Discovery Call Questions: 40 to Ask, Grouped by What You Need to Learn, With a Follow-Up for Each
Good discovery call questions do six jobs: they establish the situation, find the problem, size its impact, map the decision, confirm budget and timing, and surface risk. You do not need many. Gong's analysis of 519,000 recorded discovery calls found the most successful ones had 11 to 14 questions and went deep on 3 to 4 problems. Below are 40 questions to choose from, each with the follow-up that gets a real answer, plus a call plan builder and a 30-minute template.
Six things to learn, 40 questions to pick from
11 to 14
questions in the best calls
3 to 4
problems covered in depth
519K
calls in Gong's analysis
Bars: questions per group in this guide. Figures: Gong.
What Is a Discovery Call?
A discovery call is the first real sales conversation with a buyer who has shown interest. Its job is to learn whether the buyer has a problem you can solve, what that problem costs them, and how they would decide to fix it, so that both sides know whether a next step is worth their time.
Discovery call questions are the tool for that job. The pages that rank for this term mostly give a long flat list, 20 to 100 questions, with no way to tell which twelve to use on Tuesday's call. Few give the follow-up, and the follow-up is where the answer is. The first reply to almost any discovery question is the polite, short version.
One disclosure: SnapSDR is an AI demo agent that runs on Tough Tongue AI's agent infrastructure. It comes up once, near the end, where discovery is handled by an agent before a rep joins. The data in this guide comes from Gong, Highspot, and HubSpot pages read in November 2026, and all three sell sales software.
How Many Questions Should You Ask on a Discovery Call?
Fewer than most lists suggest. Here is what three widely cited sources say, and what each is based on.
| Source | What it says | Based on |
|---|---|---|
| Gong | 11 to 14 questions per call line up with the most success. The best calls cover 3 to 4 business problems. | AI analysis of 519,000 recorded discovery calls. Published 2017, page last updated March 2026. |
| Highspot | 8 to 12 thoughtful questions usually give enough depth. No magic number. | Practitioner guidance, updated September 2026. Not a study. |
| HubSpot | Discovery calls typically run 20 to 45 minutes, and most are 30. | Editorial guide, updated July 2026. |
Read the Gong numbers as a range, not a rule. They describe what successful calls had in common. They do not prove that asking a thirteenth question causes a win. The study is also several years old and drawn from Gong's own customers. Still, it is the largest data set anyone has published on this, and the direction is sensible: about a dozen questions in 30 minutes leaves two to three minutes per question, which is enough for an answer and a follow-up.
Where the questions fall matters as much as how many. The same Gong analysis found that top sellers spread their questions evenly through the call, while average sellers ask most of theirs at the start. It also found they ask most of their questions while the buyer is talking about problems, and fewer elsewhere.
Twelve questions in a 30-minute call, placed two ways. Each dot is one question.
Spread out
The same twelve questions, one every two to three minutes. Each answer gets a follow-up or a short reply before the next question, so it reads as a conversation. Gong found top sellers spread questions evenly like this.
Dot positions are an illustration, not Gong's data. The finding they illustrate is from Gong's analysis of 519,000 recorded discovery calls.
The practical version: take 10 to 14 questions into a 30-minute call, expect to use about ten, and never ask two in a row without responding to the first answer.
What Are the Best Discovery Call Questions?
The best question is the one that fills the biggest gap in what you know about this buyer. So the 40 below are sorted by the gap they fill. The right-hand column is the follow-up to use when the first answer is short or vague, which is most of the time.
Situation questions
How the work gets done today and why they took the call now. Keep this short; most of it should come from research.
| # | Question | Follow-up |
|---|---|---|
| 1 | Walk me through how your team handles this today, from start to finish. | Where in that does the work wait on a person? |
| 2 | What made you take this call now, and not six months ago? | What changed? |
| 3 | Who is involved day to day, and what does each of them use? | Which of those tools would you keep no matter what? |
| 4 | What have you already tried to fix this? | Why did that stop short? |
| 5 | How do you measure whether this is going well? | What is that number today? |
| 6 | What else is on your plate this quarter that this would compete with? | Where does this rank against those? |
Problem questions
What is going wrong, how often, for whom, and what they believe causes it. Spend the most time here.
