Titles
How to Write YouTube Titles That Get Clicks
How to write YouTube titles that earn clicks: the promise test, length limits that actually matter, search versus browse framing, and formulas worth reusing.
September 1, 2026 · 7 min read · By Christian Koch
A YouTube title earns clicks by making one specific promise the thumbnail supports. Titles that describe a category give a viewer nothing to want, and titles that overstate the video's contents earn a click and lose the viewer inside thirty seconds. The workable range sits between those two, and it is narrower than most channels treat it.
Key takeaways
- A title that describes the subject is a label. A title that states what the viewer gets is a promise.
- Keep the meaning inside the first 50 characters or so, since mobile and sidebar placements truncate.
- Search titles and browse titles are different jobs, and one title cannot fully serve both.
- Specific claims outperform sweeping ones, which makes accuracy the easier route to clicks.
The promise test
Read your title and answer one question: what does the viewer get? A title with no answer is a label.
| Label | Promise |
|---|---|
| "Q3 Marketing Update" | "The Q3 Campaign That Cost Us 40% of Our Leads" |
| "Thoughts on the New iPhone" | "Three Things the New iPhone Broke That Nobody Mentioned" |
| "How to Use Our Software" | "Set Up the Whole System in Eleven Minutes" |
| "Interview with Sarah Chen" | "Sarah Chen Turned Down a $4M Acquisition. Here's Why." |
Every title in the right column contains something concrete: a number, a consequence, a named person with a decision attached. The left column describes accurately and gives nobody a reason to stop scrolling.
The test takes ten seconds per title, and applying it to your last twenty videos usually finds a pattern.
What actually makes a title work
Three properties, and they compound.
Specificity. Numbers, names, timeframes, concrete outcomes. "A Lot of Money" is weaker than "$40,000," and it is weaker precisely because the vague version sounds like it might be an exaggeration. Specific claims read as verified.
Stakes. What the viewer gains or avoids by watching. This is the source of curiosity, and it works without withholding information. "The Mistake That Cost Us $40,000" tells you a great deal and still leaves you wanting the answer, which is the shape to aim for.
Legibility to a stranger. Channel shorthand, in-jokes and insider terminology narrow the title to people who already watch you. That ceiling is your current audience, which makes it the most expensive habit on this list.
Length: what matters and what doesn't
YouTube allows 100 characters. The meaningful limit is closer to 50.
Titles truncate differently depending on where they appear: mobile home feed, desktop search results, the suggested sidebar, and the watch page each cut at a different point, and the sidebar is the most aggressive. A title whose meaning depends on words 60 through 90 will be unreadable in several of the places it appears most.
The practical rule: put the meaning in the first 50 characters and treat anything after that as a bonus for the placements that show it. This is a constraint on where the important words sit, and it leaves total length open.
Two related habits worth dropping. Front-loading brand names spends the most valuable characters on something the viewer already sees in the channel name below. Stuffing keywords past the meaning produces a title that reads as machine-written, and readers respond to that.
Titles are the fastest thing to fix on most channels and the easiest to get subtly wrong. the YouTube Growth Sprint includes a full title and thumbnail rebuild across your existing library, which frequently produces the first measurable lift of an engagement.
Search titles and browse titles are different jobs
A title optimized for search leads with the query. A title optimized for browse leads with the hook. One title cannot fully do both, and pretending otherwise produces a title that does neither well.
| Search title | Browse title | |
|---|---|---|
| Viewer state | Typed a query, scanning for a match | Scrolling, no intent |
| Opening words | The query itself | The most interesting element |
| Success condition | Recognized as the answer | Interrupts the scroll |
| Example | "How to Set Up a YouTube Channel for a Business" | "We Rebuilt This Business Channel in 30 Days" |
Decide which job a video is doing before writing the title. Evergreen how-to content with real search demand should lead with the query, since a viewer scanning results for a match wants to see their own words. Content with no meaningful search volume should lead with whatever is most interesting, since nobody is looking for it by name.
