29 August 2026
Ask what someone did last time they had this problem, not what they'll do next. Past-tense questions produce dates, dollar amounts and the names of people who said no; future-tense questions produce plans, which are polite fiction until someone signs. The five minutes you spend following one honest answer are worth more than the fifteen questions left on your script.
"What would you do if this kept happening?" invites a hypothetical, and people are generous with hypotheticals because nothing is at stake. "What did you do the last time this happened?" invites a memory, and memories come with specifics attached — a vendor they fired, a spreadsheet someone still maintains by hand, a number a manager quoted in a meeting three months ago.
The tense is the whole trick. Intention is cheap; recollection has a cost, so people only report it accurately. If a prospect can't answer a past-tense question at all, that's information too — it usually means the problem hasn't actually cost them anything yet, whatever they said in the first five minutes.
A script is a list of questions in a fixed order, which means it's built to survive being ignored — the seventh question gets asked whether or not the third one revealed something worth chasing. The better habit is to treat every answer as a fork: either it opens onto a specific fact, or it's a wall of agreeable language, and either way the next question should come from what was just said, not from the sheet.
This is slower. It also means fewer calls end with "sounds great, let's follow up next week" and more end with a number, a date or a name attached to a decision. If someone says "budget's not really an issue," the next sentence out of your mouth should not be the next item on the list — it should be "who told you that, and when?"
"No problem" almost never means no problem. It means the person hasn't finished thinking, doesn't want to commit in front of a stranger, or is protecting a colleague who owns the mess. "We're pretty happy with what we have" from someone who booked the call at all is a contradiction worth naming out loud, gently: "if that's true, what made you take this meeting?"
The tell is usually a missing object. "It's fine" answers nothing — fine compared to what, fine for whom, fine until when. Ask the prospect to finish the sentence instead of accepting the adjective.
The reason reps write badly during discovery calls is that they're choosing, in real time, between listening and transcribing, and the choice shows on their face. The fix isn't a faster typing hand — it's writing down less. During the call, capture only the words a prospect actually used for money, dates and names; everything else is reconstructable afterward from memory or a recording.
AVAY runs the call in the browser and transcribes it as it happens, so the rep doesn't need to be the scribe at all — the number the prospect said out loud is captured verbatim, not paraphrased from a rushed note ten minutes later. If a rep wants to check what was said in a call three weeks ago before this one, AVAY searches across every past meeting rather than making them dig through old notes.
A discovery call has done its job when four things are on paper in the prospect's own words rather than in yours: a real problem with a cost attached, a budget with a source, a timeline with an event behind it, and the name of whoever actually has to say yes. Anything short of that is a pleasant conversation, not discovery.
| Future-tense question | Past-tense question that gets an answer | |
|---|---|---|
| Budget | What's your budget for this? | What did the last thing like this cost you, in money or time? |
| Timeline | When are you hoping to have this solved? | What has to happen before this becomes urgent again? |
| Problem | What are you looking for in a solution? | What did you actually try last time this came up? |
| Decision process | Who's involved in the decision? | Who signed off the last time you bought something like this? |
There isn't one that works cold — the best opener depends on what the prospect already said to book the call. But the reliable move is to ask about the last time they tried to solve this problem, because it forces a memory instead of an intention and usually surfaces a name, a cost or a failed vendor in the first answer.
Fewer than most scripts assume. Four facts — problem, budget, timeline, decision-maker — with real specifics attached is a successful call, even if it took twelve follow-up questions to get there rather than the fifteen items on a template.
Capture the exact numbers, dates and names during the call, in the prospect's own words, and reconstruct the connective tissue afterward from memory or the recording. Trying to write full sentences while listening is what makes reps miss the follow-up question that would have gotten a real answer.
Stop moving to the next question and name the vagueness out loud — "you said it's fine, fine compared to what?" A prospect who resists specifics twice in a row usually hasn't felt the cost of the problem yet, which is itself useful information about whether this is a real deal or a curious one.
Ask what the last attempt at this problem cost them, in dollars or in the hours someone spent managing it manually. People who won't state a budget will often state a cost, and a cost is the same number wearing a different name.
Ask what someone actually did last time, not what they plan to do next, and follow whatever answer you get instead of the next line on your script — that's where the budget, the timeline and the real problem live.
AVAY is a video meeting platform that transcribes the call itself — no bot joins, because there is nothing to join. Start one at avay.ai, read how each part works in the documentation, or see what it costs.