| # | Question | Follow-up |
|---|---|---|
| 7 | What is the hardest part of this right now? | Can you give me the most recent example? |
| 8 | When did you first notice it? | What was happening in the business at the time? |
| 9 | How often does it happen? | Is that getting more frequent or less? |
| 10 | Who feels it most? | What do they do when it happens? |
| 11 | What do you think is causing it? | What makes you say that? |
| 12 | If you could change one thing about how this works, what would it be? | What has stopped you from changing it already? |
| 13 | What workarounds has the team built? | How much time do those take in a week? |
| 14 | You mentioned that problem. What does it lead to further down the line? | And then what happens? |
Impact questions
What the problem costs and what fixing it is worth, in the buyer's own numbers.
| # | Question | Follow-up |
|---|---|---|
| 15 | What does this cost you in a typical month, in hours or dollars? | How did you get to that number? |
| 16 | What happens if nothing changes for another year? | Who would notice first? |
| 17 | Which company goal does this get in the way of? | Who owns that goal? |
| 18 | How does this affect your own work? | What would you do with that time back? |
| 19 | What would a good outcome look like twelve months from now? | How would you measure it? |
| 20 | If this were fixed, what could the team do that it cannot do today? | What is that worth to the business? |
| 21 | On a scale of 1 to 10, how urgent is this? | Why not two points lower? |
Decision process questions
Who decides, in what order, and what each person needs to see.
| # | Question | Follow-up |
|---|---|---|
| 22 | Besides you, who needs to be comfortable with a decision like this? | What does each of them care about most? |
| 23 | Tell me about the last time your team bought something like this. How did it go? | What would you do differently this time? |
| 24 | What steps sit between a yes from you and a signed agreement? | How long does each one usually take? |
| 25 | Who could say no at the last minute? | What would make them say it? |
| 26 | What will security, legal, or IT need to see? | When should we bring them in? |
| 27 | How will you decide between the options? | Which of those criteria matters most? |
| 28 | What would you need to see in a demo to feel confident? | Who else should be in that demo? |
Budget and timing questions
Whether money exists, whose it is, and what date matters and why.
| # | Question | Follow-up |
|---|---|---|
| 29 | Is there budget set aside for this, or would it need to be found? | Whose budget would it come from? |
| 30 | What are you spending on this problem today, counting tools and time? | Which part of that would go away? |
| 31 | What range did you have in mind? | What is that based on? |
| 32 | When do you need this working by? | What happens on that date if it is not? |
| 33 | What is driving that date? | Is it fixed, or a preference? |
| 34 | What could push this to next quarter? | How likely is that? |
Alternatives and risk questions
What you are up against, including doing nothing, and what could stop the deal.
| # | Question | Follow-up |
|---|---|---|
| 35 | What else are you looking at, including doing nothing or building it yourselves? | What do you like about each? |
| 36 | What worries you about making a change? | What would put that worry to rest? |
| 37 | What went wrong the last time you rolled out a new tool? | How would we avoid that here? |
| 38 | From what you have heard so far, where might we not be a fit? | How big a problem would that be? |
| 39 | What would have to be true for you to move forward? | Which of those is least certain today? |
| 40 | What should I have asked that I did not? | Who else should I be asking? |
Two that earn a place on every call. Question 2, “What made you take this call now?”, tells you what triggered the search, which is usually the real problem. Question 40, “What should I have asked that I did not?”, catches the thing your list missed. Buyers often answer it with the most useful sentence of the call.
Budget questions come late for a reason. Asked in the first five minutes, “Is there budget?” gets a guarded answer or a no. Asked after the buyer has put a number on the problem, it is a natural next question. If your team qualifies on a framework such as BANT or MEDDIC, the groups above map onto it; our guide to AI lead qualification covers where those frameworks fit when an agent does the qualifying.
How Do You Choose Which Questions to Ask?
Start from the call, not the list. How long you have and how the meeting came about decide where the time should go. A buyer who asked for a demo has already admitted to a problem. A buyer you cold-called into a meeting has not.
Pick the call. Get a time budget and how many questions to take from each group.
