Discovery Call vs Demo: Stop Running One Call That Does Both Badly

I have watched a lot of reps run what they call an "intro call." Thirty minutes. A few questions at the top, then a screen share, then a scramble at the end to book something. The rep hangs up and writes "good call, sending recap" in the CRM. Then nothing happens. The prospect goes quiet, the deal sits in stage two for six weeks, and eventually gets pushed to next quarter and then closed-lost with the reason "no decision."

That call is the problem. Not the follow-up, not the sequence, not the pricing. The call. You ran discovery and a demo in the same thirty minutes and did both of them badly.

I want to lay out what each call is actually for, who is supposed to be talking, what you are allowed to say on each, and how to tell after the fact that you demoed too early. And then the part everyone actually wants: what to say when the buyer opens with "can you just show me the product."

Why the merged call exists in the first place

Nobody merges these two calls because they think it is good practice. They merge them because of pressure, and the pressure comes from three directions at once.

The first is quota math. Two calls means two meetings booked, two calendars aligned, two chances for someone to no-show. One call means one. When you are behind, halving the number of things that can go wrong feels like a win. It isn't, but it feels like one.

The second is buyer impatience. Buyers ask to see the product. They ask early and they ask directly, and turning that down feels like the fastest way to lose the deal. So reps cave. I have caved. It is a very natural thing to do when someone with budget says "just show me."

The third is that the demo is the comfortable part. Discovery is uncomfortable. You are asking a stranger to describe something that is broken inside their company, which means asking them to admit something is broken, which many people will not do with someone they met four minutes ago. The demo has no such friction. You know the product. You have the clicks memorised. When a call gets awkward, the screen share button is right there, and sharing your screen is the fastest way to stop feeling awkward.

So the merged call is not laziness. It is three reasonable instincts stacking up into one bad meeting.

The two calls are doing completely different jobs

Here is the cleanest way I know to think about it. Discovery is a call where you are trying to find out whether a problem exists and what it costs. The demo is a call where you show that your product removes a problem you have already agreed exists. One is diagnosis. The other is prescription. You cannot prescribe before you diagnose, and if you try, the prescription sounds like guessing, because it is.

The discovery call

The goal of discovery is to leave the call able to write, in the buyer's own words, what is broken, who it hurts, what it costs them, and what happens if they do nothing. That is it. Not a demo booked because the meeting has to end somehow. Not a "they seemed interested." Four things, in their language, that you could read back to them and have them nod.

On a good discovery call, the buyer does most of the talking. If you finish a discovery call and you are the one who is hoarse, it went wrong. Your job is questions, silence, and follow-ups on the answers that were vague. The follow-up on the vague answer is where the deal actually lives. When someone says "our current process is a bit manual," the deal is not in that sentence. It is in the answer to "walk me through the last time it caused a problem."

What you are allowed to say on discovery: how other companies in their world have handled the same thing, at a level of abstraction that does not require your screen. What you are not allowed to say: anything that starts with "so what we do is." You are not there to be understood. You are there to understand. There is time later.

The structure matters more than most people think, and it changes by vertical because the buying committee changes by vertical. When I write out what a diagnostic looks like for a law firm buyer, the whole thing is organised around matter volume, write-offs, and who carries risk, because that is what a managing partner actually feels — I laid that out question by question in the 25-minute diagnostic for law firm buyers. The equivalent conversation with a distribution network is nothing like it. Asset age, outage exposure, regulatory reporting, and a completely different definition of the word "risk," which is why the energy and utilities discovery playbook asks about failure history rather than billing.

The demo call

The goal of the demo is narrower than most reps believe. You are not showing the product. You are showing the three or four things that solve the specific problems the buyer described to you on the previous call, in the order they described them, using their words as the labels.

On a good demo, you talk more than the buyer, but not that much more. If nobody interrupts you for eleven straight minutes, you are not demoing, you are presenting, and presenting is what happens when you did not know enough to demo. The best demos I have been part of get interrupted constantly, and the interruptions are all versions of "can it also do this," which is the buyer mentally installing your product in their building.

What you are allowed to say on a demo: everything. Pricing directionally, roadmap carefully, implementation honestly. You have earned the right to talk because you spent the last call listening.

What you are not allowed to do is show anything you cannot tie back to something they said. Every screen needs a sentence in front of it that starts with "you mentioned." If you cannot construct that sentence, cut the screen. That discipline is what keeps a demo from turning into a tour, and it gets harder the more people are in the room, because each person needs their own version of it. Running a demo for a managing partner, a litigation chair, and a risk committee at the same time is a genuinely hard piece of stagecraft — three different definitions of value in one Zoom window — and I broke down how to sequence it in the legal demo script.

Three signs you demoed too early

You will not always know in the moment. You will know afterwards, and these are the three tells I look for when I am reviewing a call that went nowhere.

The buyer asks feature questions with no context around them

"Does it integrate with Salesforce?" "Can you export to Excel?" "Is there an API?"

