How to Run a 25-Minute Discovery Call Without Turning It Into an Interrogation

There are two ways a discovery call dies, and most reps only worry about one of them.

The first is the interrogation. The rep has a list, the list has thirty questions on it, and the buyer can hear the list. Every answer gets a "got it, and" followed by the next unrelated question. Fifteen minutes in, the buyer stops elaborating and starts giving one-line answers, because they have figured out they are filling in a form. They will be polite. They will take the follow-up meeting off the calendar three days later.

The second is the friendly chat. This one feels good. You talk about the buyer's old company, the reorg, the weather in Denver. You laugh. You hang up thinking that went great, and then you sit down to write the recap and realize you cannot name a single quantified problem, a single person besides your contact who touches the decision, or a single reason this is happening now instead of next quarter. Rapport without qualification is just a nice phone call you paid for with an hour of pipeline time.

The fix is not "ask better questions." It is structure. A discovery call structure with a clock on it, where you know what job each block of minutes is doing, so you can stop improvising and start listening.

Here is the one I use, and the one I teach.

Why 25 minutes and not 30

Book 30, run 25. Two reasons.

The first is that your buyer's next meeting starts at the top of the hour and their calendar does not have a gap in it. Giving five minutes back is the cheapest goodwill you will ever buy, and the rep who ends early is the rep who gets the second call.

The second is that a hard 25 forces you to cut. If you tell yourself you have thirty minutes, you will spend eight of them on small talk and "a quick bit about us," and then you will rush the part that actually decides whether this is a deal. A tight clock makes you ruthless about what earns airtime.

The blocks:

• Minutes 0 to 3 — earn the right to ask.
• Minutes 3 to 15 — the diagnostic sequence.
• Minutes 15 to 20 — consequence and cost.
• Minutes 20 to 25 — the next-step commitment.

That is the whole thing. Now the inside of each block.

Minutes 0 to 3: earn the right to ask

You do not have permission to ask a stranger about their budget, their politics, or their failures. You have to buy it. The currency is a point of view, delivered fast, that proves you did work before the call.

Most reps burn this block on a company overview. Do not. Nobody has ever bought software because the logo slide had good logos on it. Instead, say what you think is true about their world and let them correct you.

The opening I use, more or less verbatim:

"Thanks for the time. I have got us down for 25 and I will give you the last five back. Before I ask you anything, let me tell you why I reached out and you can tell me how wrong I am. I saw you posted three roles on the claims side in the last month, and you moved off the legacy intake system in the spring. Usually when those two things happen together, the team is absorbing volume manually while the new system gets configured, and adjusters are doing data entry they did not sign up for. Is that anywhere close, or am I off?"

Three things are happening there. You set the time contract, so the buyer relaxes. You demonstrate you looked at something specific. And you end on a question that is easy to answer and impossible to answer with yes or no, because you have invited a correction and people love correcting you.

The correction is the gold. "Actually the intake thing is fine, our problem is on the vendor side" has just saved you eight minutes of the wrong branch.

If you are wrong about everything, say so plainly: "Okay, then I have got the wrong picture entirely. What is actually eating the most time on your side right now?" That is not a failure. That is a rep who is listening, which is rarer than it should be.

Do not spend more than three minutes here. If the buyer wants to talk about your company, say "I will earn that in about fifteen minutes — can I ask you four or five things first so what I show you is actually relevant?" Almost nobody says no.

Minutes 3 to 15: the diagnostic sequence

Twelve minutes. This is the call. Everything else is packaging.

The reason discovery feels like an interrogation is not the number of questions. It is that the questions have no visible order. When a doctor asks you fourteen questions you do not feel interrogated, because each one obviously follows from your last answer. When a rep asks fourteen questions in the order they appear on a Google Doc, you feel processed.

So the rule is: every question in this block must be built out of the buyer's previous answer. You are not working a list. You are following a thread. You should have a list — you should absolutely have a list, and I will tell you where to get one in a minute — but the list is what you studied, not what you read aloud.

