B2B Sales Discovery Questions: What to Ask and What Each Answer Tells You
On this page
A buyer says, "Our onboarding is slow." One rep writes it down and moves to the next question on the list. Another asks, "When did that last cost you a deal?" The second rep learns more from one follow-up than the first does from the rest of the list.
B2B sales discovery questions are the questions a seller asks to learn whether a buyer has a problem worth solving, why it matters now, who owns it, and what should happen next. This guide groups them by the job each one does. For every question, it shows what a useful answer sounds like, what a weak answer sounds like, and what to do with each. The questions are adapted from the written discovery guide our own sales team uses on first calls.
TL;DR
- Ask a question only if the answer will change your next question, your recommendation, or whether you keep working the deal.
- Cover six jobs: why now, the current situation, impact, people and buying process, alternatives and risk, and the next step.
- When an answer is vague, stay on it. Ask for a recent example, what happened next, and why it matters.
- Plan three to five facts you must learn per call instead of a question count.
- Ending a deal with no trigger, owner, or consequence is a valid discovery result.
What a discovery question has to do
A discovery question is worth asking when the answer changes what you do next. If the buyer's reply would not change your next question, your recommendation, or your decision to keep working the deal, the question is filler.
The discovery template our own sales team works from lists five jobs for a first call:
- Learn why the buyer took the call.
- Turn a vague symptom into a behavior or event you can describe.
- Put a business cost and a personal cost on the problem.
- Map who owns the problem, who else is involved, and what they already tried.
- Earn the next step, or learn that there isn't one.
The question groups below follow those jobs. Each table pairs a question with what a useful answer sounds like and what a weak answer sounds like. Many question lists skip the weak column, but it is the one that tells the rep when to follow up or when to stop.
A note on where these come from: our guide was written for selling to sales leaders and enablement teams, so some examples use that world. The structure carries over to any B2B sale. Swap in your buyer's problem.
Prepare before you ask anything
HubSpot's guide to discovery call questions notes that most discovery calls are 30 minutes and that reps should not spend that time asking buyers to explain what they could have researched. Company size, recent funding, open roles, the buyer's title, and how they found you are usually knowable before the call.
Before the call, write down three things:
- The one thing you most need to learn, such as whether a real trigger exists.
- What you already believe about the account, so you can test it in one question instead of asking from zero. "I saw you are hiring six account executives this quarter. Is that what prompted this?" lands better than "Tell me about your company."
- What would make you end the deal. Knowing your disqualifiers in advance makes it easier to hear them.
Questions about why now
These questions find the trigger. Without one, the buyer is curious, and curious buyers rarely move.
Question | A useful answer sounds like | A weak answer sounds like |
|---|---|---|
What made this call worth taking? | A named event: a hiring class, a missed quarter, a new product, a leadership mandate | "We're just looking around" |
Why is this on your radar now and not six months ago? | Something changed, with a date attached | "It's always been a priority" |
Who raised this first? | A specific role, such as the CRO or a frontline manager | "The team" |
What happens if you leave things as they are for the next six months? | A consequence the buyer can describe | "Probably nothing much" |
A weak answer here still tells you something. Our guide treats "we are looking into it" with no event, owner, affected group, or performance problem as a weak signal. The right move is one more question about the trigger, then a clear read on whether the deal is real.
Questions about the current situation
These questions replace the buyer's summary with an example. "Our onboarding is slow" is a summary. "Our last new hire ran her first solo call in week nine and lost the deal on pricing" is an example you can work with.
Question | A useful answer sounds like | A weak answer sounds like |
|---|---|---|
Walk me through how this works today, from start to finish. | Steps, owners, and tools in order | A one-line description |
When did this last go wrong, and what happened? | A specific case: who, what happened, when | "It happens a lot" |
What did the person do, or fail to do, in that moment? | An observable action | A trait, such as "they lacked confidence" |
How are you handling it right now? | A workaround with a cost: spreadsheets, extra meetings, a senior person stepping in | "We manage" |
Which team, region, or segment feels it most? | One named group | "Everyone" |
Our guide has a rule for this group: do not accept "our sales team" as an answer. Push until the buyer names a group and a moment, such as "the twelve new mid-market reps on their first technical discovery calls." A specific group gives you a specific problem to solve and a specific person to talk to next.
Questions about impact and the cost of waiting
These questions connect the problem to something the business already tracks. Ask them after you have a concrete example, or the buyer will answer in generalities.
Question | A useful answer sounds like | A weak answer sounds like |
|---|---|---|
How often does this happen? | A frequency: weekly, every deal at this stage | "Sometimes" |
What happens downstream when it does? | A named result: a stalled deal, a missed target, more manager hours | "It's not ideal" |
Which number is leadership already trying to move? | A metric with an owner | "Revenue, I guess" |
What does this cost you personally? | Time, credibility, or a goal the buyer owns | No answer, which is fine |
What would you expect to see first if this were fixed? | An early signal they would recognize | Only a final revenue result |
The personal question is optional. Ask it once the buyer has described the business cost, and drop it if the answer is polite.
