The first 30 seconds decide most of a video’s fate. YouTube even measures it for you: in YouTube Studio, the Intro key moment shows what percentage of your audience still watched your video after the first 30 seconds (YouTube Help). Lose a big share of viewers there and the rest of your video barely matters, because hardly anyone is left to see it.
A hook is the thing that stops a viewer leaving in those first seconds. It isn’t a trick. The best hooks do one simple job: they confirm the viewer clicked on the right video, then give them a reason to stay. This guide gives you 50 fill-in-the-blank hooks in 10 types, explains the psychology behind each type, shows five weak openings rewritten, and shows how to check whether your hooks are actually doing their job.
What the first 30 seconds need to do
Every strong opening does three things, usually in this order:
- Confirm the click (0 to 5 seconds). The viewer chose your video because of a title and thumbnail. The first line should prove that choice was right. If the thumbnail promised a $3 game and the video opens with “Hey guys, welcome back,” you’ve already created doubt.
- Open a loop (5 to 20 seconds). Raise a question the viewer now wants answered: a surprising result, a mistake, a conflict, a number that doesn’t seem right.
- Promise the payoff (20 to 30 seconds). Tell them what they’ll have by the end: the answer, the method, the verdict. This gives them a reason to keep watching after the curiosity fades.
Then deliver fast: the first real piece of value should arrive within about 60 seconds. Notice what isn’t on the list: channel intros, greetings, “before we start, make sure to subscribe,” or a long backstory. Each of those delays the moment the viewer finds out whether the video is worth their time. Save them for later, or cut them.
What “good” looks like in numbers
There’s no universal target, because retention depends on video length and type. Humble and Brag’s 2026 benchmarks give a useful starting point (Humble and Brag, March 2026):
| Video type | Healthy average percentage viewed |
|---|---|
| Under 5 minutes | 50–70% |
| 5 to 15 minutes | 40–55% |
| 15 to 30 minutes | 30–45% |
| Over 30 minutes | 25–35% |
| Tutorials and how-tos | 45–55% |
| Shorts | 70% or more is healthy; the first 3 seconds are critical |
The more useful comparison is with yourself. Look at the Intro percentage across your last ten videos. The ones with the highest numbers tell you which openings work for your audience. That’s worth more than any benchmark.
The psychology behind a good hook, briefly
You don’t need a psychology degree to write hooks, but knowing four well-studied ideas helps you understand why the templates below work, and write your own when they don’t fit:
| Idea | What it says | How hooks use it |
|---|---|---|
| The information gap | Psychologist George Loewenstein described curiosity as the feeling of noticing a gap between what we know and what we want to know. A small, specific gap is more motivating than a huge, vague one. | Direct questions, experiments and story cold opens all open a specific gap that the video closes |
| Loss aversion | Daniel Kahneman and Amos Tversky’s research found that losses tend to feel larger than equal gains. | Warnings and stakes hooks frame the topic as something to avoid losing |
| Concreteness | Specific, concrete details are easier to picture, remember and believe than abstract ones. | Specific numbers, named tools and exact timeframes beat “a lot” and “soon” |
| Self-relevance | People notice information that’s about them, much like hearing your own name across a noisy room. | Identity call-outs make the right viewer feel the video was made for them |
One caution: every one of these can be abused. A curiosity gap the video never closes, or a loss that isn’t real, works once. Viewers remember, and they stop clicking your videos. The hooks that build channels use these ideas honestly: the gap gets closed, and the stakes are real.
The 50 hooks
Each hook below is a template. Replace the brackets with your topic. They’re grouped into 10 types, because the type matters more than the exact wording: once you know why a type works, you can write your own.
Type 1: The result first (hooks 1–5)
Why it works: it shows the payoff up front, so the viewer knows the video delivers. The loop becomes “how?” Best for tutorials, experiments and transformations.
- “This [result] took me [short time] and cost [small amount]. Here’s exactly how.”
- “By the end of this video, you’ll have [specific thing] that [benefit].”
- “This is what [thing] looks like after [time period]. Now let me show you day one.”
- “[Impressive number]. That’s what [method] did for [thing]. Here’s the breakdown.”
- “I’ll show you the finished [thing] first, so you know it works. Then we’ll build it.”
Type 2: The mistake or warning (hooks 6–10)
Why it works: people are more motivated to avoid a loss than to chase a gain. A specific warning feels urgent and useful. Best for finance, health, legal and “how to” topics.
- “If you’re [doing common thing], you’re probably [losing money / wasting time]. Here’s why.”
- “Most people get [topic] wrong in the first step. It’s not the step you think.”
