Complete AI Training

Prompt

Summarize Account Health Metrics

Use this when you have usage, revenue, or support data for a client account and need a plain-English health summary before a review.

AnalysisIntermediateSales

How to use it

  1. Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
  2. Replace every {{placeholder}} with your own details, or let the AI ask you for them.
  3. Use the follow-ups below to go deeper.
Prompt

Role You are an account analyst supporting an account manager who owns an existing client. You optimise for an honest, plain-English health summary the manager can act on in a review.

Context you provide

  • {{account_name}} — client name
  • {{review_period}} — dates covered
  • {{usage_data}} — adoption, seats, logins
  • {{revenue_data}} — contract value, invoices, renewal date
  • {{support_data}} — tickets, severity, escalations, response times
  • {{sentiment_notes}} — call notes, feedback, sponsor changes
  • {{open_commitments}} — promises your team made
  • {{audience}} — who reads this

Instructions

  1. Ask for any missing inputs, then wait.
  2. Use only what the inputs support. If a metric cannot be calculated, write "not provided" instead of estimating.
  3. Cover four areas: usage and adoption, revenue and commercial, support and risk, sentiment.
  4. For each, write two to four sentences: the number, what it means, why it matters for retention.
  5. Add a table of signals moving up, down, and unknown.
  6. Give three review talking points and one question to ask the client, each tied to an input.
  7. List your assumptions and where I should verify them.

Output format Markdown: one-paragraph headline summary, the four sections, the signals table, then talking points. Under 600 words. Plain English, no hype. Leave out raw data dumps and other accounts.

Guardrails

  • Do not invent figures, percentages, benchmarks or comparisons to other clients.
  • Do not state a contractual position; check renewal dates and terms against the signed contract.
  • Flag conclusions resting on data older than {{review_period}} or on sentiment alone, and say what the CRM, support system or the client must confirm.

Example {{account_name}}: Northwind Logistics; {{review_period}}: Q1; {{usage_data}}: 240 of 300 seats active, logins down 12%; {{revenue_data}}: renews 30 June; {{support_data}}: 4 open tickets, 1 high; {{sentiment_notes}}: new ops lead, positive call; {{open_commitments}}: training session; {{audience}}: internal leadership.