How long should a LinkedIn video be
Everyone wants a number of seconds. There is no honest single number, but there is a method that gets you to the right length for each video you make. It starts with counting ideas, and it ends with reading who replied.
Count the ideas first
A video has one job. It carries one idea from your head to a stranger’s. If you can say the idea in a sentence, you know what the video is about, and everything in it either helps that sentence land or gets in its way. Length is what is left after you remove the things that get in the way.
Say your idea out loud before you record. Suppose it is that your first ten customers all asked for the same missing feature. That sentence takes four seconds. The video then needs the story of how you found out, one customer’s words, and what you did about it. That is perhaps forty-five seconds to ninety seconds of talking. If you find yourself explaining a second idea, you have two videos. Record two.
A working range for founders
Most founder videos that work sit between thirty seconds and two minutes. Under thirty seconds there is rarely room for a specific, and a video without a specific reads like an ad. Over two minutes you are usually either telling two stories or warming up. Neither needs the viewer’s time.
Treat that range as a habit. A customer story with a real turn in it can run three minutes and hold. A one-line opinion can run twenty seconds. The range is a place to start when you have no better information, and your own results will move it. After a few weeks you will see which lengths bring replies from the people you want, and that is the only length data that matters to you.
Where the time goes
Most of the wasted seconds sit at the start. Founders open with a greeting, a name, a statement of what they are about to talk about, and a breath. The viewer decides in the first few seconds whether to stay, and none of that gives them a reason. Open on the most concrete sentence you have: a number, a name, or something a customer said. Then tell the rest in the order that makes sense.
The second place is the end. When the point has landed, stop. Do not summarize what you just said, and do not thank people for watching. A clean stop feels confident. Trimming the first five seconds and the last five seconds often removes a fifth of the video without losing a single idea.
The third place is the middle, where you repeat yourself. We repeat when we are not sure the point landed. If you said it clearly once, cut the second telling. If you did not say it clearly, redo the take and say it clearly once.
Captions and the muted viewer
Many people watch with the sound off, at work or on a train. Captions turn your video into something they can read at their own speed, and the length of the video then matters a little differently. A viewer reading captions takes in words at reading pace, which is slower than a fast talker speaks. If you speak very quickly, trim harder, or slow down.
Put the hook in the captions as well as in your voice. The first line a muted viewer reads should be your most concrete sentence, large enough to read on a small screen. If the first caption is a filler word or a greeting, you have spent the moment you had.
A simple way to test length
You can learn your own best length in a month without any tools. Pick one kind of take, such as a customer story, and record it at two lengths across four weeks. Make the short version ahead of the long version one week, and swap the next. Swapping matters because the day and the topic change what happens, and you want to spread that around.
Do not compare views. Compare the people. Write down who replied, who clicked, and who followed, and note whether they look like your customers. If the long videos bring more of the right people, they are the right length for you, even when the short ones are watched by more strangers.
Be willing to be wrong about your guess. Many founders expect the short video to win and find that a slower, fuller story builds more trust with the people they want. Others find the opposite. Both results are fine. The point is to stop arguing about it in the abstract and let your own audience answer.
Keep the test small. Four weeks, one kind of take, two lengths. When it ends, set your range from what you saw, write it on a card, and get back to the more interesting work of having something worth saying.
Judge length by who replied
Views tell you how many people started watching. They do not tell you who stayed to the end or who cared. A reply from a person who matches your customer is a much better sign. Keep a short list of your videos, their lengths, and the names of people who replied or clicked. After ten or twenty videos you will see a pattern, and you may be surprised by it. Many founders find their longer customer stories bring the replies and their short opinion clips bring only views.
Then run the cheapest experiment there is. Record the same kind of take at two lengths over two weeks, and compare the names, not the counts. Use that to set your range. The method works because it starts from your audience and your product, which no general advice can know.
If you want a way to do this without a spreadsheet, Quant100 gives you one question a day, cuts the video, and sends a receipt the next morning that lists who came and which take brought them. With a few weeks of receipts, the length question usually answers itself.