How long should a before and after video be?

The honest answer is that nobody watches a clock. They watch until they stop caring, and a before and after video buys attention in a very specific way: you have promised a reveal, and they are waiting for it.

So the useful question is not how long the video should be. It is how long you can make someone wait for the reveal, and what you owe them while they wait.

The shape, not the duration

Every reel BeforeNow builds has the same five parts, in the same order. They are worth knowing whether or not you use the app, because the shape is what does the work.

Hook. The first beat. Not a title card, not a slow zoom on nothing. Something that states the promise: the worst shot, the honest line, the thing that makes the change worth watching.

Build. The middle. The work, the ordinary days, the setback. This is the part that makes the result believable and the part everyone wants to skip.

Reveal. The turn. Before becomes now. This is the moment the whole video was buying time for.

Payoff. What the reveal means. A number, a reaction, a sentence you could not have said at the start.

Outro. Get out. One beat, then stop.

Notice that only one of those five is the after shot. If you cut a video that is mostly the after shot, you have made an advert, and people are very good at not watching adverts.

Where the reveal goes

Put the reveal roughly two thirds of the way in.

Earlier than that and there is nothing left. You have shown the answer and the rest of the video is a person talking after the punchline. Later and you have run out of goodwill: the build has to hold attention on its own, and the build is the weakest material you have.

Two thirds gives you a long enough middle to make the change credible and enough room after the reveal to say what it meant. If your video is twenty four seconds, the reveal lands around sixteen. If it is forty, around twenty seven.

Everything else follows from that. You are not choosing a duration, you are choosing how much build your footage can carry before it starts repeating itself.

What that means in practice

Most people's footage supports twenty to forty seconds. That is not a rule, it is an observation about how much genuinely different material a normal person films.

If you have three clips, you do not have forty seconds. You have maybe fifteen, and stretching it will show. Three clips cut tight beats three clips padded out with slow motion and a title card.

If you have three months of daily captures, you still probably do not want three minutes. You want the four or five moments that are actually different from each other, which is usually the start, one bad day, one turning point, and the now.

The test for any shot in the middle: does this show something the previous shot did not? If not, it goes. Two similar clips do not add up to twice the evidence, they add up to a viewer who has worked out that nothing new is coming.

What to cut first

In order, when a video is running long:

1. The setup. Anyone explaining what they are about to do. Start after that sentence.
2. Duplicate progress. Week three and week four look the same. Keep one.
3. The second half of every clip. People finish sentences long after they have finished the point.
4. The outro. If it is more than one beat, it is a goodbye nobody asked for.

What survives all four cuts is usually the video.

The part length cannot fix

A video that is the wrong length is a fixable problem. A video where the before and the now do not match is not, because the reveal will not land no matter how well you time it. That is worth getting right at capture, and it is a different problem: see how to film a before and after video people actually watch.

If you already have the footage and you would rather not make these decisions yourself, this is what the editor is for. It reads what you said, finds the five parts, and puts the reveal where it belongs. Try it with your own clips at beforenow.ai/create, or start from the Before & After Video Maker if you just have the two shots.