The last question in the table matters more than it looks. Our guide ranks evidence from early to late: the target behavior improving, a manager agreeing it improved, the behavior showing up on live calls, a leading metric moving, and only then revenue. Buyers who can describe an early signal are easier to run a pilot with, because you can both tell whether it is working before the quarter ends.
Questions about people, process, and buying
These questions map the deal. A deal with one contact can stall when the one contact goes quiet, changes roles, or turns out not to own the decision.
Question | A useful answer sounds like | A weak answer sounds like |
|---|---|---|
Who is responsible for this day to day? | A name and a role | "I do," from someone who clearly doesn't |
Who else would need to trust the result? | Specific people outside the call | "Just me" in a company with a buying committee |
How did the last purchase like this get approved? | Steps, people, and how long it took | "We'd figure it out" |
What would security, legal, or procurement need to see? | Known requirements or a person to ask | "Let's not worry about that yet" |
Is there budget for this, or would it need to be found? | Either answer, stated plainly | Evasion |
Who could stop this even if you wanted it? | A name | "Nobody" |
"How did the last purchase like this get approved?" asks about something that already happened, which is easier to answer truthfully than a question about a future process. If the buyer cannot describe a past approval, they probably cannot run this one alone, and your next step should include the people who can.
Questions about alternatives and risk
These questions show what you are competing with, which is often doing nothing.
Question | A useful answer sounds like | A weak answer sounds like |
|---|---|---|
What have you already tried? | Past attempts and why they fell short | "Nothing really" for a long-standing problem |
Are you looking at other options, including building this yourselves? | Named alternatives | "We're open" |
What would make you stay with what you have? | A real reason: cost, timing, change fatigue | "Nothing" |
What would make this fail even if the product works? | Adoption, time, a skeptical manager, a competing priority | "I can't think of anything" |
The last question comes from the pilot section of our guide. If the buyer has watched an earlier initiative stall, their answer tells you what to plan around.
Questions that earn a next step, or end the deal
These questions turn the conversation into a decision. A booked meeting is a next step only if the buyer has a reason to attend.
Question | A useful answer sounds like | A weak answer sounds like |
|---|---|---|
Based on what you've told me, here is what I heard. What did I miss? | A correction or an addition | "Sounds right" with no detail |
What would you need to see to take this to [the person they named]? | A named piece of evidence | "Just send me something" |
If we started small, which team and which moment would you pick? | A group, a moment, and a date | "Let me think about it" |
Who should be on the next call? | Names beyond the current contact | Only the current contact |
Is there a reason this shouldn't go further? | A blocker you can now see | A pause, which is worth sitting through |
The summary question carries the most weight. Restating the buyer's problem in their words, then asking what you missed, checks your understanding and gives the buyer a chance to add the detail they held back.
Ending a deal is a valid result of discovery. If there is no trigger, no owner, and no consequence, say so: "It sounds like this isn't a priority right now. Would it make sense to reconnect after your hiring plan is set?" That is a better outcome than three more meetings with a buyer who was never going to move.
What to do with a vague answer
Discovery often goes wrong at the second question. The buyer gives a broad answer, and the rep moves to the next question on the list instead of following the answer.
When an answer is vague, stay on it. Three follow-ups cover most cases:
- Ask for the example: "Can you walk me through the last time that happened?"
- Ask what happened next: "And what did that lead to?"
- Ask why it matters: "Why does that matter to you right now?"
Use the buyer's words in the follow-up. If they said "our reps freeze on pricing," ask "When they freeze, what does the buyer hear?" instead of "Can you tell me more about your pricing challenges?"
Our discovery call roleplay guide lists the full seven-question follow-up sequence our team uses, plus the ten questions we ask on almost every first call.
Adjust the questions to the person on the call
The question groups stay the same across roles. The emphasis changes. Our guide keeps a short set of follow-ups for each role our team sells to, and the pattern below generalizes it.
Who is on the call | Spend more time on | A question that fits |
|---|---|---|
Executive or economic buyer | Why now, impact, and what would make it worth the cost | "Which problem is most visible to you right now?" |
Functional leader or program owner | Current situation, what they are accountable for, and where past efforts stalled | "Where do initiatives like this usually stall between launch and results?" |
Frontline manager or daily user | The recent example and the workaround | "Which deal or project do you wish you had time for this week?" |
Technical evaluator | Requirements, integration, and risk | "What would need to be true for you to sign off?" |
Procurement or security | Process, documents, and timing | "What does your review usually require, and how long does it take?" |
Gong's discovery questions article makes a related point about seniority: senior executives tend to have less patience for a long run of questions. With an executive, ask fewer questions, bring more of your own research, and get to impact sooner.