Call length
Type of call
Plan 12 questions for this call. They came to you, so they already believe there is a problem. Spend less time on the situation and more on who decides and by when.
The time split is our suggestion, not a benchmark. Question counts assume about two minutes per question with its follow-up and are capped at 14, the top of the range Gong found in successful calls.
Then cut by what you already know. Before the call, write one line under each of the six groups: what you know and how. A blank line is where your questions go. A line that says “from their website” needs a confirmation, not a question.
Keep the list where you can see it and hold it loosely. If the buyer starts describing a problem in detail in minute four, stay there. The plan exists so you notice at minute twenty that you still do not know who else decides.
What Should You Not Ask on a Discovery Call?
Anything you could have looked up. HubSpot's guide makes the same point: most discovery calls are 30 minutes, and reps should not spend them asking buyers to explain what research would have shown. Every looked-up fact also gives you a better question.
| Do not ask | Find it here | Ask this instead |
|---|---|---|
| What does your company do? | Their website and last two press releases | I saw you launched in two new regions this year. How has that changed this process? |
| How big is your team? | LinkedIn headcount for the department | It looks like the team has roughly doubled since last year. What broke first? |
| What tools do you use? | Job posts, the form they filled in, tech-lookup tools | Your job posts mention two tools for this. Where do they overlap? |
| What is your role? | Their LinkedIn profile | You moved into this role in the spring. What did you inherit? |
| How did you hear about us? | CRM source field or the form | You came in through the pricing page. What were you hoping to find there? |
| Are you the decision maker? | You cannot research this one | Besides you, who needs to be comfortable with a decision like this? |
Three other kinds of question to drop: leading questions that have your product as the only answer, stacked questions that ask three things at once, and anything the buyer already answered on the form they filled in. Re-asking a form field tells them nobody read it.
How Should You Phrase Discovery Questions?
Phrase them so the buyer talks for a while. Gong found a direct link between long, uninterrupted buyer answers and successful calls, and names three openers that produce them: “Can you help me understand,” “Can you walk me through,” and “Talk to me about.” All three ask for a story, and a story has the detail you need.
| Weak | Why it fails | Stronger |
|---|---|---|
| Do you have problems with reporting? | Yes or no. Ends the thread. | Walk me through how the weekly report gets built. |
| Wouldn't it help to automate that? | A pitch dressed as a question. | What would you do with the hours that takes? |
| What's your budget, timeline, and who signs off? | Three questions. You get an answer to the easiest one. | When do you need this working by? |
| Why haven't you fixed this? | Sounds like blame. | What has got in the way of fixing it so far? |
| Is this a priority? | Everyone says yes. | Where does this rank against the other things on your list this quarter? |
One question, then wait. The pause after a hard question feels longer to you than to the buyer. If you fill it with a second question or a multiple-choice list of answers, you get the easy answer and lose the true one.
Give something back between questions. A short reply (“That matches what we hear from other finance teams of your size”) or a one-line summary keeps the call from feeling like an interview. It is also how questions end up spread through the call instead of bunched at the start.
What Does a 30-Minute Discovery Call Template Look Like?
Six steps. The times are for an inbound call; shift minutes from decision to problem for a meeting you booked cold.
| Time | Step | What to say or do |
|---|---|---|
| 0:00 to 0:03 | Open and agree the agenda | I have 30 minutes. I would like to understand how this works for you today, share what we do if it is relevant, and agree a next step either way. What would make this useful for you? |
| 0:03 to 0:06 | Situation | One or two questions that research could not answer. Confirm what you found instead of asking for it. |
| 0:06 to 0:15 | Problem | Three or four questions with a follow-up each. Stay on a problem until you have an example, a frequency, and who it affects. |
| 0:15 to 0:21 | Impact | Two or three questions. Get one number in the buyer's words: hours, dollars, deals, or a date. |
| 0:21 to 0:26 | Decision, budget, and risk | Two or three questions. Who else decides, by when, and what could stop it. |
| 0:26 to 0:30 | Recap and next step | Here is what I heard: [problem], costing [number], needed by [date], with [people] involved. Did I get that right? The useful next step is [step]. Does [day] work? |
The recap is the step reps skip and the one that does the most. Saying the problem, the number, the date, and the people back to the buyer shows you listened, lets them correct you, and gives you the opening lines of the follow-up.
