Skill

Meeting to CRM Notes

Convert a customer call transcript into clean CRM-ready notes: summary, next steps, sentiment, and key fields to update.

Goal

Transform a raw customer call transcript into structured, CRM-ready notes including a concise summary, action items with owners, sentiment assessment, and a list of specific CRM fields to update.

Trigger

Invoke after any customer call when a transcript is available and CRM notes need to be logged.

Steps

  1. 1

    Analyze the transcript provided below: $transcript Read it fully before proceeding. Identify the customer name, company, participants, call date, and primary purpose of the call (e.g. discovery, renewal, support, QBR).

    • Customer name and company extracted
    • All participants and their roles identified
    • Call purpose clearly categorized
    $transcript
  2. 2

    Write a 3–5 sentence executive summary of the call. Cover: what was discussed, any problems or concerns raised, decisions made, and the overall outcome. Write in third-person, past tense, suitable for a CRM activity log.

    • Summary is 3–5 sentences, no more
    • Decisions and outcomes are explicitly stated
    • Tone is professional and neutral
  3. 3

    Extract all next steps and action items mentioned or implied. For each, assign an owner (customer or internal team member by name/role if mentioned), a clear action description, and a due date or timeframe if stated. If no due date was given, flag it as 'TBD'.

    • Every action item has an owner assigned
    • Due dates or 'TBD' are included for each item
    • Both customer-side and internal actions are captured
  4. 4

    Assess the customer's sentiment based on language, tone, and content in the transcript. Classify as Positive, Neutral, Mixed, or Negative. Provide 2–3 specific quotes or signals from the transcript that justify the classification. Note any risk signals such as churn indicators, escalation language, or budget concerns.

    • Sentiment is classified with one of the four labels
    • At least 2 supporting signals or quotes are cited
    • Risk signals are explicitly flagged if present
  5. 5

    Identify specific CRM fields that should be updated based on the call. For each field, state the field name, the current implied value, and the new value to enter. Common fields include: Deal Stage, Close Date, Customer Health Score, Primary Use Case, Key Pain Points, Products Mentioned, Contract Renewal Date, and Executive Sponsor.

    • Each recommended update names the specific CRM field
    • New field values are concrete, not vague
    • Only fields with evidence from the transcript are included
  6. 6

    Compile the final CRM note package in a clean, copy-paste-ready format: sections for Summary, Next Steps, Sentiment & Risk, and CRM Fields to Update. Keep language concise and scannable for a sales rep reviewing it in 60 seconds.

    • All four sections are present and clearly labeled
    • Total output is scannable and free of filler
    • Format is ready to paste directly into a CRM activity log

Output format

Return four clearly labeled sections: (1) Summary — 3–5 sentences in third-person past tense; (2) Next Steps — bulleted list with owner, action, and due date; (3) Sentiment & Risk — classification label, 2–3 supporting signals, and any risk flags; (4) CRM Fields to Update — a two-column list of field name and new value. Use plain text formatting suitable for pasting into any CRM.

Use it everywhere

Copy this skill into your library to inject it into Claude, ChatGPT, and Gemini — or install your whole library as / slash commands in Claude Code and Cowork.

Get started free →