A title has one job: convert an impression into a click without promising what the video can't deliver. The formulas below keep doing that job because they're built on stable mechanics — truncation, curiosity gaps, specific numbers — not on whichever phrasing is trending. Worked example: "7 Photoshop Tricks Nobody Tells You (Beginner Guide)" scores 90/100 under the transparent scoring formula in this guide, and every number in that score is reproducible.
YouTube's hard limit is 100 characters, but nobody sees your hundredth character. Search results truncate around 60, the mobile home feed can cut sooner, and suggested sidebar placements are tighter still. The working budget is 40-60 characters with the hook and keyword in the first half.
| Length | What happens | Verdict |
|---|---|---|
| 0-39 | Survives everywhere | Usually too vague — wasted space |
| 40-60 | Survives desktop search | The target zone |
| 61-70 | Cut in most search results | Acceptable if the front half carries it |
| 71-100 | Cut on home feed and mobile | Only for search-intent titles |
Front-loading is the part people skip. Whatever survives truncation must still sell the click on its own: "7 Camera Settings Nobody Tells Beginners…" can lose its tail and still work; "Here's Everything You Need to Know About…" can't.
Every pattern is fill-in-the-blank by design. You bring the topic; the structure does the mechanical work of being scannable, specific, and truncation-proof.
Enter a topic, get all ten patterns filled and scored 0-100 — then score your own drafts against the same formula.
YouTube Title Generator →The title generator scores deterministically, so the same title always gets the same number. Four weighted dimensions, two penalties:
"7 Photoshop Tricks Nobody Tells You (Beginner Guide)": 52 characters, so length = 100. One power word ("nobody"), so power = 40 + 20 = 60. Contains a digit, so number = 100. Has a bracket, so structure = 100. Weighted: 0.35×100 + 0.25×60 + 0.15×100 + 0.15×100 + 0.10×100 = 90/100, no penalties triggered.
Compare "How I Saved $1,200 on Car Insurance in One Afternoon": also 52 characters, same power word count, has a digit, but no bracket — 0.15×60 instead of 0.15×100 lands it at 84. And a 109-character title scores 37: the length penalty floors at zero (100 − 49×2.5 is deep underwater), and nothing else recovers it. Both are reproducible in the tool's score box.
The last point is worth internalizing: on YouTube, the click is only half the transaction. Titles that win the click and lose the viewer are financing this quarter's CTR with next quarter's reach.
Titles are editable forever, so treat publishing as the start of the test. The discipline: change one variable at a time, keep the target keyword stable so ranking signals don't reset, and give each version about a week (or a statistically meaningful impression count) before judging. Pair the title work with thumbnail testing — YouTube's Test & Compare runs thumbnail experiments natively, and the thumbnail tester scores contrast, saturation, and clutter before anything goes live. For the payoff side of the equation — what a CTR improvement actually earns — the YouTube earnings calculator converts views to revenue by niche.
The hard limit is 100 characters, but search results truncate around 60 and the mobile home feed can cut even earlier. Treat 40-60 characters as the working budget, with the keyword and hook in the first half so the meaning survives any cut.
They work as one unit — the thumbnail carries the emotional hook and the title closes it with specifics. A strong title fixes a weak thumbnail less often than the reverse, because viewers scan imagery first. Test both together; YouTube's built-in Test & Compare handles thumbnails, and title changes are just edits you measure over a week.
Curiosity-gap and loss-aversion words: nobody, mistake, stop, truth, why, how, versus, before, hidden. Specific numbers outperform them all when they're odd and concrete — 7 beats 10, $1,200 beats "a lot". Avoid stacking more than two power words; it reads as clickbait and hurts retention.
Yes, if early CTR is weak relative to your channel norm — impressions without clicks is the signal. Change one variable at a time (usually the title or the thumbnail, not both), keep the keyword intact so ranking signals don't reset, and give the change a week before judging.