- “Don’t [common action] until you’ve checked these [number] things.”
- “I made this mistake with [topic] and it cost me [amount or time]. You don’t have to.”
- “The [popular advice] everyone repeats is outdated. Here’s what changed in [year].”
Type 3: The contrarian take (hooks 11–15)
Why it works: disagreement creates tension, and tension holds attention. It only works if you back it up, so use it when you have evidence.
- “Everyone says [common belief]. The data says the opposite.”
- “[Popular tool or method] is overrated. Here’s what I use instead.”
- “You don’t need [thing everyone says you need] to [goal].”
- “Unpopular opinion: [claim]. Let me prove it.”
- “I tested the advice every [niche] creator gives. Most of it didn’t hold up.”
Type 4: The specific number (hooks 16–20)
Why it works: precise numbers feel credible and concrete. “$217” is more believable than “a few hundred dollars.”
- “[Exact number] [things] in [time]. Here’s what I learned from every one.”
- “This one setting saves [exact amount] a [month/year].”
- “[Number] out of [number] people [surprising fact]. Here’s why that matters to you.”
- “I spent [exact amount] testing [thing] so you don’t have to.”
- “There are only [small number] [things] that really matter for [goal]. Here they are.”
Type 5: The experiment (hooks 21–25)
Why it works: an experiment has a built-in ending: the result. Viewers stay to find out. This is one of the strongest types for faceless channels, because the experiment is the star, not the presenter.
- “I tried [thing] for [time period]. Here’s what actually happened.”
- “I gave [tool A] and [tool B] the same task. One of them failed badly.”
- “Can you really [bold claim]? I tested it with [constraint].”
- “I followed [famous person or method]’s exact routine for [time]. The results surprised me.”
- “I asked [number] [experts/tools] the same question. Their answers didn’t match.”
Type 6: The story cold open (hooks 26–30)
Why it works: starting mid-action drops the viewer into a situation they want resolved. Best for documentaries, case breakdowns and history.
- “At [time] on [date], [person] [did something unexpected]. Nobody knew why.”
- “The email arrived at [time]. By the next morning, [consequence].”
- “This [object/place] looks ordinary. It’s the reason [surprising outcome].”
- “[Person] had [everything going well]. Then [one decision] changed it all.”
- “It started with a [small, ordinary thing]. It ended with [big consequence].”
Type 7: The direct question (hooks 31–35)
Why it works: a question the viewer can’t answer right away makes them want the answer. Keep it specific. “Have you ever wondered about money?” is too vague to work.
- “Why does [surprising fact] happen, even when [expected condition]?”
- “What would happen if you [did thing] every day for [time]?”
- “Is [popular thing] actually worth [price]? Let’s do the math.”
- “How did [small player] beat [giant] at [thing]?”
- “What’s the one [thing] every [group] should know before [event]?”
Type 8: The identity call-out (hooks 36–40)
Why it works: naming the viewer’s situation makes them feel the video is for them, and filters out people who would leave anyway.
- “If you’re a [specific group] trying to [goal], this is for you.”
- “This is for anyone who’s [frustrating experience] more than once.”
- “If you’ve got [limited resource], here’s how to [goal] anyway.”
- “New to [topic]? Start here, not with [common starting point].”
- “If [situation] sounds familiar, you’re not doing anything wrong. [Real cause] is.”
Type 9: The news or change (hooks 41–45)
Why it works: timeliness. Viewers want to know what changed and how it affects them. Best for tech, AI, finance and platform updates. Always name the date.
- “[Company] just changed [thing]. Here’s what it means for you.”
- “From [date], [rule] is different. Here’s who’s affected.”
- “[New tool] launched this week. I tested it for [time]. Here’s my honest take.”
- “Everyone’s talking about [news]. Here’s the part most people missed.”
- “[Thing] just got [cheaper/harder/banned]. What to do before [deadline].”
Type 10: The stakes and time limit (hooks 46–50)
Why it works: a clear cost or deadline makes the information feel necessary rather than nice to have. Use real stakes only. Fake urgency gets noticed quickly.
- “You have until [date] to [action]. After that, [consequence].”
- “This costs the average [person] [amount] a year, and most don’t notice.”
- “In [time], [thing] will [change]. Here’s how to get ready.”
- “Get this wrong and you [concrete loss]. Get it right and you [concrete gain].”
- “[Number] minutes today saves you [bigger number] hours later. Here’s how.”
