How to write titles that get clicks
A title is read in about half a second, alongside a thumbnail, by someone who is not looking for you. It has one job: make the video look worth the next thirty seconds.
Most weak titles fail the same way. They describe the video accurately and give no reason to watch it, which is a summary rather than a title.
You end with a title you can defend against the alternatives you rejected.
Steps
Lead with the interesting part
Whatever is most surprising, most useful or most specific goes first, because the end of a title is frequently truncated and always read with less attention. "Why most faceless channels fail in month three" works; "A discussion of common problems faced by faceless channels" says the same thing and buries it.
Be specific where you can be
Numbers, timeframes and concrete nouns outperform general claims, because they signal that the video contains something definite. "Three things" beats "some tips". A specific title also sets an expectation the video can actually meet, which protects retention in a way vague titles do not.
Write five, then pick
The first title is almost never the best one, and the gap between the first and the fifth is usually larger than any other edit you could make in the same time. Writing alternatives is fast, and it is the highest-return five minutes in the whole upload process.
Check it against the thumbnail
Title and thumbnail are read together and should not repeat each other. If the thumbnail already says the words, the title has been wasted on duplication. Use it to add the second half of the idea instead, so the two together say more than either alone.
Generate options when you are stuck
Feed the topic into the free title generator to get a list to react against. Reacting to candidates is consistently faster than producing them from nothing, and the ones you reject sharpen what you actually want the title to do.
What you end up with
A shortlist of five titles for one video, with the strongest leading on the specific detail rather than the category, sitting under sixty characters so it survives truncation on mobile.
Try the title generatorWhen this isn't the right approach
Titles matter less on Shorts than most advice implies. The feed surfaces videos by behaviour rather than by click decision, and the first frame is doing the work the title does on long-form. If you are producing Shorts at volume, a serviceable title on every upload beats an agonised one on a third of them. Save the five-option treatment for long-form, where a click is an actual decision someone makes.
Ready to make one?
Try the title generatorFrequently asked
How long should a YouTube title be?
Under about sixty characters if you want it whole on mobile, where most of your audience is. Longer titles are not penalised, but the tail gets cut, so anything past that point should be information you can afford to lose. Front-loading the interesting part solves this automatically rather than requiring you to count.
Do keywords in the title help ranking?
Somewhat, and much less than click and watch behaviour. A title stuffed with search terms that nobody clicks performs worse than a compelling one that omits them, because the behaviour signal dominates. Include the term if it fits naturally; do not bend the title around it.
Should I use capital letters or clickbait punctuation?
Sparingly. Full capitals and stacked exclamation marks read as low quality to a lot of viewers, and the attention they buy is usually from people who leave immediately. A specific, confident title outperforms a shouted vague one, and it does not cost you the audience that finds shouting off-putting.
Is it worth changing the title of an old video?
Yes, on videos that get impressions but few clicks — that is exactly the signal a title can fix. Retitling something with almost no impressions changes nothing, because the problem is upstream. Change one thing and watch the click rate rather than changing title and thumbnail together.