What do people actually do when you send them a demo?

What people do with a demo link, from never opening it to passing it on: where each behaviour shows in your own room, and what to do about it.

On this page

What people do, and where you see it
Why are there no benchmarks in this guide?
Do people open the link?
How long do they spend, and on what?
How far do they take the tour?
Do they come back and pass it on?
What do they choose next?
What the room does not see
What should you read first in your own room?
What should you do with this?
Questions people ask
Can I compare my figures with other people’s rooms?
Is time counted when the tab is in the background?
Does the room record clicks inside my demo?
Can I see all this on Free?
About the author
Related guides

By Kaz Kato · Published 13 October 2026 · Updated 13 October 2026

Some open it at once, some days later, and some never. Those who open it look at a screen or two, take part of the tour or go off on their own, and some come back or pass it on. Your room records which of those each viewer did, person by person, so read that rather than anyone’s average.

This is a field guide to those behaviours. For each one it says what it looks like, where it shows in Roomi, and what to do about it. Every measure is defined in analytics and the pipeline, and sharing a link covers how each person gets a link of their own.

What people do, and where you see it

Each row is one thing a viewer can do with your demo, what it looks like in the room, where you see it, and what to do about it.

Behaviour	What it looks like	Where you see it	What to do
Did not open it	The link still in Sent	The pipeline	Check it reached them, then resend with a line of context
Opened it once, briefly	One short visit	Their journey	Wait for a second visit before reading much into it
Opened it on a phone	A visit made on a phone	Their journey	Check the first screen reads well in a phone frame
Finished the tour	“Finished the tour”	Their journey, and the tour’s drop-off	Ask what is missing
Stopped part way	“Took the tour to step 3 of 5”	The tour’s drop-off	Rewrite the step where people stop
Left the tour to explore	Left it to explore on their own	The tour’s drop-off, and their journey	See what they went looking for
Read a page	Time on a page under Read more	Pages opened, and their journey	Answer what that page raises, such as price
Read a PDF	Its time, page by page	Analytics	Follow up on the pages they stayed on
Ran a scenario	The scenario, by name	Scenarios run, and their journey	Talk about the part they tried
Came back	A visit an hour or more after the last	Their journey, and Hottest right now	Call while they are thinking about it
Passed it on	A new name under the link, marked Forwarded	Their pipeline card, and Activity	Offer to walk the new person through
Left a comment	A comment pinned to the screen it is about	The Feedback tab	Reply: they read it in the room next time
Reacted	A heart, a clap or a fire, by name	The Activity tab	A light touch, not a call
Answered a next step	Their answer, with its detail	Next steps chosen, and the pipeline	Do what they asked
Went quiet	No visit since the last one	Their journey and pipeline card	Set a follow-up date on the card

Why are there no benchmarks in this guide?

Because the only numbers worth acting on are your own, and your viewers’ data is not ours to pool. Roomi uses what your viewers do to run your rooms and to show you what they did. So this guide tells you where to look in your own room, not what other people’s rooms looked like.

That is what our terms say, in the section on viewer data. Where I give a view of my own below, I say that it is mine. It is a view, not a measurement.

Do people open the link?

Not all of them, and not always soon. A link can sit in an inbox for days, be opened on a phone between meetings, or never be opened at all. Look at how many of your links were opened, and how soon after you sent each one. That tells you whether the link landed, before anything else.

Where you see it. On the pipeline, a link nobody has opened sits in Sent, and one visit moves it to Opened. A project’s Overview counts how many people it was shared with, and how many opened it.
How soon. Each viewer’s journey lists their visits with when each began. Compare the first with the day you sent the link.
What to do. A link still in Sent after a while has not landed. Check it went to the right person, then send it again with a line on why it is worth a look. Copy link beside their name gives you the same address.

An open is not a read. One short visit shows the link arrived, not that it held them.

How long do they spend, and on what?

Time inside is counted per app and per page, only while the room’s tab is visible, so it measures attention rather than an open tab. In your own Analytics, Apps used and Pages opened show what held people. In each journey, you see where one person’s time went, visit by visit.

Apps used lists time per app, largest first, with how many viewers used each.
Pages opened lists opens per page, with how many viewers opened it and how long it was read.
Where attention goes, on a project’s Analytics tab, is a heatmap of time on each app and page, a day at a time, with each viewer’s own split below it.
A journey shows each visit’s time split across apps and pages, and its moments in order, such as “Read Why Corner” and “Left from Corner HQ”.
A PDF in the room, such as a brochure, has its time kept page by page, so you see which pages were read and for how long.