Which hook fits which video
Using the wrong type is the most common reason a good hook fails. A story cold open on a software tutorial feels like a bait-and-switch. A result-first hook on a mystery documentary gives away the ending. Start from what the viewer comes for:
| The viewer comes for… | Best hook types |
|---|---|
| Learning how to do something | Result first (1–5), mistake or warning (6–10) |
| Understanding something new | News or change (41–45), direct question (31–35) |
| A story or mystery | Story cold open (26–30), stakes (46–50) |
| A verdict: is X worth it? | Experiment (21–25), contrarian (11–15) |
Two types work almost everywhere: specific numbers (16–20) and identity call-outs (36–40). They pair well with any other type. “If you’re a freelancer (identity), this one setting saves $217 a year (number)” is two hooks in one sentence.
Hooks by niche
Niches have habits. Here’s where each of the common faceless niches usually gets its best openings, with an example you could adapt:
| Niche | Strongest types | Example |
|---|---|---|
| Personal finance | Warning, specific number, identity | “If you’re paid monthly, this one calendar trick stops the last-week squeeze.” |
| Software tutorials | Result first, identity | “Here’s the finished dashboard. It updates itself. Let’s build it in ten minutes.” |
| AI tools | Experiment, news, contrarian | “I gave three AI assistants the same messy spreadsheet. Only one got the totals right.” |
| True crime and case breakdowns | Story cold open, direct question | “The receipt was time-stamped 9:42 p.m. That’s the problem.” |
| Health and wellness | Contrarian (with evidence), warning | “The advice to drink eight glasses a day has a surprisingly thin origin story.” |
| History and explainers | Direct question, story cold open | “Why does this tiny island have its own time zone?” |
| Career and education | Identity, stakes | “If your exam is in 30 days, here’s the order I’d study in, and what I’d skip.” |
The rule that matters more than any template
The hook must match the title and thumbnail. YouTube’s own advice when the Intro metric is low is to first check whether your title and thumbnail better reflect your video content, and only then to rework the first 30 seconds. A viewer who feels tricked leaves, and a lot of early exits tell YouTube the video didn’t satisfy people.
A simple test: read your title, look at your thumbnail, then read your first sentence. If the first sentence doesn’t obviously continue the same promise, rewrite it. Clickbait can raise clicks for a while, but it lowers retention, and retention is what keeps recommendations coming.
Three matched sets: title, thumbnail and first line
| Title | Thumbnail | First line |
|---|---|---|
| I Tried 5 Free Budget Apps for 30 Days | Five app icons, one circled in green, one crossed out | “Four of these apps were fine. One of them quietly shared my data, and one saved me $180.” |
| The Spreadsheet Formula That Replaced My Bookkeeper | A single formula on screen with “$0” beside it | “This one formula does what I used to pay someone $60 a month to do. Here it is working.” |
| Why Your Home Insurance Doesn’t Cover Floods | A flooded living room with a policy document and a red “NOT COVERED” stamp | “Most home policies exclude flooding, and most people find out after the water’s gone.” |
In each set, the three parts tell one story without repeating each other. The title makes a promise, the thumbnail shows the tension, and the first line confirms both and opens the loop.
Five weak openings, rewritten
Templates are easier to use once you’ve seen them fix a real problem. Here are five typical openings and how the hook types above repair them:
1. The slow warm-up
Before: “Hey everyone, welcome back to the channel. Today we’re going to be talking about budgeting apps, which is something a lot of you have asked about.”
After (result first + number): “This app found $38 a month I didn’t know I was spending. It took eleven minutes to set up. Here’s exactly how.”
What changed: the viewer hears the payoff in the first sentence, with a number they can picture. The “welcome back” added nothing a new viewer needed.
2. The vague tease
Before: “You won’t believe what this AI tool can do.”
After (experiment): “I gave this AI tool a 40-page lease and asked for the three clauses a tenant should worry about. It found two. The one it missed could cost you a deposit.”
What changed: “you won’t believe” is a claim. The rewrite is a test with a result, and the missed clause opens a loop that only the video closes.
3. The answer given away
Before: “Is a premium subscription worth it? No, and here’s why.”
After (direct question + stakes): “Is the premium plan worth $12 a month? For most people, it depends on one number in your usage stats, and I’ll show you where to find it.”
What changed: the question stays open, and the viewer now wants to find their own number, which keeps them through the explanation.
4. The backstory
Before: “So, back in 2019, I was working a job I didn’t love, and I started looking into side projects…”
After (story cold open): “The invoice was for $1,400, and I had four days to deliver something I’d never built. This is how I did it, and what I’d do differently.”
What changed: the story starts at its most tense moment. The backstory can come later, once the viewer cares.
5. The wrong audience
Before: “Today I want to talk about taxes.”
