Conference recap video: three days of footage into one cut

Most conference recaps are souvenirs. They prove the team went somewhere. The useful version is planned before the flight and cut while the event is still being talked about.

September 14, 2026 · 9 min read

The team comes home with four hundred photos, eleven vertical clips at different exposures, and a voice memo of most of a panel. Someone is asked for a recap. Two weeks later a video appears: a wide shot of a busy aisle, the logo on the backdrop, three seconds of a handshake, an upbeat track, a card at the end that says thank you.

It proves the team attended a conference. That is all it does. Nobody who was not in the hall learns anything from it, and the people who were there already know. It gets a few likes from staff and the agencies who were at the same event, then it stops.

This is the cheapest content a trip produces and the only asset that reaches the majority of your market, which did not fly anywhere. It ends up as a souvenir because nobody decided what it was for before the flight, and by the time the footage is on a drive, that decision can no longer be made.

A recap is an argument, not a souvenir

The default mode for filming an event is memory: get a bit of everything, sort it out later. That is how holidays are filmed. It produces material that has no centre, because every clip is equally important and therefore none of it is.

A recap that earns its edit time makes one claim. There are only three worth making, and they need different footage, which is exactly why the choice cannot wait until the edit.

• The room took us seriously. The stand full, the queue, the talk with people standing at the back. This is social proof for the people who have never heard of you.
• We said something there. You announced a mainnet date, a listing, a partner. The recap is that announcement for the ninety five percent who were not in the room.
• There are people behind this. Faces, names, two founders answering a real question. Worth more than either of the above for an anonymous protocol.

Most crypto teams should pick the second one, because a conference is usually booked around a moment that already exists. The recap then has a spine: here is the thing we said, here is the room we said it in, here is what happens next.

The shot list is decided before the flight

Unbriefed footage is what makes a recap impossible. Not bad footage, unbriefed footage: forty clips of the same aisle and nothing of the moment that mattered. A shot list fixes this in about twenty minutes of work, done at the desk, weeks before anyone packs.

Then assign it to one named person, in writing, with a charged phone, a spare battery and permission to interrupt conversations. Not the whole team, not a shared album. A group instruction to film things produces the same aisle six times and none of the talk.

That person needs time budgeted for it, which means they are not also the one running demos for three days. This is the point where most capture plans quietly fail: the job exists on paper and belongs to somebody who is already busy. Deciding what the booth itself needs is a separate exercise, covered in our post on crypto conference booth content.

The nightly pass beats the post event pile

A conference is three days long, which means you get two chances to fix what you missed. Nobody takes them, because the footage is not looked at until everyone is home.

1. 01Offload every night. Phone to laptop to cloud, in one folder per day. Storage full on day two is the most common way footage is lost.
2. 02Log the three best moments while you remember them. A note with the file name and a few words is enough. Nobody can reconstruct this two weeks later from a timeline of near identical clips.
3. 03Flag the one clip the cut cannot exist without. Usually the announcement, or the room at its fullest. If it does not exist yet, you now know tomorrow has a job in it.
4. 04List what is missing and shoot it the next morning. Empty stand plates, a founder saying the one line properly, the signage. The hall is quiet before the doors open and the light is better.
5. 05Send the day one selects to whoever is editing. They can start assembling on day two, which is how a recap goes out in the window rather than after it.

This is the whole difference between a recap that is assembly and a recap that is archaeology. The footage is the same; one version has been sorted by someone who was standing there.

One capture effort, three cuts

Teams tend to imagine one recap video. What a properly shot event actually yields is three deliverables from a single capture effort, each landing somewhere different, and none of them requiring a second production.

• The vertical cut, twenty to thirty seconds. Muted, subtitled, posted during the event or within two days of it. This is the one that carries the reach.
• The follow up film, sixty to ninety seconds. Sound on, sent in the email to everyone the team met, where a person is sitting down and has chosen to watch.
• The stills set. Ten to fifteen frames that keep working for weeks as post images, deck slides and site proof, long after the video has stopped circulating.

One build, several cuts, is the same economics that makes a quarter of content affordable rather than three separate shoots, and it only works when the cuts are named in the brief. Asked for afterwards, each one is a new job at a new price. We lay out how those slots fill a month in our post on the video content calendar for startups.

How the cut is structured

A recap fails in the first two seconds more often than anywhere else, because it opens the way a corporate video opens: logo sting, slow fade, establishing shot of a building. On a feed that is three chances to scroll before anything has been said.

1. 01Open on the room at its fullest. The busiest frame you have, first, with no build up. It is the only shot that does the social proof job on its own.
2. 02Put the claim on screen as text inside three seconds. The announcement, the number of meetings, the name of the event. Sound is off, so text is the voice.
3. 03Give the announcement one clean line of real audio. Ten seconds of a founder saying it on stage beats a caption saying the same thing.
4. 04Cut in one human moment. A reaction, a laugh, two people arguing at the screen. This is what separates a recap from a slideshow of a stand.
5. 05End on what happens next, not on thank you. A date, a link destination, the next city. A recap that ends in gratitude asks for nothing.

Subtitles are not optional and neither is the vertical framing on the feed cut. Both are decided in the edit, so both are free if the shoot was framed loosely enough to allow a reframe, and expensive if it was not. That is another reason the shot list matters more than the camera.

Publish inside the window

Conference attention has a shape: it peaks during the event, holds for two or three days while people post their own photos, and is gone inside a week. A recap that arrives two weeks later is competing with nothing, which is why it looks like it performed badly. It was simply late.

So cut the short one first. The order is almost always reversed: the long film gets the attention because it feels like the real deliverable, and the feed cut is squeezed out of what is left at the end. Invert that. The vertical goes out inside forty eight hours, the follow up film lands in the week after, and if something has to slip, it is the long one.

The same week usually has a second, unrelated need: the announcement you made on stage wants a feed asset that is not event footage. Worth naming the fast option honestly, since it is ours: ShipTeaser, built by the same founder as this studio, turns a product URL into a 15 second 1080p motion graphics teaser for muted feeds, with an optional music bed and the first one free without a card. It never touches your footage, it has no timeline and no voice-over, so it is no use for the recap itself. It covers the product side of that week while the edit is still open, which is often exactly what is missing.

After that, the footage keeps paying out quietly. The stills feed the weekly rhythm we describe in crypto community content, and the strongest frames end up in the case work you can see across the work.