Long time on a screen is not always interest. It can be someone stuck. A comment pinned there, or a quick exit after it, tells you which. A short visit on a phone can be a first look rather than a verdict, so see whether a second visit follows.

How far do they take the tour?

Some finish it, some stop part way, and some leave it to explore on their own. In your own Analytics, the tour’s drop-off shows how many people reached each step, in order, so the step where people stop stands out. Each journey says how far that one person went.

The drop-off sits beside how many people started the tour, reached its end, and left it to explore. A person counts once at each step they reached, however many visits it took.
A journey says “Finished the tour”, “Took the tour to step 3 of 5” or “Has not taken the tour”, across every visit.

A step where the count falls away is the one to rewrite. Make it say one thing, show one screen, and stand on its own. The demo tour script template has the shape. Leaving the tour is not a failure, either: someone who goes off on their own is looking for something, and their journey shows what.

Do they come back and pass it on?

Some do, and those are the people to call. A return is a visit an hour or more after the one before. A pass-on shows when someone opens the link and says they are someone else. Both are recorded against the link you made, so you see who came back and who brought in whom.

Coming back. Each journey lists every visit. Hottest right now ranks whoever was seen most recently first, then whoever came back most, with the evidence in words, such as “seen today, back on Pricing”. The app also tells you when someone comes back to the room.
Passing it on. Every browser that opens a link is asked “Are you Priya Shah?” first. Someone who says no gives a name, and appears under the link as Forwarded, on its pipeline card and in its Activity.
The same name, more than once. “Opened as Priya Shah in 3 separate sessions” can be Priya on another day or device, or someone she passed it to. The room cannot tell which.

With a return, call while they are still thinking about it. With a pass-on, offer to walk the new person through. The buying signals that a prospect is ready to talk ranks these against each other.

What do they choose next?

Some answer one of the next steps you offer, and some leave without answering. The steps run from “Book a meeting” to “Not for me”, and you choose which to offer. In your own Analytics, Next steps chosen counts the answers to each, and every answer moves that person’s card on the pipeline.

“Book a meeting” or “Yes — send the proposal” moves them to Said yes.
“Get back to me” comes with when: in a week, a month or next quarter. They land in Later, so set the follow-up date they chose.
“Bring in a colleague” carries the colleague’s name. “Ask a question” carries the question.
“Not for me” can carry a reason: price, timing, fit, or already solved. The reason is worth more than the no.

No answer is not a no. With no answer, the pipeline works out a stage from what they did: two visits, or two minutes in the room, is Engaged.

What the room does not see

The room tells you what each viewer did in your demo, and nothing about them beyond that. Knowing its edges keeps you from reading too much into it.

No clicks inside the demo. It counts time per app and page, not clicks or keystrokes.
No address or location. It records no IP address and no browser details, so it cannot tell you which company a visitor is from.
No background tabs. A tab left open behind other windows adds no time.
No checked names. A name is the one you gave the link, or the one a newcomer typed.
Nothing from your own preview. Preview as viewer records nothing, so your own checks never count.

What should you read first in your own room?

Read the people before the totals. My own order starts with whoever answered, then whoever came back or passed it on, then comments waiting for a reply. Next come links nobody has opened, and only then the tour and where time went. That order is my view, not a measurement: it puts the calls first and the fixes after.

Answers — someone who booked a meeting, said yes or asked a question has told you what to do.
Returns and pass-ons — they went away and came back, or brought someone in. Call them next.
Comments — each one is pinned to a screen and waiting for a reply.
Links nobody has opened — resend with a line of context, before the link goes cold.
The tour’s drop-off — the step where people stop is your next edit.
Where time went — the apps and pages that held people, for the next build.

Your own Claude can read the same records through the connector. Ask it who has not opened, or who came back, and it answers from your room.

What should you do with this?

Act on what each person did, one person at a time. Call the people who answered, came back or passed it on. Resend to the ones who have not opened. Then fix the tour step where people stop and deploy again: everyone opening a link gets the new build, and nobody’s link changes.

For what to say when you call or write, see how to follow up after a demo. Pricing shows which plan carries each person’s journey and the pipeline.

See what your own viewers did.