Skip to main content
the optional sales view turns discovery calls into structured meddpicc qualification. this page is the practical version - how to enable it, what to listen for, and how to read the output. for the feature overview see ai insights.

​
the letters

​
before the call

  1. open the insights more menu and choose sales before you hit record. this enables sales analysis for live and final updates
  2. connect google calendar so attendee names + domains flow into the prompt - this materially improves economic_buyer and champion attribution
  3. if you know the account ahead of time, jot what you already know in the notes panel - it won’t feed the ai, but it keeps you anchored

​
during the call

sales updates each field live as the transcript grows. you don’t need to ask every question explicitly, but here’s a prompt set that tends to surface the fields: metrics
  • “if this worked, what would you measure three months in?”
  • “what’s the number you’re trying to move?”
economic buyer
  • “who else needs to sign off on something this size?”
  • “where does the budget for this sit?”
decision criteria
  • “when you’ve evaluated tools like this before, what mattered most?”
  • “what would make you rule a vendor out?”
decision process
  • “walk me through what happens between today and a signed contract”
  • “have you bought something at this price point before? how did that go?”
paper process
  • “any security review or legal step we should plan for?”
  • “do you have a vendor onboarding process?”
identified pain
  • “what happens if you don’t solve this this quarter?”
  • “how are you handling it today?”
champion
  • “who else on your side is feeling this pain?”
  • “who’d be using this day-to-day?”
competition
  • “what else are you considering?”
  • “what would you do if we didn’t exist?“

​
reading the output

the sales view shows each meddpicc field as a short summary with the supporting quote. good signs:
  • concrete metrics - numbers, timeframes, current-vs-target
  • named economic buyer - a person, not a role
  • champion with a reason - “sarah, because her team is blocked on this”
  • competition includes “doing nothing” - a real risk, not just other vendors
red flags the ai will surface as vague or empty fields:
  • metrics field says “improve productivity” - press for a number next call
  • economic buyer is null after 30 minutes - you’re talking to the wrong person
  • decision process is null or “they’ll get back to us” - no real process = no real deal

​
after the call

  1. stop the meeting. miniti saves the final transcript before finishing insights in the background
  2. review the sales view. if a field looks wrong, check the transcript quote - the ai only uses what was actually said
  3. export to attio or Twenty (macos only) - meddpicc fields ride along with the note. see attio and Twenty CRM
  4. for your next touch, the empty fields are the agenda

​
regenerating

if the call ran long and early qualification got buried, open the saved meeting, select sales, and use update to re-run against the full transcript. sales re-extracts all 8 fields.

​
using questions mode alongside

after a discovery call, switch to questions on the saved meeting. it generates incisive follow-ups referencing specific things said - useful as prep for the next meeting or as the body of your follow-up email. see questions mode.

​
webhook output

meddpicc fields are included in the webhook payload under meeting.meddpicc after sales analysis has run. full shape in webhooks.

​
not a replacement for judgment

sales extracts what was said. it can’t tell you whether the champion is actually a champion or whether the economic buyer was bluffing. use it as a checklist and a memory aid, not an oracle.