Key Takeaways

• A leave-behind demo is the follow-up asset a seller sends after a live call, and most of the people who open it were never on that call.
• Call recordings make poor leave-behinds because almost nobody outside the original meeting watches more than a couple of minutes of a 45-minute video.
• A strong leave-behind demo covers one problem for one persona and runs about 20 to 25 clicks, or 2 to 3 minutes for the viewer.
• Feature overload works like carrying pints across a bar: past two or three key areas, buyers do not forget only the last one, they lose all of them.
• Adding a story intro to an existing feature tour does not make it a story. The screens have to be chosen to prove the story, not the other way around.
• Calls to action belong in the middle of an interactive demo as well as at the end, because many viewers never reach the last screen.

The person who decides your deal often never sees your demo. They see whatever your champion forwards them afterwards. In most deals that is a recording link, a slide deck, or a short note saying “this looked good, take a look.” The leave-behind demo is that forwarded asset, and it usually gets a fraction of the effort that went into the live call.

Gartner found that 67% of B2B buyers now prefer a rep-free buying experience, and buyers consult an average of seven information sources per purchase (Gartner, 2026). The live call is one of those sources. The asset that keeps circulating after the call ends is another, and it does a lot of the selling on its own.

This post draws on a webinar Walnut ran with Demodia in November 2026, “From product tour to buyer story.” Simon Harvey (CEO of Demodia), David Bordewyk (Clinical Solutions Strategist at Waystar) and Brian Fogg (Head of Customer Experience at Walnut) spent 45 minutes on what separates demos that move deals from demos that just get watched. What follows is a practical guide to building the leave-behind: who it is really for, how long it should be, how to structure it, and where AI helps. You can watch the full webinar recording here.

What Is a Leave-Behind Demo?

A leave-behind demo is a self-guided version of your product story that a seller sends after a live meeting. The buyer can click through it without a presenter, share it internally, and come back to it when the deal moves forward. The most common form today is an interactive demo (a clickable, guided walkthrough built from real product screens), though plenty of teams still send a call recording or a PDF.

Unlike a website demo or a presenter-driven sandbox, the leave-behind sits in the middle of the deal. It exists because the buying committee is bigger than the meeting.

Brian Fogg put it plainly in the webinar: after a live demo, the leave-behind “might be forwarded on to people that weren’t on the call, so they’re not going to understand what this demo is showing.” That one sentence changes how you should build it.

Why Call Recordings Fail as Leave-Behinds

The default leave-behind is the recording of the call. It is easy to send, it is complete, and almost nobody watches it.

Simon Harvey shared a story from a sales leader who told him that if anyone other than the people in the room watched more than two minutes of their recorded demo, the deal was as good as won. Brian had the same experience from the other side: buyers ask for the recording, he always sends it, and he has always wondered how many go back to a 45-minute call when a two-minute guided version exists.

Recordings fail for three reasons.

• They assume context. A forwarded viewer lands in the middle of a conversation they never heard the start of.
• They carry everything. Every tangent and every “let me just show you one more thing” is still in there.
• They cannot be skimmed. A CFO who cares about one outcome has to scrub through the whole video to find it.

The recording is an archive. A leave-behind has to work as a pitch, and it has to do that without you in the room.

Who Actually Opens Your Leave-Behind?

Build for the forwarded viewer, not the person you sent it to. Your champion already saw the live demo. The leave-behind is for their manager, their finance partner, the end user who will live in the product, and the executive who signs.

Those people share three traits. They did not hear discovery. They do not know your product’s vocabulary. And they will give it a couple of minutes at most before deciding whether it deserves more.

That means every guide step has to explain itself. Brian named the trap that catches the people closest to the product: the “burden of knowledge.” When you know how a platform works, you write tooltips that make sense to you and to nobody else. His advice for leave-behinds was to “shorten them up, make them more concise but more clear,” so the viewer understands the value “instantly from those little bubbles that pop up.”

A useful test: hand it to a colleague who has never seen your product. If they cannot name the problem it solves after one pass, the forwarded buyer will not either.

How Long Should a Leave-Behind Demo Be?

About 20 to 25 clicks, which works out to 2 to 3 minutes for the viewer. That is the guideline Walnut’s customer experience team uses internally, and Brian was specific about how to count: clicks, not screens. Every “next,” every interaction, every branch choice counts.

The number comes from watching what buyers actually do. Brian described the pattern he sees in engagement data for feature-led demos, the ones that read “click next, here you can see this feature, click next, here you can see that one.” Viewers make it through around 23 screens and then drop off, “because they don’t want to know the feature function. They want to know what am I getting from this?”

