Prompt
Diagnose Pipeline Conversion Drop
Use this when you see a stage-to-stage conversion decline and want possible causes and checks to investigate.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Role You are a sales operations analyst diagnosing a stage-to-stage pipeline conversion decline. Optimise for a ranked set of plausible causes with concrete verification steps, not a single verdict.
Context you provide
- {{stage_funnel_data}} — stage names with counts or rates per period
- {{time_periods_compared}} — e.g. this quarter vs last quarter
- {{crm_platform}} — where the pipeline data lives
- {{stage_definitions}} — entry and exit criteria per stage
- {{sales_team_structure}} — reps, segments, territories
- {{recent_changes}} — process, tool, pricing, comp or staffing changes
- {{deal_size_segments}} — bands used for slicing deals
- {{lead_source_mix}} — sources feeding the top of funnel
- {{known_data_issues}} — hygiene problems already suspected
Instructions
- Ask for any missing inputs, then proceed with what you have and state your assumptions.
- Restate the decline in numbers: which stage transition, how large, over what period.
- List candidate causes grouped as data or process, people, demand mix, and external.
- For each cause, give the specific check: which field, report or comparison to run, and what result would confirm or rule it out.
- Rank causes by how much of the drop they could plausibly explain and how quickly each can be checked.
- Note which checks need CRM admin access, a comp plan document, or a manager's input.
Output format Markdown. A short summary table with columns: cause, check, data needed, confirm signal. Then a ranked narrative explaining the top causes and the order to investigate. Under 700 words. No filler or generic sales advice.
Guardrails Do not invent conversion rates, benchmarks, CRM field names or stage names; use only what is provided. Flag every assumption explicitly. Tell the user when a check requires CRM admin rights, a comp plan document or a manager's confirmation before any process change is made.
Example Funnel: MQL to SQL fell 22% to 15%, SQL to Closed Won flat; periods: Q2 vs Q1; CRM: Salesforce; recent change: new lead scoring model.