Within the twelve minutes I am chasing four things, in this order:

1. The current state, in their words and their nouns

"Walk me through what happens today when a claim comes in and the documentation is incomplete." Not "do you have a process for exceptions." Make them narrate. You are listening for the nouns they use, because those nouns are what you will use for the rest of the cycle. If they say "pended," you say "pended." If they say "stuck loads," you say "stuck loads."

2. The break point

Somewhere in that narration is the step where it goes wrong. Find it and stop there. "You said it goes to a shared inbox at that point — what happens if nobody picks it up that day?" You want the specific joint where the process fails, not a general sense that things are hard.

3. The workaround

This is the question most reps skip and it is the most valuable one on the call. "So what do you do today to stop that from happening?" Every real problem has a workaround. A spreadsheet, a Slack channel, a person named Denise who just knows. The workaround tells you the true cost of the problem, because someone is already paying for it in labor. It also tells you your real competitor, which is almost never the vendor on your battlecard. It is Denise and her spreadsheet.

4. The blast radius

Who else feels this. "When that happens, who hears about it?" Then: "And when it gets escalated, who is in that conversation?" You are mapping the buying group without ever using the words "who else is involved in the decision," which makes people defensive and produces a name they gave you to end the topic.

Four threads, twelve minutes, three minutes each. That pace is doable if you are not talking. Your job in this block is to say about a fifth of the words.

Change the question order based on what they are bleeding

The four threads stay constant. The entry point does not. Which one you lead with depends on what kind of pain the account is actually in, and getting this wrong is why a good question sequence can still land flat.

There are three broad kinds of bleed, and they want to be asked about in different orders.

The buyer bleeding money

This is your classic efficiency case. Cost per transaction, margin leak, spend you cannot see. A revenue leader watching CAC climb. A brokerage watching margin compress on a lane.

Lead with the current state and go hard at volume and unit economics early, because a money buyer thinks in a denominator. "How many of those come through in a week?" "And what is the split between the ones that clear on the first pass and the ones that come back?" Get the shape of the volume before you get the emotion, because with this buyer the emotion follows the arithmetic. If you lead with "how frustrating is that," a CFO-adjacent buyer will mentally file you under vendor.

With freight and 3PL buyers this shows up almost immediately in load-level economics, and the sequence I use for brokerage conversations is laid out question by question in the freight and 3PL discovery playbook if you want the literal list before your next call.

The buyer bleeding people

Burnout, attrition, a team that is doing work beneath its pay grade. This is enormously common in clinical operations and in support-heavy orgs, and it is the one reps handle worst, because they hear "my team is exhausted" and jump straight to a demo.

With this buyer, lead with the workaround, not the current state. "Who is holding this together right now?" is a better opening thread than a process walkthrough, because the answer is a person, and the person is the problem. Then go to the break point, then volume, then blast radius. You are establishing that you understand the human cost before you start quantifying it, because if you quantify a person's team as a cost line in minute four you will lose them.

Watch your language too. "Inefficient" is an insult when the inefficiency is a nurse doing double documentation. The healthcare version of this sequence, including how to ask a CMIO about clinical burden without sounding like you are calling their staff slow, is in the healthcare discovery playbook.

The buyer bleeding regulatory exposure

Audit findings, compliance deadlines, a regulator with a date on it. This buyer is not optimizing. They are avoiding a specific bad outcome.

Invert the whole thing. Lead with the blast radius. "When the finding came back, who ended up owning the remediation plan?" With exposure, the political map is the diagnosis, because the deal will be won or lost on who is accountable for the risk. Then go to the break point, then the workaround, then current state last. Volume barely matters. One failure is enough to fund the project.

The tell for this buyer is that they will use passive voice about the trigger — "it was flagged," "concerns were raised." When you hear that, stop asking about process and start asking about people and dates. The insurance and claims playbook goes deeper on the exposure-first ordering, and the SaaS version for CROs and RevOps buyers covers the money-first ordering in the detail a growth buyer expects.

