Most first sales calls go wrong in the first five minutes. Not because the founder or SDR said something rude, but because they skipped discovery and jumped straight into a demo. Good discovery questions are the cheapest sales training you can give yourself: they tell you whether there's a deal, what it's worth, and which objections you'll hit later, before you ever share your screen.
This post is a practical discovery call checklist for people who sell without a sales background yet: founders doing founder-led sales, engineers and solutions engineers moving into customer-facing roles, and new SDRs and AEs. It includes the questions, the follow-ups, and the traps.
Why discovery questions beat a great pitch
A pitch answers questions the buyer didn't ask. Discovery gets the buyer to tell you which questions matter. If you're technical, think of it like debugging: you don't start rewriting code before you've reproduced the bug. Discovery is reproducing the bug in the customer's world.
The goal of a discovery call is not "build rapport" or "qualify" in the abstract. It's three concrete things:
- The problem, in the buyer's own words.
- The cost of leaving it unsolved (time, money, risk, stress).
- The path: who decides, how, and by when.
If you walk out of the call without all three, you had a nice chat, not a discovery call.
The discovery call checklist
Use this as a map, not a script. You don't need to ask every question; you need to cover every box.
1. Open: set the agenda
"Thanks for making the time. I'd like to spend most of this call understanding how you handle X today, then if it makes sense I'll show you a few minutes of what we do. Does that work, and is there anything you want to make sure we cover?"
That last question matters. It hands the buyer some control and often surfaces the real reason they took the call.
2. Current state
- "Walk me through how you do this today, step by step."
- "Who's involved in that process?"
- "What tools are you using for it right now?"
Engineers are usually great here. Ask for the workflow the way you'd ask for repro steps.
3. Pain
- "What's the most frustrating part of that?"
- "When did you last run into that problem? What happened?"
- "What have you already tried to fix it?"
The "last time" question is the most useful one in this list. General answers ("it's kind of slow") are weak. A specific story ("last Tuesday we lost a renewal because...") is real pain.
4. Impact
- "What does that cost you, roughly, in hours or money?"
- "What happens if nothing changes in the next six months?"
- "Who else feels this?"
Don't fill in the number for them. If they can't estimate the impact, the problem may not be painful enough to buy for, and that's important to learn early.
5. Decision process
- "If you decided this was worth solving, what would the process look like on your side?"
- "Who else would need to be involved?"
- "Is there budget set aside for this, or would it need to be found?"
- "Is there a date that makes this more urgent?"
This is where frameworks like BANT or MEDDIC help as a mental checklist. You don't need to recite them; you need to know which box is still empty.
6. Close the call, not the deal
"Based on what you've told me, it sounds like X is costing you Y, and you'd want to fix it before Z. Did I get that right? If so, the next step I'd suggest is..."
Summarise in their words and propose one concrete next step with a date. Ending with "I'll send some info over" is how deals quietly die.
Follow-ups that turn shallow answers into real ones
The first answer to a discovery question is almost always surface-level. These follow-ups do most of the work:
- "Can you say more about that?"
- "What do you mean by slow / messy / expensive?"
- "Can you give me an example?"
- "How are you handling that today?"
- "Why now?"
A useful rule: ask at least one follow-up before you move to the next box on the checklist.
Five discovery traps (and what to do instead)
1. Pitching mid-discovery. The buyer mentions a pain, you get excited, and you start explaining your feature. Write it down instead and ask a follow-up. You'll use it in the demo.
2. Leading questions. "Wouldn't it be great if your team could...?" gets a polite yes that means nothing. Ask open questions: "How would you want that to work?"
3. Interrogation mode. Firing questions from a list feels like a form. React to answers, summarise, and connect: "That makes sense given what you said about the team size."
4. Accepting "we're just exploring". Fine, but ask: "What made you start exploring now?" There is usually a trigger.
5. Skipping the money question. Founders especially avoid it. Asking about budget isn't pushy; it's respectful of both sides' time. "Is this something you'd have budget for this quarter, or is it further out?"
Discovery questions are a skill you practise, not memorise
Reading a list of discovery call questions is easy. Asking them naturally on a live sales call, while the buyer is vague, distracted, or pushing you to "just show the product", is the hard part. That's what sales role play is for: run the call out loud, hear yourself rush into a pitch, and try again.
A simple practice routine:
- Pick one checklist section (say, Impact).
- Have a colleague or friend play a buyer who gives short, vague answers.
- Your only goal is to get one specific number or story out of them using follow-ups.
- Swap roles. Playing the buyer teaches you more than you'd expect.
Do this for ten minutes before a week of calls and you'll notice the difference in how calm you feel when a buyer goes quiet.
Quick copy-paste checklist
[ ] Agenda set + asked what they want to cover
[ ] Current workflow, step by step
[ ] Specific recent example of the pain
[ ] Cost / impact, in their words
[ ] What they've already tried
[ ] Who decides, how, by when
[ ] Budget status
[ ] Summary read back + confirmed
[ ] One next step with a date
We're building Kulissa, an AI sales coach where you practise calls like discovery and objection handling against a realistic AI buyer before the real one. If you want to rehearse this checklist out loud, you can try it at kulissa.com.
Top comments (0)