Multimedia Audience Signal Miner

The Prompt

You are an audience research analyst. Extract actionable audience signals from mixed-format content evidence.

AUDIENCE SEGMENT:
{{audience_segment}}

SOURCE MATERIAL:
{{source_material}}

CONTENT FORMATS:
{{content_formats}}

BUSINESS GOAL:
{{business_goal}}

TIME HORIZON:
{{time_horizon}}

CONSTRAINTS:
{{constraints}}

Return:
1. Signal map: recurring themes, unmet needs, and friction points.
2. Format-specific findings (image/video/audio/text behavior differences).
3. Opportunity hypotheses ranked by confidence and impact.
4. Suggested experiments for content direction validation.
5. Fallback approach if only limited evidence is available.

Rules:
- Keep claims tied to provided evidence.
- Separate observation from interpretation.
- Avoid tool-specific assumptions.

When to Use

Use this when planning multimedia content and you need more than intuition about audience behavior. It helps turn scattered evidence into prioritized hypotheses and testable next actions.

Variables

• audience_segment: The group you are targeting.
• source_material: Comments, transcripts, analytics snippets, briefs, or notes.
• content_formats: Which media types are in scope.
• business_goal: Why this research matters operationally.
• time_horizon: Near-term campaign vs longer-term strategy.
• constraints: Data availability, compliance, brand, and timing constraints.

Tips & Variations

• Ask for separate “quick wins” and “strategic bets” outputs.
• Request contradictory signals explicitly to avoid one-sided narratives.
• If data is noisy, ask for a confidence-weighted synthesis only.
• For low-capability models, split work into two passes: extraction first, interpretation second.

Example Output

A solid output includes a clear signal map, confidence-tagged opportunities, and a practical experiment list that teams can use to refine multimedia strategy.