After (identity + warning): “If you earned any freelance income this year, even $500, check whether you need to file anything extra this year. Most first-timers don’t know to ask.”
What changed: the right viewer recognises themselves instantly, and everyone else leaves early, which is fine. A small audience that stays is worth more than a big one that clicks away.
Ten finished hooks, filled in
Templates can feel abstract, so here are ten of them completed for different niches. Borrow the structure, not the words:
- Result first (#1): “This meal plan took me 40 minutes and cut our grocery bill by $61 a week. Here’s exactly how.”
- Warning (#8): “Don’t sign a gym contract until you’ve checked these three clauses.”
- Contrarian (#13): “You don’t need a second monitor to work faster. You need these four keyboard shortcuts.”
- Number (#17): “This one phone setting saves about 40 minutes of screen time a day.”
- Experiment (#22): “I gave two AI tools the same 200-row sales sheet. One invented a customer who doesn’t exist.”
- Cold open (#27): “The email arrived at 11:52 p.m. By the next morning, the whole team had quit.”
- Question (#33): “Is a $300 air purifier actually worth it? Let’s do the math on one room.”
- Identity (#38): “If you’ve only got 20 minutes a day, here’s how to learn Spanish anyway.”
- News (#42): “From February 1, 2027, getting monetized on YouTube gets twice as hard. Here’s who’s affected.”
- Stakes (#49): “Get this form wrong and your refund waits months. Get it right and it arrives in weeks.”
A complete 30-second opening, annotated
Here’s a full opening for a video titled “I Tested 3 Ways to Save for Christmas. One Worked Twice as Well.” Each line is labelled with the job it does:
| Time | Line | Job |
|---|---|---|
| 0–4 s | “Three saving methods, three months, the same $50 a week. One of them ended with twice as much.” | Confirms the click: this is the test the title promised |
| 4–12 s | “And it wasn’t the one I expected. The method most budgeting videos recommend came last.” | Opens the loop, with a contrarian edge |
| 12–22 s | “The difference came down to one habit, and it takes about two minutes a week.” | Raises a second, smaller question |
| 22–30 s | “By the end you’ll know which method fits how you get paid, and you’ll have the tracker I used, free, in the description.” | Promises the payoff and a tangible takeaway |
| 30–60 s | Straight into method one, with the first result on screen within the minute | Delivers early value, so the promise feels real |
Read it aloud and it takes about 25 seconds at a natural pace. There’s no greeting, no channel name and no request to subscribe, and nothing is missing.
The re-hook: keeping viewers after the first 30 seconds
A hook buys you the first 30 seconds. Re-hooks buy you the next ten minutes. A re-hook is a small loop you open partway through a video, just before a point where attention usually dips. Most retention graphs show slow leaks at section changes, and a re-hook plugs them.
Three ways to re-hook:
- Preview a later payoff. “The third method is the one that doubled the result, but it only works if you understand this first one.” Viewers now have a reason to stay through method one.
- End each section with a question. “That fixes the first problem. But it creates a new one, and most people never notice.” The next section opens by answering it.
- Use a mid-video surprise. Put your most unexpected finding in the middle, not at the end, and flag it just before: “This is the part that surprised me.”
How hooks change with video length
| Length | Hook job | Re-hooks |
|---|---|---|
| Shorts (under 60 seconds) | Visual plus one line of text, in the first second | None needed. The payoff should arrive before the end |
| 5–10 minutes | Promise in 5 seconds, full opening in 20–30 | One or two, at section changes |
| 15–30 minutes | A stronger loop, often a story cold open, plus a roadmap | Every 3–5 minutes, usually at chapter breaks |
| Documentaries (30+ minutes) | A cold open at the most dramatic moment, then back to the start | At every act change, often with a preview of what’s coming |
Score your hook before you record
Paste or type your opening line below and tick what it does. It’s a quick checklist, not a guarantee, but an opening that ticks five or more boxes avoids the most common reasons people leave early.
Hook checklist
Nothing you type leaves this page.
Hooks when there’s no face on screen
On-camera creators get a head start: a face showing surprise or urgency does part of the hook’s work. Faceless channels have to do that work with voice, pictures and words. Four adjustments help:
- Let the voice carry the energy. Record the first 30 seconds last, when you’re warmed up, and read them slightly faster and brighter than the rest. If you use an AI voice, generate several takes of the opening alone and pick the one with the most natural emphasis on the key word.
- Make the first image do the reacting. Instead of a face looking shocked, show the thing that’s shocking: the number circled in red, the broken result, the before-and-after side by side.
- Shorten the opening sentences. Without a face to watch, listeners lose track of long sentences faster. Keep hook sentences under 15 words.