Simon pushed it further. Some of the best-performing interactive demos he builds are short use-case demos: one persona, one problem, five to ten clicks, two minutes at most. If you have three personas on the buying committee, three short demos beat one long one.

Your own completion data will show your ceiling. Tracking every click tells you the exact screen where attention drops.

The Four-Pint Rule: Why Fewer Features Win

Think of every key feature area you show as a pint of beer the buyer has to carry across a bar. One pint is easy. Most people can carry two without spilling. Three, and they start to struggle. Four, and the problem is not that they drop one. When the first glass slips, the whole lot goes on the floor. Buyers who see too much do not remember the first feature or the last feature. They get confused and remember nothing.

Jargon makes it worse. Simon compared it to smearing washing-up liquid on the glass: every unfamiliar term makes the pints harder to hold. And in a leave-behind, the viewer is very likely seeing your product for the first time.

David Bordewyk’s version of the same rule, when asked what demo builders should stop doing tomorrow: resist the urge to show one more thing. “Show that and stop, and if they have questions they’ll ask or they’ll come back.” He framed it as respect. The buyer gave you their time for discovery. Show them what they told you mattered, then let them get to their next meeting.

For the leave-behind, that means two key areas, sometimes three. Anything else belongs in the next conversation.

How to Structure a Leave-Behind Demo So It Tells a Story

Simon uses a five-part framework that maps cleanly onto an interactive demo. It spells SPORT.

Situation

Introduce the hero and what they want. Not your product. The buyer. “A finance lead who needs to see days sales outstanding at a glance” is a situation. “Welcome to our dashboard” is not.

Problem

Introduce what stands in their way, and do it early. Simon’s point here is the one most demos miss: the problem is where the story begins. Think of any film. Nobody leans forward until the hero hits an obstacle. If the problem never appears, the viewer has no reason to care about anything that follows.

Offer

This is the body of the demo, and the buyer is still the hero. Write the guide steps from their perspective. Not “click here to open the report,” but what the report now lets them do. Every label and tooltip is a chance to say what the buyer gets.

Rallying Point

Give the hero a clear, unambiguous next step. Most feature tours forget this entirely. In a leave-behind, the next step might be booking a follow-up, looping in a colleague, or going deeper on one area.

Transformation

Show what changes. What their week looks like after, and what it keeps looking like if they stay on the current process. This is the part a forwarded executive actually needs.

If you want to pressure-test whether the story you chose is the one buyers want, our guide to demo storytelling covers how to test the order and the stakes against real buyer behavior.

Why You Can’t Paste a Story Onto a Product Tour

The most common shortcut is to take an existing feature tour and add a story intro to the first screen. It does not work, and Brian called it the mistake he chuckles at most.

The original tour was built with a purpose, even if that purpose was “show every feature.” Its tooltips point at specific things for specific reasons. When a new narrative is pasted on top, the story says one thing and the screens still say another. “There’s no really like through line from the beginning to the end. The screens don’t match up with the story and the story doesn’t really show a value.”

Simon sees the same pattern when he audits interactive demos: still 50 screens, still every feature. It is the same harbor tour with a different opening slide.

The fix is to reverse the order of work. Write the story first, using the five beats above. Then choose the screens that prove each beat, and cut everything that does not. If you start from an old template, watch the full flow again and rewrite every step so it matches the new story. Reusing screens is fine. Reusing the old logic is not.

Put a Call to Action in the Middle, Not Just the End

Most demos save the call to action for the final screen. Plenty of viewers never get there.

Simon’s recommendation is to place a decision point partway through as well: “Have you seen enough? Talk to a real person, or keep going.” Without a clear next step, as he put it, “heroes never act.”

Brian’s team goes a step further with branching. At a key moment the demo asks the viewer what matters most to them, and offers two or three paths based on what came up in discovery. Not ten paths. Two or three. The viewer feels the demo is answering their questions, because it is.

Walnut’s Branching playlists are built for exactly this. You ask a question at the start of a Walnut playlist, and each answer sets a completely different path through the demos, videos and documents inside it. A finance viewer and an end user can open the same link and each follow the path that answers their question.

Keep the Data Consistent Across Every Screen

Forwarded viewers are the people most likely to notice when something is off. They were not on the call, so they have nothing to tell them which details matter and which they can ignore.

David’s team at Waystar sells revenue cycle software to hospitals and health systems, which means demoing to clinical audiences who notice when a lab value does not fit the diagnosis. Every demo is built around one realistic patient. The trends, lab results, dates and provider notes all line up, and the patient is the same on every screen. Brian, who helped build those demos, described the standard simply: “If you click somewhere next on the screens, the data is the same, because the patient or the customer is the same.”