If the next step is a demo, the answers decide what you show. Our sales demo script starts from exactly these notes: the two or three problems the buyer named, in their words.
What Do You Do With the Answers After the Call?
Write the six lines within the hour. One line per group: situation, problem, impact, decision process, budget and timing, risk. Use the buyer's words and numbers. Blank lines are your agenda for the next conversation. Then send the recap the same day; our follow-up email templates include one built on this recap.
Score the call, not just the deal. Discovery is a skill, and it improves when it is measured. A sales call scorecard for discovery checks a short list: did the rep agree an agenda, get a number for impact, learn who else decides, and recap before closing. Managers and reps scoring the same recording against the same card is the fastest way to agree on what good sounds like.
Practice the follow-ups. Reading a list of questions is easy. Asking “How did you get to that number?” of a senior buyer who just gave a vague answer is not. That is worth rehearsing with a manager, a peer, or an AI roleplay tool; our sales role play scenarios include discovery calls, one of them under time pressure.
When an AI agent runs the first call. Some teams now put an AI demo agent on inbound first meetings. SnapSDR, for example, joins a Google Meet or Zoom call, shows the product, and asks discovery and qualifying questions along the way, then hands the deal to a rep with a transcript and notes. In that setup the rep's first call is the second conversation. The six lines are partly filled in, so the rep's job is to confirm them and go deeper on impact and decision process, not to start again from question one.
Frequently Asked Questions
What is a discovery call?
A discovery call is the first real sales conversation with a buyer who has shown interest. The seller uses it to learn whether the buyer has a problem the product can solve, what that problem costs, who would decide to fix it, and by when. It usually runs 20 to 45 minutes, most often 30, and ends with both sides agreeing whether a next step such as a demo is worth it.
How many questions should you ask on a discovery call?
About a dozen in a 30-minute call. Gong's analysis of 519,000 recorded discovery calls found that 11 to 14 questions lined up with the most success, and that the best calls went deep on 3 to 4 business problems. Highspot suggests 8 to 12. Treat these as a range, not a rule: the numbers describe what good calls had in common, and they do not prove that one more question wins a deal.
What are the best discovery call questions?
The best ones fill the biggest gap in what you know about this buyer. Six groups cover most calls: situation, problem, impact, decision process, budget and timing, and alternatives and risk. Two questions earn a place on almost every call. "What made you take this call now?" shows what triggered the search. "What should I have asked that I did not?" catches what your list missed.
What questions should you avoid on a discovery call?
Avoid anything you could have looked up, such as what the company does, how big the team is, or what the buyer's role is. Use the research to ask a sharper question. Also avoid leading questions that have your product as the only answer, stacked questions that ask three things at once, and anything the buyer already answered on the form they filled in.
How long should a discovery call be?
Most are 30 minutes. HubSpot's guide puts the typical range at 20 to 45 minutes. A 15-minute call is enough to qualify but not to diagnose, so use it to decide whether a longer call is worth booking. A 45-minute call should go deeper on the same three or four problems, not cover a longer list of questions.
What is the difference between discovery questions and qualifying questions?
Qualifying questions decide whether to keep going: is there a fit, a budget, a decision maker, a date. Discovery questions explain the problem: what is going wrong, what it costs, and what a fix would be worth. A good discovery call does both, with the problem and impact questions first. Budget and authority questions land better once the buyer has put a number on the problem.
Pick Twelve, Not Forty
A discovery call goes well when the buyer does most of the talking about a problem they care about, and you leave knowing what it costs, who decides, and by when. That takes about a dozen questions, placed through the call, each with a follow-up.
Our recommendation: before your next call, write one line under each of the six groups from research, pick ten to twelve questions for the blanks with at least three on the problem, and add questions 2 and 40. Recap before you close. Then score the recording against those six lines and see which one stayed empty.
Let Reps Start at the Second Conversation
SnapSDR joins Google Meet or Zoom, runs the first demo through your slides or a live browser, asks your discovery questions, and hands each deal to a rep with the transcript and notes.
Related Reading
SnapSDR Team
AI Sales Automation Experts