Discovery Call Questions Template

Research the account first, so you spend the buyer's time on what only they can tell you. Huthwaite International, the training firm behind SPIN Selling, says questions such as company size no longer need asking¹ because the answers are online.

Then write down three hypotheses: what you think their situation is, what is going wrong, and what is forcing a decision now. Pick the questions that would test each one. Decide in advance which you'll drop if the call runs short, and never put the why-now question on that list.

An account research brief gives you material for all three. Check notes from earlier contact too, so the buyer never has to repeat themselves.

Agree on the purpose in the first two minutes: what you hope to learn, what they want and how long they have. Then take the parts below in order, since a question about cost lands better once the buyer has described the work.

In a 2017 analysis of 519,000 recorded discovery calls, Gong, which sells call-recording software, found that asking 11 to 14 questions correlated with the greatest success². The most effective calls uncovered three or four business problems. Top sellers spread their questions evenly through the call, while average ones front-loaded them like a to-do list. So plan about 12 questions, not 30, and phrase them to draw long answers.

Group your questions by who is on the call. An operational owner knows how the work gets done, a technical evaluator what a product must pass, and an economic buyer where the money comes from. Under each question, note its purpose, what a good answer contains and one follow-up for a thin answer.

Skip "Are you the decision maker?" It invites a polite yes, and few purchases have only one. Forrester's State of Business Buying 2026 report found that a typical buying decision involves 13 internal stakeholders and nine external influencers³. Procurement staff were among the decision makers in 53% of buying cycles.

Phrase questions as if the problem exists. Julia Minson of Harvard Kennedy School and three colleagues tested this in a 2018 study. Questions that presupposed a problem got people to disclose more undesirable information⁴ than ones that presupposed none or asked generally. Stick to common problems, though: assuming an uncommon one made people judge the asker more negatively.

Let the buyer talk. Don't answer your own questions, don't rush to fill a pause, and don't demo. If they ask to see the product, make that the next step and spend the call learning what to show.

Label every answer by how you know it. Quoted: they said it, and their words go in your notes. Paraphrased: they said something that plainly means it. Inferred: nobody said it, so it's an open question, not a score. Not asked: a gap worth counting.

A recording makes quoting easy, but get consent right. In the US, federal law lets a party to a call record it⁵. The Reporters Committee for Freedom of the Press counts about 11 states that mainly require every party's consent⁶, including California, Florida and Pennsylvania. Its guide, written for journalists, advises assuming the stricter law applies when a call crosses state lines.

End by proposing the step that follows from what you heard, with a date and names. If it's a demo, build it from the call: a demo script built from your discovery call gives the most minutes to what the buyer raised most.

Walk away if the buyer confirms any of these in their own words:

• Nothing they described has a cost or consequence attached
• No budget line, and nobody willing to create one
• The event that made it urgent has already passed
• A must-have you can't provide and won't build
• Your one contact won't introduce anyone else after two asks

Send a short recap the same day: what you heard, in their words, the agreed next step and what you still need. Write it so they can forward it to someone who missed the call. Keep your CRM note separate, with commitments quoted exactly; the sales call to CRM notes tool drafts it from the transcript.

A question list is guesswork until you check it against results. Once deals start closing, compare the win rate on deals where each question was asked with the rate where it wasn't. Keep the losses in, because wins alone can't show what the losing deals lacked.

On 30 deals, one deal moves a win rate by more than 3 points, so agree on a minimum and write "too few to judge" below it. Look for other explanations too: a question only your best rep asks will look powerful, because that rep wins more anyway.

When a question keeps going unanswered, find out why. If calls run out of time first, fix the structure. If buyers dodge it, fix the wording. If one rep never asks it, that's coaching, not the question's fault.

To skip the setup, open this template with Edit with AI. River drafts the question set for each role and scores your past calls from their transcripts.