Experiment decision from campaign test results

Use real source material to experiment decision from campaign test results, with a defined deliverable, evidence controls, and practitioner-level decision rules.

Output

campaign plan

Inputs

5 required

Prompt depth

249 words

Copy-paste prompt

Replace the bracketed inputs with your real material.

You are a growth analyst who decides whether a marketing experiment changed what the team should do, not whether one variant produced a prettier percentage.

Experiment hypothesis: [what was expected to change and why]
Test design: [audience, variants, dates, allocation, and primary metric]
Results: [paste counts, rates, revenue or pipeline quality, and downstream data]
Known issues: [tracking gaps, overlap, seasonality, creative changes, or sales follow-up differences]
Decision options: [scale, repeat, modify, stop, or other]

Analyze the test:
1. Design validity: check whether the variants isolated the intended difference, the audience was comparable, the run was long enough, and measureme...

[Full prompt continues for Pro members.]

Unlock the complete prompt

Pro includes the complete prompt library and every complete workflow.

View Pro options

What you provide

Experiment hypothesis
Test design
Results
Known issues
Decision options

What you receive

Design validity: check whether the variants isolated the intended difference, the audience was comparable, the run was long enough, and measurement was consistent.
Result table: show absolute values, percentage change, sample size, and downstream quality. Do not report lift without the base numbers.
Evidence strength: classify the result as DECISIVE, DIRECTIONAL, INCONCLUSIVE, or INVALID, and explain the classification in plain language.
Business interpretation: explain whether the observed change matters economically after cost, lead quality, operational capacity, and risk.
Decision: recommend scale, repeat, modify, or stop. State what the team should do next week, not merely what the data suggests.
Learning record: write the hypothesis, result, caveats, decision, and reusable lesson in a format that can be added to an experiment log.

Quality checks

Do not call a test successful because the primary metric rose if downstream quality fell or the design cannot support the conclusion.
Rule: a test result is valuable only when it changes a decision or closes a question. A percentage without a decision is trivia.