Writing a YouTube description that does something

Most descriptions are one of two things: a wall of hashtags, or the title copied out with some links underneath. Both waste the field, and the field is worth more than the tag box that people spend far longer on.

The description does three separate jobs, and they happen in different places: the two lines shown before “more”, the chapters, and everything below. Treating it as one block of text is why it usually does none of them.

The two lines that get read

Above the fold you get roughly two lines of text. On a phone it is less. That is the part almost everybody sees and the only part most people ever read.

Those lines are read by two audiences at once. A viewer deciding whether this video is the one they wanted, and a system trying to work out what the video is about. Both are served by the same thing: a plain sentence saying what the video actually covers.

What goes wrong there is predictable.

The title, again. The title is directly above. Repeating it spends your most visible text on something already read.

A link, first. Anything clickable in the first two lines is competing with your own video for attention, at the exact moment somebody is deciding to watch.

A greeting. “Welcome back to the channel” tells nobody anything.

The version that works: say what the title promised, and add the specific thing the title did not have room for. If the title is a claim, the opening lines are the scope — what is covered, for whom, and anything that would otherwise be a disappointment discovered at minute four.

That last point matters more than it sounds. A description that quietly narrows the promise costs you a few clicks and saves you the far more expensive kind of viewer, the one who clicks, discovers the video is not what they wanted, and leaves in thirty seconds. The first fifteen seconds decide the video covers what happens after the click; the description is where you can stop the wrong click happening at all.

If you are unsure how much text you actually have before the cut, the word counter will give you a character count, and the same discipline that applies to titles applies here — the title analyzer shows where a title is cut at four different widths, which is the same problem one field higher.

Chapters, next

Timestamps become chapters. Chapters appear on the progress bar, and they become their own entries in search results, which means one video can surface for several different questions.

They also change how people watch. A viewer who can see that the part they want is at 4:12 is a viewer who stays instead of leaving to find a shorter video.

The format is simple and unforgiving: each line starts with a timestamp, the first must be 0:00, they must be in order, and there need to be at least three. Get one wrong and YouTube silently declines to make chapters at all, with no error anywhere. The timestamp generator produces the block in the exact format YouTube parses, which removes the most common reason chapters do not appear.

Write the labels as things somebody would search for, not as structural markers. “How the payout is calculated” is a chapter. “Part 2” is not.

Everything below the fold

Below the chapters, in descending order of usefulness:

• Links the viewer actually wants — the tool you demonstrated, the source you cited, the previous video in the series.
• Credits and licences. Where footage, music and images came from. This is dull and it is the record you will want if a claim ever arrives. Finding b-roll that doesn’t look like stock covers keeping that record while you download rather than reconstructing it a year later.
• A short channel line. One sentence on what the channel is, for the person who arrived from search and has never seen you before.
• Up to three hashtags, if they are accurate. The first three appear above the title.

What does not belong: a keyword list. Stuffing the description with search terms is a tactic from the same era as tag stuffing, and it has the same problem — you are adding text that describes your video less accurately. Do YouTube tags still matter? goes through where that effort is better spent. If you want the genuine phrasing variants your audience uses, the tags and keyword generator works from your actual topic, and the useful ones belong in the opening lines as normal prose, not in a pile at the bottom.

A description template that is not a template

[One sentence: what this video covers, specifically.]
[One sentence: the scope, the caveat, or who it is for.]

0:00 [Chapter]
1:24 [Chapter]
...

Links
- [thing shown in the video]

Credits
- footage: [source, licence]
- music: [source, licence]

[One line about the channel.]

That is five minutes of work per video, and four of those minutes are the chapters.

The short version

• Two lines get read. Do not spend them on the title or a greeting.
• Say what the video covers and narrow the promise where it needs narrowing.
• Chapters are worth more than the whole tag field, and they break silently if the format is off.
• Links, credits and licences go below the chapters.
• Three accurate hashtags, at most. No keyword pile.