David made a second point that often goes unsaid. Inconsistent data does not only lose the buyer. It throws the presenter. When you know the data looks off, you start apologizing with “it’s just demo data,” and the story slips. Clean, credible data works like a script: a well-built persona tells the buyer what the problem is before anyone explains it.

In a leave-behind, nobody is there to explain an odd number. One customer, one persona, one dataset, start to finish. Building personalized demos for each deal is how you make sure that dataset reflects the buyer, not a generic sample account.

Where AI Helps Build a Better Leave-Behind

AI has changed how this work gets done, and both practitioners described the same shift. In the past, building a demo story meant writing patient histories or customer scenarios from scratch, often in a spreadsheet. Now, as David put it, you get to be “more of an editor” than a creative writer.

Brian’s workflow starts with what came out of discovery. He gives an AI assistant the persona, what they said on the call or by email, and asks it to draft a value-based story for that persona. Then he moves into AI Mode inside Walnut, gives it the story, and has it rewrite the guide steps on the existing screens so every tooltip supports the story and points at the right place.

Simon has seen teams give an AI model an existing demo, the five-part framework and a description of the ideal customer, then ask which problem to introduce and which four or five clicks prove the solution. It does a good job of turning a 50-step tour into a short story.

What AI does not do is decide which problem matters to this buyer, or which two pints they can carry. That judgment comes from discovery, and it stays with the person who ran the call.

Who Should Own the Leave-Behind Demo?

Demo ownership is, in Brian’s words, “a heated topic.” His answer was to let subject matter experts build the demos, whichever department they sit in, because they understand both the value and how the product shows it. David added that the people who present a demo repeatedly should have the final say. Presenting someone else’s demo is like presenting someone else’s slide deck.

The model that scales is for experts to build the core story once and let everyone else reuse it, which is why SEs should author once and everyone else should reuse. A rep should be able to send a strong leave-behind without first becoming a product expert.

A Quick Leave-Behind Demo Checklist

Before you hit send, check the demo against this list.

• It focuses on one persona and one problem from discovery.
• It covers two key areas, three at most.
• It runs 20 to 25 clicks or fewer.
• The problem appears in the first few steps.
• Every guide step makes sense to someone who was not on the call.
• No step uses jargon the buyer would not use themselves.
• The customer, persona and data are consistent on every screen.
• There is a call to action partway through, not only at the end.
• The final screens show the transformation, not another feature.

Frequently Asked Questions

What is a leave-behind demo?

A leave-behind demo is a self-guided version of a product demo that a seller sends to a buyer after a live meeting. The buyer can click through it on their own and share it with colleagues who were not on the call. The most effective leave-behinds are short interactive demos focused on the one problem the buyer raised in discovery, rather than full call recordings.

How long should a product demo be?

For a self-guided or leave-behind demo, aim for about 20 to 25 clicks, which takes a viewer 2 to 3 minutes. Very focused use-case demos can be as short as 5 to 10 clicks. Set length by the one story you need to tell, not by how much the product can do.

Should I send the call recording after a sales demo?

You can, but do not rely on it. Few people outside the meeting watch more than a couple of minutes of a long recording. A short interactive demo built around what you discussed is far more likely to be viewed and forwarded to the rest of the buying committee.

How do I turn a feature tour into a demo story?

Start from the story, not the existing screens. Define who the buyer is, what they want, and the problem stopping them. Then decide what they get, what next step you want them to take, and what changes for them. Only then choose the screens that prove each part, and rewrite every guide step so it matches the new story. Adding a story intro to an unchanged feature tour does not work.

What is feature overload in a demo?

Feature overload is when a demo shows more key areas than a buyer can absorb. Beyond two or three, buyers do not just forget the extra features. They lose track of the whole demo and remember very little of it. Cutting to the two or three areas that prove value for that specific buyer leads to better recall and a clearer next step.

Can AI help build a leave-behind demo?

Yes. AI is useful for drafting a persona-specific story from discovery notes, simplifying jargon-heavy guide text, and suggesting which problem to lead with and which clicks prove the solution. The human still needs to decide which problem matters to this buyer and to check that the story, screens and data line up.

Build for the Person You Never Met

The live call is where you earn the champion. The leave-behind is where the champion tries to win over everyone else, using whatever you gave them. Build it for the person who wasn’t in the room: one problem, two pints, a story that explains itself, and a clear next step. That person may well be the one who signs.

Ready to see what a focused, personalized demo can do for your pipeline? Start for free with Walnut.