The B2B discovery call guide
How to run a discovery call that qualifies the deal instead of collecting answers. What to ask, what the answers actually mean, and when to stop.
Most discovery calls collect answers. Very few of them qualify anything, because the seller accepted the first version of each answer and moved on.
What you are actually trying to establish
Four things, and they are the same four whether you call the framework MEDDIC, BANT, or nothing at all:
| What you need | What a weak answer sounds like | What a real answer sounds like |
|---|---|---|
| A quantified problem | “We want to be more efficient” | “Two reps spend a day a week on this” |
| The economic buyer | “I’d be the one deciding” | “I’d propose it, my VP signs anything over $20k” |
| The decision criteria | “We’re looking at a few options” | “It has to pass security review and import our data” |
| A forcing date | “Sometime this quarter” | “Our current contract renews on the 14th” |
The left column is what you get if you ask once. The right column is what you get if you ask the follow-up.
The one habit that changes the call
After every answer, ask one more question about the same thing before moving on. That is it. The qualifying detail is almost always one layer underneath the answer that sounded complete.
“We want to be more efficient.” → “Efficient in what specifically? Is there a particular part of the week where it’s most obvious?”
Knowing when to stop
Discovery ends when you can state, in one sentence, what problem this company has, what it costs them, who has to agree to fix it, and by when. If you cannot say that sentence out loud after the call, you did not finish discovery. You finished the time slot.