The judgment call belongs at the idea stage, and we cover the search side in YouTube SEO.
Title formulas worth reusing
Formulas are scaffolding for the promise. They work when the specifics underneath them are strong, and they fail when they become a substitute for having something to say.
- The specific consequence. "[Action] Cost Us [Number]" - grounds a claim in a real outcome.
- The count with stakes. "Three Things [X] Broke That Nobody Mentioned" - a count sets scope, the qualifier adds tension.
- The named decision. "[Person] Turned Down [Specific Thing]. Here's Why." - works when the person or the thing is recognizable to your audience.
- The corrected assumption. "[Common Belief] Has Been Wrong Since [Year]" - sets up a reversal without overstating it.
- The constrained how-to. "Do [Task] in [Timeframe]" - the constraint is the promise.
- The direct query. "How to [Task]" - for genuine search demand, where clarity beats cleverness.
- The comparison. "[A] vs [B]: [The Deciding Factor]" - the second half is what makes this work, since the bare comparison promises nothing.
The failure mode is running a formula on a video with nothing specific in it. "Five Things About Marketing" is formula 2 with the substance removed, and it performs like a label.
Titles and thumbnails as one unit
The title carries the stakes. The thumbnail carries the situation. Repeating one inside the other wastes half the packaging.
Four words of thumbnail text duplicating four words from the title spends your most valuable visual space on redundancy. When they divide the work, a thumbnail showing a specific moment and a title stating why that moment matters combine into something neither delivers alone. Full frame in packaging a YouTube video, and the thumbnail side in the thumbnail half of the pair.
Changing titles after publishing
Changing a title carries no penalty and resets nothing. Videos with strong retention and weak click-through rate are the best candidates, since they already prove the content satisfies the people who arrive.
YouTube's native A/B test compares up to three titles on a live video and decides the winner by watch time, since the platform states it optimizes tests "for overall watch time over other metrics, like click-through-rate" (YouTube Help). Testing the title alone with the thumbnail held constant is the cleanest way to learn something about title framing you can apply to the next twenty videos. Mechanics in testing titles and thumbnails.
One caution: give a manual title change two to three weeks before judging it. Click-through rate on a few hundred impressions swings enough to produce false readings in either direction, and impressions only register when a thumbnail is shown for over a second with at least half visible (YouTube Help).
FAQ
How long should a YouTube title be?
YouTube permits 100 characters, and the meaning should sit inside the first 50. Mobile feeds, search results and the suggested sidebar each truncate at different points, with the sidebar cutting most aggressively. A title whose sense depends on later words becomes unreadable in several of the places it appears most often.
Should I put keywords in my YouTube title?
Yes, when the video serves real search demand, and the keyword should lead the title so a viewer scanning results sees their own words first. For videos with no meaningful search volume, keywords add nothing and cost you the opening characters, which are worth more spent on the hook.
Do YouTube titles affect the algorithm?
Titles influence which queries a video surfaces for and how many people click, and both feed distribution. YouTube's own A/B testing tool resolves title tests on watch time, which indicates that the downstream effect on viewer satisfaction matters more than the click itself.
Can I change a YouTube title after publishing?
Yes, with no penalty. The best candidates are videos with strong audience retention and weak click-through rate, since those already demonstrate the content works for people who arrive. Change one thing at a time and allow two to three weeks before drawing a conclusion.
What makes a YouTube title clickbait?
Promising something the video does not deliver. Specificity and accuracy are the defence, and they are also the higher-performing route, since a title that overstates earns a click and loses the viewer within thirty seconds, which reduces the video's distribution.
Where to start
Run the promise test on your last twenty titles: for each one, write down what a viewer gets. The titles where you cannot answer are your fastest available improvement.
If you want that rebuilt across the whole library with a testing plan attached, the YouTube Growth Sprint covers it.