Most sellers know they should "ask good questions". Fewer can say what makes a question good, and fewer still can ask one smoothly when a real person is waiting on the line. On the phone there are no facial cues, so your questions carry most of the conversation.
This article is about the questions themselves: the kinds worth knowing, how to chain them, and how to say them so they sound like curiosity and not an interrogation. For the preparation routine before a call, see the guide on preparing for a discovery call. Here we focus on what happens once the customer picks up.
What a good discovery question does
A good question does at least one of three things:
- It reveals a fact you did not have ("How many people handle this today?").
- It reveals a reason ("What made you look at this now?").
- It lets the customer hear their own situation out loud, which sometimes changes how they see it ("What happens if nothing changes?").
A weak question mostly confirms what you already assumed, or steers the customer towards the answer you want. "You would like to save time, wouldn't you?" gives you only a yes or no, and little information.
Four kinds of question to keep in your toolkit
Open questions
They start with how, what, tell me or walk me through. They invite a paragraph rather than a word.
- "How does this work at the moment?"
- "What does a normal week look like on your side?"
- "Walk me through the last time that happened."
Open questions are your default for the first half of a call.
Closed questions
They can be answered with yes, no or a number. They are not bad; they are for confirming and narrowing.
- "Is that handled by one team or several?"
- "Roughly how many requests a month?"
Use them after an open question, not instead of one.
Clarifying questions
These check that you understood, and they often unlock the most useful detail.
- "When you say 'slow', what does that look like in practice?"
- "Can you give me an example?"
- "What do you mean by 'complicated'?"
Customers use vague words. Whenever you hear one, such as "expensive", "difficult" or "busy", ask what it means for them.
Impact questions
These connect a problem to its consequences, which is where priority comes from.
- "What does that cost you, in time or money?"
- "Who else feels the effect?"
- "What happens if it stays as it is?"
Ask them gently and only after the customer has described a problem they actually care about.
A simple sequence that works on the phone
You do not need a rigid script, but a loose order helps you avoid jumping around:
- Situation: "How are you handling this today?"
- Problem: "What would you change if you could change one thing?"
- Impact: "What does that mean for you in practice?"
- Priority: "Why look at it now rather than next quarter?"
- Decision: "Who else would have a say in this?"
Treat it as a map, not a track. If the customer starts talking about impact in the first minute, follow them there, then come back for what you missed.
How to ask so it sounds natural
Ask one question at a time
Stacking questions ("What do you use now, how long have you had it, and are you happy with it?") leaves the customer answering the easiest one. Ask, then stop.
Leave the silence
After a question, count to three in your head. On the phone, silence feels longer than it is, and many sellers fill it with a second question or an answer for the customer. The best information often comes after the pause.
Use their words
If they say "it keeps falling through the cracks", ask, "When something falls through the cracks, what happens next?" It shows you are listening, and it keeps the conversation in their language and not yours.
Soften the transition
People get defensive when a conversation turns into a questionnaire. A short bridge helps: "To make sure I focus on what is useful to you, may I ask a few questions about how this works today?"
Follow the answer, not the list
The most valuable questions are usually not on your list. They are the ones that begin with "You mentioned..." or "Tell me more about...". If you notice yourself looking at your next prepared question while the customer is still speaking, you are not listening.
Questions to avoid
- Leading questions: "Wouldn't it be great if...?" The customer will say yes to be polite.
- Why questions that sound like blame: "Why did you choose them?" can feel like a challenge. Try "What led you to choose them?"
- Multiple choice you invented: "Is it more of a cost problem or a quality problem?" may force a false answer. Ask openly first.
- Questions you could have researched: asking what the company does is a quick way to lose credibility.
- Questions that only serve your pitch: if a question exists to set up your product, the customer usually feels it.
What to do with the answers
Discovery is not finished when the customer stops talking. Three habits turn answers into progress:
- Take short notes in their words. You will want exact phrases later.
- Summarise before you move on. "So the main issue is the handover between teams, and it matters because it delays invoices. Have I got that right?" Correcting you is easy for them, and it confirms you understood.
- Link the next step to what they said. "You mentioned the handover is the main delay, so it would make sense to look at that with your operations lead." The next step then follows from their problem, not from your calendar. The companion article on asking for the next step covers this in detail.
Mistakes that show up in real calls
- Pitching after the first problem. You hear one pain point and jump to your solution. Ask a few more questions first; the first problem is rarely the biggest.
- Accepting a surface answer. "We are fine" often means "I do not know you yet". A follow-up such as "What would make it even better?" can open the door gently.
- Interrupting. Even to agree. Let the sentence finish.
- Talking about yourself in your questions. "At my last company we..." takes the floor back.
- Ending the questions too early because it is uncomfortable. Some of the best answers appear after an awkward pause.
Examples: weak questions and stronger versions
| Weak | Stronger |
|---|---|
| "Are you looking to improve your sales?" | "What is the hardest part of your team's sales calls right now?" |
| "Do you have budget?" | "How would you usually fund something like this?" |
| "Is this a priority?" | "Where does this sit among your top three things this quarter?" |
| "Do you want to grow?" | "What would need to change for the team to hit its number?" |
| "So you have a problem with onboarding?" | "How long does it take a new rep to get comfortable on the phone?" |
| "Would you be interested in a demo?" | "If we could show you X in fifteen minutes, would that be useful?" |
The pattern: weak questions are closed, leading or obvious. Strong ones are open, specific and invite a story.
How to practise discovery questions out loud
Reading a list is easy. Asking these questions while someone answers unpredictably is the skill. Some ways to build it:
- Rewrite your questions in your own words. If a question sounds stiff when you say it, change it until it sounds like you.
- Record yourself asking your first four questions, then listen for stacked questions, filler words and any place you answered your own question.
- Practise follow-ups. Have three ready: "Tell me more", "Can you give me an example?" and "What happened next?"
- Work from an objection or a vague answer. Ask a friend to answer "We are fine at the moment" and practise finding a way forward without pushing.
You can also practise against a voice that answers back. In DIRI AI you choose a situation, set the difficulty and speak with an AI customer who plays only the customer and reacts to your questions, including vague or short replies. Afterwards, a report points to moments from the call and suggests one thing to try next. The sample call and report page show how it looks, and how we score explains what is assessed. If you want a script to start from, the call script generator produces a draft you can edit.
Related: How to Prepare for a Discovery Call, Healthcare Appointment Scheduling Call Training for Front Desks, MEDDIC Discovery Call Questions for B2B Sales. Free tool: Sales Glossary.
FAQ
How many discovery questions should I ask?
There is no magic number. Plan five to seven in a rough order, and expect to improvise follow-ups. Quality of follow-up matters more than quantity of questions.
What if the customer gives very short answers?
Slow down, ask for an example, and make the question easier to answer: "Could you tell me about the last time this came up?" Short answers can also mean they do not yet see why they should share, so briefly explain why you are asking.
Should I ask about budget in discovery?
Ask about how decisions are made and what the cost of the problem is before asking for a number. Budget questions land better once the customer has described why the problem matters.
Is it fine to read my questions from a sheet?
Yes, as a reminder. Do not read them word for word, and keep an eye on the conversation, not the page.
Try it free
Pick three questions from this article, say them out loud, then try them on a customer that answers back. DIRI AI has a free plan with 3 voice calls a month of up to five minutes each and no card needed. Start free and choose a discovery scenario.