How many discovery questions to ask
There is no correct number. Gong's discovery questions article recommends 11 to 14 targeted questions per call and says that 15 or more starts to feel like an interrogation, with the caveat above for C-suite buyers.
A count is a weak standard on its own. A rep who asks eight questions and follows every vague answer to an example will learn more than a rep who asks fourteen questions from a list. A more useful plan for a first call is three to five facts you must learn, each with one opening question and a follow-up ready.
Map the questions to your sales framework
If your team uses a qualification framework, the question groups line up with it. Score only the parts that belong on the call in front of you. A first discovery call rarely covers every element.
Question group | MEDDIC elements | BANT |
|---|---|---|
Why now | Identify pain | Need, timeline |
Current situation | Identify pain | Need |
Impact and cost of waiting | Metrics | Need |
People, process, and buying | Economic buyer, decision criteria, decision process, champion | Authority, budget |
Alternatives and risk | Decision criteria | Not covered |
Next step | Champion, decision process | Timeline |
For drills on each MEDDIC element and a call scorecard built around it, see our MEDDIC sales training guide.
Discovery questions to stop asking
Some questions are easy to answer politely without revealing anything. These are adapted from the questions-to-avoid list in our guide, generalized beyond our own market:
- "What is your biggest challenge?" It invites a rehearsed, general answer. Ask about a recent event.
- "Would faster results be valuable?" Every buyer says yes. Ask what the slow version costs them today.
- "What features do you want?" The buyer designs your product instead of describing their problem.
- "Who is the decision maker?" before you understand how the work gets done. Ask how the last similar purchase was approved.
- Anything you could have learned from their website or LinkedIn.
The common thread: hypothetical questions get hypothetical answers. Questions about what already happened get evidence.
A copyable discovery question map
Fill this in before the call. It fits on one page and keeps the rep from reading a list.
- Call goal: what you need to know by the end of this call.
- Must-learn facts (three to five): for each, write one opening question, the follow-up you will use if the answer is vague, and what you will do differently depending on the answer.
- What you already believe: one sentence per assumption you want to test.
- Disqualifiers: the answers that would end the deal.
- Next step you are aiming for: the action, the owner, and why the buyer would want it.
- After the call: one sentence describing the buyer's problem in their words, who owns it, what it costs, and what happens next. If you can't write it, you need more discovery.
Practice the follow-up, not the list
Knowing the questions is the easy part. The hard part is the moment a buyer gives an answer you did not plan for and you have a second to decide whether to follow it or move on. Reading a list does not train that. Repetition against a buyer who answers back does.
A practical way to practice:
- Pick one question group, such as impact, and one buyer role.
- Have a colleague or an AI buyer play the role with a few facts they reveal only when asked a good follow-up.
- Score whether each vague answer got a follow-up and whether the call ended with a next step the buyer agreed to. A sales call scorecard helps keep the review tied to what was said.
- Retry the weakest moment immediately, not the whole call.
In Quotain, reps practice against an AI buyer built from your playbook and real calls. They can run the full discovery call or drill the one question that keeps stalling, and raise the difficulty so the buyer pushes harder. The scorecard marks the line that cost the rep, so the retry starts at that moment. After live calls, real-call grading scores the conversation against your team's scorecard, so you can see whether the follow-ups showed up with real buyers.
For ready-made discovery scenarios with private buyer instructions and pass criteria, start with our discovery call roleplay guide.
Frequently asked questions
What are good discovery questions for B2B sales?
Good discovery questions ask about something that already happened and lead to a follow-up. Examples: "What made this call worth taking?", "When did this last go wrong, and what happened?", "What happens downstream when it does?", "How did the last purchase like this get approved?", and "What would make this fail even if the product works?" Each one has a job, and the answer tells the rep what to ask next.
What is the difference between discovery and qualification questions?
Discovery questions learn the buyer's situation, problem, and reasons for acting. Qualification questions decide whether the deal is worth pursuing, using criteria such as budget, authority, and timeline. In practice they overlap on the same call: the people, process, and buying questions above do most of the qualification work, and the answers to the discovery questions tell you whether those criteria matter.
How do you ask discovery questions without sounding like an interrogation?
Ask fewer questions and follow the answers. Use the buyer's own words in the next question, summarize what you heard before changing topics, and bring what you already know about the account so the buyer is not repeating public facts. A call where three answers got real follow-ups feels like a conversation. A call that runs down a list feels like a form.
What should you do when a buyer won't answer discovery questions?
Ask why they took the call, then offer something in return: a short point of view on what you have seen with similar teams, or a direct question about what they would need to learn from you first. If the buyer still won't share context, treat that as a signal about the deal. Ask whether a later conversation would make more sense, and agree on what would need to change.
Build the map, then rehearse it
Write the question map for your next discovery call: the facts you need, the opening question for each, and the follow-up you will use when the answer is vague. Then practice the follow-ups, because the follow-up is the part a list cannot teach. Book a free personalized practice call and run your question map against an AI buyer who answers back.