Library › Sales AI › Sales and customer success
Turn a call transcript into CRM tasks as a CSV block
- For Sales and customer success
- You practice Asking for a format you can use
- Time about 5 minutes
- Level Intermediate
About this exercise
You want to import the next steps from today's call (Tuesday, Nov 11, 2026) into your CRM, which takes CSV. Put them in exactly 1 code block with the header owner,action,due_date, one row per commitment, dates as YYYY-MM-DD. The kickoff moved during the call, so use the final date. Keep any text outside the block under 20 words.
Common mistakes when you ask AI for this
- Getting a prose recap. The AI writes a friendly summary with bullets instead of rows you can import. Try: Say: "Output exactly 1 code block of CSV with the header owner,action,due_date."
- Leaving dates as words. The AI writes "by Thursday" or "the 21st", which a CRM import can't read. Try: Add: "Convert every date to YYYY-MM-DD. Today is Tuesday 2026-11-11."
- Missing the moved kickoff. The AI may keep the first date mentioned (the 25th) instead of the final one. Try: Add: "Use the final kickoff date agreed in the call."
- Adding explanation around the block. The AI adds an intro and notes that you'd have to delete before importing. Try: Add: "Text outside the block under 20 words."
- Skipping small commitments. Side promises like checking the Spanish portal get dropped from summaries. Try: Add: "One row per commitment, including small ones."
What a strong prompt includes
- Asks for CSV
- Gives the header
- Sets the date format
- Gives today's date
- Limits the extra text
Related exercises
- Accept the buyer's counteroffer on three stated conditions (Sales AI · Sales and customer success)
- Adapt one ad image to 3 sizes with Midjourney (Image & Video AI · Marketing and content)
- Age open requisitions as of today and flag the old ones (Finance AI · FP&A and financial analysts)
- Announce a discount-approval process change in three bullets (Sales AI · Managers and team leads)
- Announce a new win in Slack without revealing the price (Sales AI · Sales and customer success)