The Buyer Interrupts on Slide Two. Now What?

The interruption is not the problem

You are four minutes in. You have done the intro, you have said the thing about how you want to make this useful rather than just talk at them, and you have advanced to slide two. Then someone you have not heard from yet says, "Sorry, before you keep going — does this work with our existing system, or is this a rip and replace?"

Watch what most reps do next. They do one of two things, and both of them are wrong.

The first is the retreat. "Great question, I'm actually going to get to that in a few minutes." Then back to the deck. The rep has technically survived the moment and has, in the same sentence, told the buyer that the agenda matters more than the buyer does. Nobody says anything. The call proceeds. It also ends without a next step.

The second is the collapse. The rep abandons the plan entirely, follows the question into the weeds, and forty minutes later realises they have shown three screens, none of which were the ones that matter, and the two people who actually control the budget have not spoken since the introductions. Everyone says thanks, this was really helpful. Nothing happens.

The interruption itself is neither of those outcomes. The interruption is information you did not have thirty seconds ago and would not have got any other way. Someone in that room cared enough about something to break the social norm of letting the presenter present. That is a signal. In my experience running and sitting in on demos, the calls where nobody interrupts are the ones I worry about, because a silent room is either a polite room or an absent one, and you cannot tell which until the deal disappears.

So the question is not how to prevent interruptions. It is what to do in the three seconds after one lands.

Classify before you answer

Every interruption in a demo is one of three things. Scope question, credibility challenge, or a hidden requirement surfacing. They sound similar. They demand completely different responses. If you answer a credibility challenge like a scope question you will lose the room, and if you answer a hidden requirement like a credibility challenge you will spend ten minutes defending something nobody was attacking.

The classification takes about a second and a half once you have practised it. Here is how each one behaves.

Type one: the scope question

A scope question is someone trying to build a mental frame for what you are showing them. "Is this cloud or on-prem?" "Does this replace what we have or sit on top of it?" "How many users is this priced for?" "Is this the same product your website calls the enterprise tier?"

The tell is that the question is about boundaries, not about value or risk. The person is not doubting you. They are trying to work out which shelf to put you on so they can listen properly. Until they have that shelf, everything you say after slide two is going into a void, because they are still running the boundary question in the background instead of watching your screen.

The response is short. One sentence. Then park the rest and keep moving.

"It sits on top — you keep your current system as the record, we read from it. Happy to go deeper on the integration if it's useful, want me to flag that for the back half?"

That is the whole thing. Answer, offer, move. The mistake reps make with scope questions is over-answering them, because a scope question is the easiest kind to answer well and it feels good to be competent. So you take the boundary question and turn it into a four-minute architecture monologue, and now the CFO in the corner who does not care about architecture has checked out permanently.

One sentence. If they want more they will ask again, and the second ask is a different animal entirely — that is no longer a scope question, that is a requirement, and you handle it as type three.

Type two: the credibility challenge

A credibility challenge sounds like a question and is not one. "We looked at something like this two years ago and it didn't stick." "How is this different from what we already pay for?" "Who else in our industry is actually using this?" "Yeah, we've seen demos like this before."

The tell is tone plus history. There is almost always a reference to the past in it — a previous vendor, a failed rollout, a tool nobody adopted. The person is not asking you to explain a boundary. They are telling you they have been burned and they are checking whether you are going to be honest with them or sell at them.

Do not answer this one and keep going. Stop the demo. Actually stop — stop sharing if you have to, or at minimum stop advancing slides and let the screen sit still. The visual signal matters. It tells the room that you heard something worth stopping for.

Then address it directly, and address it without defending.

"Okay, that's worth stopping on. What happened with the last one — was it a functionality thing or did people just not use it?"

That is the move. You ask about the failure. Not because you need the detail, though you do, but because the fastest way to establish that you are not the last vendor is to be curious about what the last vendor got wrong instead of immediately explaining why you are better. Almost every rep I have watched jumps straight to differentiation here, and differentiation before diagnosis reads as a script.