Those look like buying signals. They are not. They are the questions people ask when they have no frame for what they are looking at. A buyer who has told you their problem asks contextual questions instead: "when our paralegals do that at 6pm on a filing deadline, does it still hold?" That question has a person, a time, and a stake in it. It could only be asked by someone who has already mapped your product onto their own building.

The bare feature question is the buyer trying to be a good participant in a meeting they do not understand the purpose of. They are being polite. Politeness is not pipeline.

There is no named business problem in your notes

After the call, open your notes and try to write one sentence in this form: [Named person or team] cannot [do specific thing] because [specific cause], and it costs them [specific consequence].

If you cannot fill in all four slots from what you actually heard — not what you assumed, not what you know about their industry, what they said — then you did not run discovery. You ran a product overview with a warm-up.

This is the sign that catches the most experienced reps, by the way, because we are good at filling the gaps ourselves. We know the vertical. We know what a firm that size is probably dealing with. So we write the sentence from our own knowledge and it reads convincingly and it is entirely our invention. The buyer never said it. And when the buyer takes your business case to their CFO, they cannot repeat a sentence they never said.

There is no second stakeholder

The hardest tell and the most reliable one. If you demoed and you are still talking to exactly one person, the demo did not work, regardless of how it felt.

A demo that lands produces a name. Someone on the call says "I'd want Priya to see this" or "the risk committee would have questions about the audit trail." That is the sound of a deal becoming real, because it means the buyer has started doing internal selling on your behalf, and internal selling is the only kind that closes anything.

When no second name appears, one of two things is true. Either the problem is not big enough to justify pulling a colleague into a meeting, or your champion cannot articulate the problem well enough to convince a colleague to come. Both of those are discovery failures. Neither is fixed by demoing again, which is what most reps do next.

"Can you just show me it?"

This is the moment. It usually arrives in the first three minutes, sometimes in the calendar invite reply before the call even happens.

The wrong response is to refuse outright. "I'd love to, but first I need to understand your business" is a sentence buyers have heard from every vendor who ever called them, and it tells them the next twenty minutes are for your benefit, not theirs. You will get short answers and a call that ends early.

The other wrong response is to say yes and share your screen. Then you are demoing blind, and blind demos are where reps show the seven features they find most impressive, none of which the buyer cares about.

What works is naming the trade honestly. Here is roughly how I say it:

"Yes, absolutely, and I'd rather show you the part you'll actually use than the whole thing. Give me ten minutes of questions first so I know which three screens matter to you, and I'll spend the rest of the call on those. If it turns out we're not a fit, I'll tell you in those ten minutes and give you the twenty back."

Three things are doing work in there. You said yes, so you are not the vendor who withholds. You gave a reason that benefits them, not you — they get the relevant screens instead of the generic tour. And you offered to disqualify yourself, which almost nobody does, and which buys you an enormous amount of credibility in a single sentence.

Then you have to honour it. Ten minutes means ten minutes. If you are still asking questions at minute nineteen, you have taught them their time does not matter, and they will not give you a second call.

The compromise: five minutes of screen, earned by ten minutes of questions

When you genuinely cannot get a two-call process — inbound demo request, competitive deal, a buyer who will simply not book twice — run this instead of the merged mush.

Ten minutes of tight questions. Not fifteen. Ten. Which means you cannot ask everything, so you ask the ones that determine what you show: what triggered them looking now, what happens today without you, who else is affected, and what the consequence is of leaving it alone another quarter.

Then five minutes of screen. Only five, and you say so out loud: "I'm going to show you three things, about five minutes, and then I want to hear which one you'd want to look at properly." The constraint is a feature. It forces you to pick, and picking is the whole skill. A five-minute demo built on ten minutes of real answers beats a twenty-five minute tour every time, because the buyer sees their own problem on the screen instead of your product.

The last ten minutes are for the next meeting, and the next meeting is the real demo, with the second stakeholder in it. That is what you are trading five minutes of screen for. Not goodwill — an invitation to the person who has not been in the room yet.

Some verticals make this harder because the technical depth arrives immediately. Show a distribution network engineer an asset risk model and you will be in the weeds on data sources inside ninety seconds, which is fine if you planned for it and fatal if you did not; I set out how to hold that line in the energy and utilities demo script.

What to change on Monday

Go back through the deals that stalled last quarter and check them against the three signs. Feature questions with no context. No named business problem in the buyer's own words. No second stakeholder. My guess is most of them fail at least two, and the common cause is one call trying to do two jobs.

Then fix the calendar before you fix the script. Book discovery as its own twenty-five minute meeting with an agenda in the invite that says questions, not demo. Buyers accept that far more often than reps expect, because a twenty-five minute call with a clear purpose is an easier yes than a vague hour.

The part that takes actual work is the ten minutes of questions, because holding your own line while someone with budget asks to see the screen is uncomfortable, and you will not get good at it by reading about it. That is the specific thing I built DrillCall for — running the same discovery opening again and again against a buyer who pushes back, until saying "give me ten minutes and I'll show you the three screens that matter" stops feeling like a risk and starts feeling like the obvious thing to say. If I were starting over as an SDR, that is the rep I would rather be.