Prompt Experiment Readout for Media

The Prompt

You are an experimentation analyst. Turn media prompt test results into a decision-ready readout.

EXPERIMENT GOAL:
{{experiment_goal}}

VARIANTS TESTED:
{{variants_tested}}

EVALUATION DATA:
{{evaluation_data}}

OBSERVED FAILURES:
{{observed_failures}}

DECISION DEADLINE:
{{decision_deadline}}

NEXT ITERATION CONSTRAINTS:
{{next_iteration_constraints}}

Return:
1. Executive summary (what worked, what did not, what to do next).
2. Variant comparison table with signal confidence notes.
3. Failure pattern analysis and likely root causes.
4. Recommended next test plan (max 3 focused experiments).
5. Fallback plan for teams with sparse data.

Rules:
- Keep conclusions tied to evidence.
- Distinguish strong signals from weak/ambiguous signals.
- Stay tool-agnostic and avoid vendor-specific assumptions.

When to Use

Use this after running prompt trials for generated images, videos, or audio outputs. It helps teams avoid endless tweaking and move to clear next-step decisions.

Variables

• experiment_goal: Core hypothesis the test aimed to evaluate.
• variants_tested: Prompt variants and setup details.
• evaluation_data: Scores, reviewer notes, acceptance rates, timing, cost, etc.
• observed_failures: Recurrent quality or compliance issues.
• decision_deadline: When the team must choose a direction.
• next_iteration_constraints: Budget, timeline, or resource limits.

Tips & Variations

• Ask for “minimum viable next test” when time is limited.
• Include confidence labels (high/medium/low) to prevent overclaiming.
• For small teams, request a one-page readout version with top 3 actions only.
• If data quality is poor, ask for additional instrumentation recommendations.

Example Output

A useful output contains a concise decision summary, evidence-backed variant table, failure clusters, and a short prioritized experiment plan that the team can execute immediately.