Let them answer. Usually you get something specific: the data was never clean, the training was one session and then nothing, the champion left. Then you respond to the actual thing they said, honestly, including the parts where you are not sure.

"That tracks. If the data going in isn't clean, this has the same problem — we don't fix that, we just surface it faster. What I'd want to show you is the piece that makes the surfacing worth something, but I'd rather do that than the rest of what I had planned. Can I skip ahead?"

Then — and this is the part reps skip — ask permission to continue. "Does that answer it well enough to keep going, or is there more there?" You are giving them the chance to say no. Most of the time they will say yes and the temperature in the room drops noticeably. Occasionally they say no, there is more there, and that is the best outcome available on the call, because whatever they say next is the real objection and you now have thirty minutes to work on it instead of finding out in an email two weeks later.

Some industries interrupt this way as a default posture. If you sell into carriers, the room is often staffed with people whose job is finding the failure case before it costs them, and the insurance claims demo playbook is written around that assumption rather than treating it as an exception. Same with freight, where the operations lead who has been through two failed TMS implementations is not being difficult, they are doing their job — the freight and 3PL demo script sequences the demo so the breakage question comes up early on your terms instead of late on theirs.

Type three: the hidden requirement

This is the valuable one and the one reps most often mishandle, because it looks like a tangent.

A hidden requirement surfaces when someone interrupts with a scenario. "What happens if the file comes in as a PDF?" "Can you do this for someone who isn't a licensed user?" "We have three locations that run on completely different processes — how does that work?" "Does it handle the case where the customer disputes it after we've already closed the ticket?"

The tell is specificity that you did not prompt. Nothing in your first two slides invited a question about PDF ingestion. They brought it from somewhere. That somewhere is their actual working day, and the thing they described is a requirement that never made it onto the discovery call, either because you did not ask or because they did not think to say.

Drop your agenda and follow it.

I mean that literally. The plan you built was based on what you knew before this moment. You now know more. The plan is out of date. Reps resist this because the deck feels like the safe path and because they are worried about running out of time, but a demo that covers your agenda and misses their requirement is a demo that produces a polite thank-you email.

"Okay, tell me more about that — how often are those coming in as PDFs, and what do you do with them today?"

Let them describe the workaround. There is always a workaround, and it is always worse than they admit at first, and the description of it is the most useful three minutes of the entire call. Then show the thing that touches it, even if it is not the thing you rehearsed, even if you have to click through a screen you do not usually show.

The risk here is real: you can follow a hidden requirement into a dead end where it turns out you genuinely do not handle it. That is fine. Say so. "We don't do that today. What we'd end up doing is X, which is not as good as what you're asking for." You have lost nothing except a fantasy, and you have bought enormous credibility for everything else you say on the call.

The exact language for parking a question

Parking is a skill and most reps are bad at it, because the standard phrasing is transparently evasive. "I'll get to that" means "stop talking." Everyone knows it. It costs you.

A good park has three parts: a partial answer, a named reason to defer, and a commitment with a return point.

"Short answer is yes, it does — the reason I want to come back to it is that the way it works depends on which of these two setups you'd use, and I'll know that in about five minutes. Can I hold it until then and you pull me back if I forget?"

Partial answer: yes, it does. They now have the shelf. Named reason: the answer is conditional on something you have not established yet, which is true and is not the same as "the deck says so." Commitment with a return point: five minutes, and you have handed them the authority to enforce it.

That last clause does more work than the rest combined. "Pull me back if I forget" converts the park from a dodge into an agreement. It also means that when they do pull you back, it is a collaboration rather than a gotcha.

Write the parked question down where they can see you write it. On a shared screen if you have one, on a notepad if you are on camera. The physical act of recording it tells the room that questions do not evaporate here.

