A description is not a menu
One link, chosen because it matches what the video promised, usually out-pulls a description full of them; a strong default worth testing on your own channel.
Open your last video, expand the description and count the links. You already know roughly what the number is, and that it is too high, because every one arrived the same way: something new happened, a line got added, and nothing has ever come off.
That list is being read at one particular moment. Somebody has just watched you do the thing, they are mildly convinced, and for a moment they are willing to act. What they meet at that moment is a menu. Menus produce browsing, and browsing at the end of a video produces the same result as closing the tab.
Why does a list of links stop somebody who was about to act?
Because choosing is work, and they were not intending to do any.
The decision the viewer had just made was small and specific. They wanted the thing you mentioned. When they arrive at a column of eleven options they have to re-open a decision that was already closed, work out which one is the one, and rank the others in case they were wrong. None of that is difficult. All of it is friction, and it lands where you can least afford any.
There is a second cost people miss. Every link on that list is a claim about what matters, and eleven claims of equal weight is the same as none.
How much this is worth is not the same on every channel. The direction is reliable, because the friction is real, but the size of the gain depends on your audience and on what you are asking of them, and the only honest way to find your own number is to cut a few descriptions to one link and read what the clicks do. Treat it as a strong default to test, not a law to obey.
Which one destination?
The one that matches what the video promised. Not the one you most want them to visit.
If the video taught somebody how to do a thing, the destination is the longer version of it, or the tool they will need. If the video showed a finished piece of work, it is where the rest of that work lives. If the video answered a question, it is where the next question gets answered.
The test is uncomfortable to apply honestly: would somebody who has just watched this video, and nothing else on your channel, want this? Your newsletter usually fails it. Your booking page usually fails it. Both are good destinations for somebody who has watched six videos, which is not who this description is written for.
One link per video. A different one next time, when the next video promises something else.
Where does it actually go?
Above the point where the description collapses.
Descriptions are truncated on every device your viewers use, and the visible portion is short. Anything below the fold needs a tap to expand, which means it is only being read by people who were already going to act. Put the one destination in the visible part, early, on its own line, with plain words saying what it is.
Not the raw address, which tells nobody anything. Not the word here. The name of the thing.
What are the first two lines for?
Telling somebody what this video contains, in the words they would use.
The first lines of your description are the only ones most people ever see, and they are read by two audiences at once. A person deciding whether this video is the one they meant to open. And the platform, which uses your text to work out what the video is about and who else might want it.
Both want the same thing, which is a relief: a plain sentence saying what the video covers and who it is for. Write it as though a friend had asked what you made this week. If your first line is your channel tagline, or a greeting, or the phrase in this video, you have spent the most valuable text on the page on nothing.
Do timestamps belong in the description?
Yes, if the video is long enough that somebody would want to return to one part of it.
Timestamps do two jobs. They give the platform a set of chapters, so the progress bar becomes navigable rather than a single stripe. And they let somebody who watched last week find the bit about the awkward step without scrubbing for it.
Write them for the person, not for the algorithm. Sauce, not step three. Fixing the buzz, not troubleshooting. A chapter title is a small promise, and the same rule applies to it as to everything else you write: if it does not tell somebody what they will see, it is decoration.
What about keywords?
Use the words your viewers use, in sentences, and then stop.
The honest version of this is very short. Say what the video is about using the phrases somebody would actually type. If your video is about stopping a room sounding hollow, the word echo belongs in your description, because that is the word people use. Nothing clever is required.
What does not work, and has not for a long time, is the block of comma-separated phrases at the bottom. It reads as spam to a person, it is worth nothing to the platform, and it is sitting in the space where your one destination should be. Delete it and nothing bad will happen.
What can you take out this afternoon?
Open your five most recent videos and do the following to each, in order.
Delete the keyword block. Delete every link that is not the one destination for that video. Move the survivor above the fold. Replace the first line with a plain sentence saying what the video contains. Add timestamps if it is long enough to need them, written as things rather than as sections.
Roughly ten minutes per video, and the highest return per minute available in this part of the job.
The interesting part happens after that, and it is not a writing task. Once each video has one destination, you will start noticing which videos have no honest destination at all. That is worth sitting with. A video with nowhere to send anybody is usually a video that was made because it was Tuesday, and that is a planning problem, not a description one.
Keep reading
This is a service, and the method is written down.
Everything above came out of doing the work rather than writing about it. If you want the method instead of the story, it runs in order on one page.
Start with a free audit
Tell us where your content is now. We will come back with what we would change and what result to expect.
A person reads the channel and writes the audit by hand: a considered read typically takes three working days. That is the usual shape, not a promised turnaround. We use these details only to reply to you: no lists, no lurking.
What you will get
A fit snapshot: where your channel stands, and whether we are a match.
Two to three opportunities: specific, prioritised, yours to keep.
A recommended next step, even if that step is not us.
The audit is free and commits you to nothing: nobody follows up with a call you did not ask for.