel plan is drafted and everyone likes it — which is exactly when it should be red-teamed. Run the adversarial pass before the money moves.

You are a skeptical marketing consultant hired to find the flaws in a campaign plan before launch. I will paste the plan. Produce:

A) FAILURE MODES ranked by expected cost, each tied to a specific line of my plan with its evidence status (shown in my inputs vs hypothesis): audience-channel mismatch, budget too thin to clear an effective frequency on a channel, overlapping audiences paying twice for the same person, measurement gaps that will make results unreadable after the fact, and creative-format mismatch with placement.
B) The THREE plan changes with the best cost-to-risk-reduction trade, argued from my numbers.
C) A PRE-LAUNCH CHECKLIST: tracking verified end-to-end, naming conventions, holdout or baseline agreed, and budget-pacing alerts phrased as decision rules.

My plan: [PASTE: channels, budget split, flighting, audiences per channel, objective, creative formats, measurement setup]

Rules: Do not invent, estimate, or fabricate any statistic, benchmark, or performance figure — if a number is not in the material I give you, write "not provided" and flag it. Mark every claim I should verify in my analytics or source systems before it is published or presented. Never include customer personally identifiable information or client-confidential terms.

Why it works — Plans fail on things nobody wrote down: too little weight to matter on a channel, two line items buying the same user, a KPI that will be unreadable post-launch. A ranked failure-mode read costs one prompt; discovering the same list in week three costs the quarter.

Catch AI's mistakes by asking what it assumed

Confident-but-wrong is the #1 AI risk at work. Surfacing its hidden assumptions is a 10-second insurance policy on every important output.

After any answer you'll act on, ask: 'What did you assume that I didn't tell you? What would change your answer if it were wrong?'

1. Get the answer.
2. Ask for its assumptions + what would flip them.
3. Verify the load-bearing ones.