And then actually return to it. Unprompted, at the point you promised. "Before I go further — Dave, back to your integration question." If you park three questions and return to zero of them, you have taught the room that parking is a euphemism and everything after that is a negotiation about attention.

One rule: never park a credibility challenge. Type one, yes. Type three, sometimes, if the room is large and the requirement is niche and one person owns it. Type two, never. Parking a credibility challenge confirms the suspicion that produced it.

When the interrupter is the real decision maker

Here is the situation that costs deals. Someone books the call. They are engaged, they are friendly, they have done the discovery with you, they introduce everyone at the top. Ten minutes in, a person who has said almost nothing interrupts with a question that cuts across everything you have shown.

The person who booked the call is often not the person who decides. The interrupter frequently is.

The signals are consistent and they are not about job titles, because titles in the calendar invite lie constantly. Watch for these instead.

The interruption is about consequence rather than mechanics. Mechanics questions are "how does it do that." Consequence questions are "what happens to my team when it does that," "who owns this after go-live," "what does this cost us in the second year." People who ask about consequence are people who will absorb it.

The room defers. This is the loudest signal available and it is entirely nonverbal. Someone interrupts, and the person who booked the call stops talking mid-sentence, or looks at them before answering, or says "yeah, that's a good point" in a slightly different tone. On video you can watch the eye movement. Everyone glances at the person who matters before they commit to a position.

The interruption comes with a scar. "We tried this before" carries organisational memory. Junior people rarely reference failed projects from three years ago, partly because they were not there and partly because it is not their place. The person who invokes history usually owns the history.

And they interrupt in the first place. In most rooms, seniority buys the right to interrupt. The person who feels entitled to cut across a vendor presentation without apologising has usually earned that entitlement internally.

When you spot it, redirect the demo toward that person without abandoning the one who booked it. Something like: "That's the more important version of the question, so let me answer yours first and then come back to where we were." You have flagged the priority without embarrassing your champion. Then keep checking in with the interrupter by name for the rest of the call. "Does that get at what you were asking?" Three or four of those and you will know exactly where you stand.

This is standard in some rooms and rare in others. Plant managers interrupt because they are being paid to run a line, not to watch slides, and the manufacturing demo script for plant managers who interrupt starts from that premise. Principals in real estate do something similar for different reasons — they are the decision, there is no committee, and the real estate demo playbook is built for the version of the call where the person interrupting can also sign.

What to do after you get it wrong

You will misclassify. You will treat a credibility challenge as a scope question, answer it in one crisp sentence, and feel the room go flat.

You can recover, but only quickly. Within a minute or two, go back.

"Actually — I answered your question about the previous rollout too fast. Can I come back to it?"

That sentence has never once cost me anything. It reads as attentive rather than uncertain, because a person who tracks their own misses is a person who tracks other things too. The reason reps do not do it is that going back feels like admitting a mistake in front of a room, which it is, and which is precisely why it works.

What you cannot do is notice at the end and address it in the follow-up email. By then the person has already decided what kind of vendor you are.

Practise this before the call, not during it

The classification has to be automatic. Under pressure, with a screen shared and six faces in a grid, you do not have the working memory to think "is this type one or type two." It has to be a reflex, and reflexes come from repetition rather than from reading about them.

The cheapest version is to sit with a colleague, have them fire twenty interruptions at you in a row — mixed types, no warning — and force yourself to say the classification out loud before you respond. Scope. Credibility. Requirement. It feels ridiculous for the first five and then it starts to click.

If you do not have a colleague with an hour to spare, this is exactly the kind of thing I built DrillCall for: running the same demo opening over and over against a buyer who interrupts on slide two, in different ways, until the three responses stop being a framework you remember and start being something you just do. That is what I would go and do this week if I had a demo booked with a room I could not read.

The buyer who interrupts on slide two is telling you what the call is actually about. Most reps spend the next forty minutes talking about something else.