What discovery is actually for
Discovery has one job: find out whether there is a real problem you can solve, and what happens if it goes unsolved.
Everything that goes wrong in these meetings comes from quietly replacing that job with a different one, which is convincing them to buy. The moment the goal shifts, the questions become leading, the listening stops, and you end up with a proposal built on what you hoped was true.
The one-sentence test
By the end, you should be able to state their situation in one sentence they would agree with word for word. If you cannot, discovery is not finished, regardless of how well the conversation seemed to go.
A 40-minute structure
Set the frame, two minutes
Tell them what the call is for and what it is not. "I want to understand how you handle this today. I am not going to pitch anything, and at the end I will tell you honestly whether I think we can help."
This changes the conversation more than it should. People are braced for a sales call, and naming it otherwise moves them off script.
2 minutesWalk the current process, fifteen minutes
How do you do this today? Step by step, who does what, what happens when it goes wrong.
This is the most productive part of any discovery call. People describe processes accurately and describe problems abstractly, so the walkthrough surfaces friction they would never think to name.
15 minutesThe highest-value sectionFind the cost of it staying the same, ten minutes
What does this currently cost in time, money, missed work, or people's patience? What happens in six months if nothing changes?
If there is no answer to this, there is no deal, however friendly the conversation is. Problems without a cost do not get budget.
10 minutesWhere deals qualify or dieEstablish the decision, eight minutes
Who else is affected, who signs off, what has to be true for this to be a priority this quarter, and whether there is budget or a route to it.
Ask directly. People answer these questions honestly far more often than salespeople expect, and guessing produces the deals that stall at proposal.
8 minutesAsk directlyConfirm and close the loop, five minutes
Say their situation back to them and ask if that is right. Then be honest about fit, including saying so if it is poor.
Agree one specific next step with a date before ending the call. "I will send a summary tomorrow, and we speak again on the 3rd."
5 minutesAlways end with a dated next step
Questions that find the real problem
| Question | What it gets you |
|---|---|
| Walk me through how you do this today | Friction they would never name directly |
| What have you already tried? | Prevents you proposing something they rejected |
| Who else is affected when this goes wrong? | The other stakeholders, without asking for a list |
| What happens if you do nothing for six months? | Whether there is urgency or just interest |
| What made you look into this now? | The trigger event, which is usually the real driver |
| If this were solved, what would be different on a Monday morning? | Their actual success criteria |
| What would stop this happening? | Objections, surfaced early instead of at proposal |
The last one is worth the discomfort. Asking what would prevent this from going ahead brings budget cycles, competing projects, and internal politics into the open while you can still do something about them.
Five mistakes that ruin discovery
Talking more than they do. The most common. Usually caused by answering a small question with a long explanation. Answer briefly and hand the conversation back.
Demoing too early. A demo before you understand the problem is a feature tour, and it moves the conversation permanently onto your product rather than their situation.
Accepting the first problem stated. The first thing someone names is rarely the expensive one. It is the one that is easiest to articulate. Keep going.
Leading the witness. "So it sounds like the real issue is X, right?" gets you agreement, not information. People say yes to be agreeable, and you build a proposal on your own assumption.
No dated next step. A call that ends with "let me know" has no forward motion, and the follow-up becomes chasing. Set the next step while you are both still on the call.
What to do in the hour afterwards
Two things, and both decay fast.
Write it down while it is fresh. Their process, the cost, who decides, the trigger, and their exact words for the problem. Their language matters: using their phrasing in the proposal is far more persuasive than translating it into yours.
Send a short summary the same day. Three or four lines of what you understood, and the agreed next step. This does more work than it appears to: it gives them a chance to correct you cheaply, it proves you listened, and it means the internal conversation they have without you is anchored to your framing rather than their recollection.
If the fit is poor, say so in that message rather than going quiet. It costs you a deal that was never real, and it is remembered.
Frequently asked questions
- How long should a discovery call be?
- Thirty to forty-five minutes is enough for a first conversation, and shorter calls tend to be better attended. If you genuinely need ninety minutes, you are probably combining discovery with a demo, which are better kept apart.
- Should I demo the product in a discovery call?
- Not usually. A demo before you understand the problem is a generic tour, and it shifts the conversation from their situation to your features. If they insist, show one narrow thing tied to something they have already told you.
- How much should I talk in a discovery call?
- Roughly a quarter of it. If you are talking more than the client, you are running a pitch. The most common cause is answering a question with a five-minute explanation instead of a sentence.
- What if the client does not know what their problem is?
- That is common and useful information. Move to the current process: walk through how they do the thing today, step by step. Problems surface as friction in the walkthrough much more reliably than in answer to a direct question.
- When should I disqualify a prospect?
- When there is no real problem, no budget, no timeline, or no route to a decision, and you cannot see any of those changing. Ending it early is better for both sides than a proposal that was never going to convert, and it protects your forecast.