Most accounts bleed in more than one way. Lead with the loudest one and pick up the others in the consequence block.

Minutes 15 to 20: consequence and cost

You now know what is broken. Five minutes to find out whether anyone cares enough to spend money.

This is where deals get disqualified, and disqualifying at minute eighteen of the first call is the single most profitable thing an SDR or AE can learn to do. The pipeline you never build is the forecast you never miss.

Three moves.

Size it without asking them to size it. Do not say "what is that costing you." Nobody knows, and asking makes them feel audited. Instead, do the math out loud from their own numbers and let them correct it. "So if I have this right — you said it is a couple of hundred a week, and roughly a third come back for rework, and someone on your team is spending most of a day on those. Is that the right shape?" They will either confirm it or tighten it, and either way the number is now theirs, not yours.

Find the deadline. "What happens if this is still the situation in six months?" If the honest answer is "we cope," you do not have a deal this quarter and you should stop spending on it. If the answer references a date, a board meeting, a contract renewal, a go-live, or an audit response, you have a compelling event and everything after this gets easier.

Find out what they have already tried. "Has anyone looked at fixing this before?" A previous failed attempt is not bad news. It means budget existed once, and it tells you exactly what objection is going to surface in front of the committee. It also tells you if you are being used as a stalking horse to renegotiate with an incumbent.

If all three come back soft — no size, no date, no prior attempt — say so on the call. "Honestly, it sounds like this is annoying but not urgent. I would rather come back to you in a quarter than have us both sit through a demo. Fair?" Buyers remember that. Some of them call you back.

Minutes 20 to 25: the next-step commitment

Do not end on "I will send over some information." That is where pipeline goes to die.

Spend the last five minutes doing three things. Play back what you heard, in their nouns, in about four sentences, and ask what you got wrong. Then propose a specific next step with a specific attendee and a specific purpose: "The natural next step is 40 minutes with you and whoever owns the intake queue, where I show you exactly how the rework loop gets closed. Is that Denise, or someone else?" Then get the calendar out while you are both still on the phone, because a next step without a date on it is a hope.

And give the five minutes back. Say the words: "That is everything I needed, and I have got you five minutes early. Thanks for being straight with me."

The transitional lines that stop it feeling like a form

The difference between diagnosis and interrogation is almost entirely in the connective tissue. Here are the lines I use, and I use them constantly.

To change direction without a jump cut: "Can I pull on that thread for a second?"

To go deeper without repeating yourself: "Say more about that part." Four words, and it outperforms almost any question you could construct.

To get a number without an audit: "Ballpark is fine, I am not going to hold you to it."

To ask something sensitive: "This might be a dumb question." It is not, and it gives them room to answer generously.

To check you have not lost them: "Am I in the right area here, or is the real problem somewhere else?"

To move to the next block: "That is the picture I needed. Can I ask you about the consequences of it staying that way?" Naming the transition out loud makes a structured call feel considered rather than scripted.

And the one that saves the most calls: "I am going to shut up for a second." Then shut up. In my experience, the most important sentence on any discovery call arrives about four seconds after the rep would normally have started talking again.

Where the reps I watch actually break

It is almost never the questions. Reps who fail discovery usually know the right questions and cannot deliver them under live pressure. They rush the opening because silence is uncomfortable. They accept the first vague answer because pushing feels rude. They hear a compelling event and fail to nail it to a date. They get to minute twenty-two and lose their nerve on the ask.

You cannot fix any of that by reading a playbook again. You fix it by running the sequence out loud enough times that the transitions become automatic and your attention is free to actually listen.

That is the gap DrillCall exists to close — you drill the block you are weakest in against a buyer who pushes back, so the first time you handle a vague answer at minute nine is not on a real account. If I were starting Monday, I would pull the question list for my patch, block twenty minutes, and run the consequence-and-cost segment five times before I dialed anything live.

Structure does not make you robotic. Having no structure is what makes you robotic, because you fall back on the list. Know what each block is for, follow the thread instead of the page, and give the last five minutes back.