Library › Data AI › Sales and customer success

Pull the overdue deals out of a pipeline export

Practice: Giving the right context. About 5 minutes.

About this exercise

It's Wednesday, November 5, 2026. Your sales manager wants a Slack list of overdue deals: the expected close date has already passed and the stage is not Closed Won or Closed Lost. Last-activity dates don't count. Give her a numbered list of only those deals, biggest value first, one line each with just the deal name, owner and value, no intro, under 60 words, plain text. Write a prompt that gets that list from the export.

Common mistakes when you ask AI for this

  • Asking "which deals need attention?". The AI invents its own idea of "attention", like old activity dates, and flags deals that aren't overdue. Try: Define it: "Overdue means the close date is before today and the stage isn't Closed Won or Closed Lost. Ignore activity dates."
  • Not giving today's date. The AI doesn't reliably know today's date, so it can't tell which close dates have passed. Try: Add: "Today is November 5, 2026."
  • No sort order. The AI keeps the export's order, so your manager has to hunt for the big ones. Try: Add: "Biggest deal value first."
  • Trusting the AI to count words. AI is bad at counting. Ask for "under 60 words" and it often lands well over. Try: Cap each part instead, like "each line under 12 words," so the total lands well under 60.
  • Getting a report instead of a list. AI adds a heading, a table and a paragraph of advice. In Slack the bold shows up as asterisks. Try: Add: "One line per deal: name, owner, value. No intro. Plain text, no bold, no table."

What a strong prompt includes

  • Gives today's date
  • Defines overdue
  • Sets the sort order
  • Says what each line shows
  • Asks for plain text

Practice this now ↑

Related exercises