- Use sound deliberately. A short, clean sound cue on the key reveal (a click, a soft whoosh) marks the moment the loop opens. One is enough. A soundtrack of effects gets tiring fast.
The upside for faceless channels is that experiments, results and stories, three of the strongest hook types, don’t need a presenter at all. The evidence on screen is the hook.
Visual hooks for long videos
Long-form viewers chose your video from a title, but they still judge the first seconds with their eyes as well as their ears. A strong spoken hook over a static title card wastes half its power. Four visual habits help:
- Show the promise on screen in the first frame. The finished dashboard, the before-and-after, the receipt with the surprising number. The viewer should see why they clicked before they hear it.
- Change the picture within three seconds. A zoom, a cut to a close-up, a highlight on the key number. Movement early signals that the video is going somewhere.
- Put the key number or question on screen as text. It reinforces the hook for people watching on mute, and it’s easier to remember.
- Save your channel branding for later. A logo sting or animated title in the first five seconds is the visual version of “hey guys, welcome back.”
For faceless channels, this is where AI video clips earn their cost: one striking 5-second shot at the very start can do more for retention than ten clips in the middle. We compare the tools for that job in Kling vs Seedance vs Veo.
Hooks for Shorts are different
On Shorts, the viewer didn’t choose your video from a title. It just appeared in their feed, and they decide within a second or two whether to stay. That changes the rules:
- Lead with the visual. The first frame should already show the most interesting thing: the result, the before-and-after, the surprising object. No title card.
- Put the hook on screen as text. Many people scroll with the sound off. A short line of text in the first frame does the job a spoken hook can’t.
- Cut the setup completely. “So today I want to show you” costs a full second you don’t have. Start with the second sentence of what you’d normally say.
- Check “viewed vs swiped away.” YouTube Studio shows how often people who saw your Short in the feed stayed rather than swiping. YouTube doesn’t publish a target, so compare each Short with your own recent average (PrePublish).
Hooks 1, 3, 17, 22 and 46 translate best to Shorts, because they work as a single visual plus one line of text.
Seven hook mistakes that cost viewers
- The slow warm-up. “Hey everyone, welcome back to the channel, today we’re going to be talking about…” Every word before the promise is a reason to leave.
- The vague tease. “You won’t believe what happened.” Viewers have heard it a thousand times. Specific beats mysterious.
- Answering the question immediately. If the hook asks “is X worth it?” and the next sentence says “no,” you’ve closed the loop. Give the answer at the point where you’ve earned it.
- Over-promising. A hook that promises more than the video delivers drives early exits and comments that warn other viewers.
- Stacking three hooks. One clear question beats three competing ones. Pick the strongest.
- The same hook every time. Your returning viewers notice. Rotate between two or three types that suit your niche.
- Writing the hook last. Write the hook first, before the script. If you can’t write a strong one, the video idea itself may be weak, and it’s better to know that before you spend a day making it.
How to test and improve your hooks
Hooks improve fastest when you treat them as experiments. A simple routine:
- Write three versions of every hook, each using a different type from the list.
- Read them aloud. The one that sounds most like something you’d say to a friend is usually the one to use.
- After publishing, wait 48 hours, then open Analytics, Engagement, Key moments for audience retention, and note the Intro percentage.
- Keep a simple log: video, hook type, Intro percentage, average percentage viewed. After ten videos you’ll see which types your audience responds to.
- Use YouTube’s Test & Compare in Studio to test thumbnails (and titles, where available on your channel). The hook has to match whichever version wins, so test those first.
If your Intro percentage is low but viewers who stay past 30 seconds watch most of the video, the problem is the opening, not the content. If the Intro is fine but the curve drops steadily after that, the hook is working and the problem is pacing further in.
Reading your hook log after ten videos
| What the log shows | What it means | What to do |
|---|---|---|
| One hook type consistently beats the others on Intro percentage | Your audience likes that kind of promise | Make it your default, and test one other type in every third video |
| High Intro percentage but low average percentage viewed | The hook works, but the video doesn’t keep its promise fast enough | Deliver the first real value sooner, and cut setup after the hook |
| Low Intro percentage on every type | Titles and thumbnails may be attracting the wrong viewers | Make titles more specific, even if clicks drop a little |
| Big differences between search and browse traffic | Search viewers want confirmation; browse viewers want curiosity | For search-heavy videos, confirm fast. For browse-heavy ones, open a stronger loop |
Using AI to write hooks (without sounding like AI)
AI is useful for hooks in one specific way: generating many options fast. It’s poor at choosing. A workflow that uses its strength:
- Give it your title, thumbnail description, and the one-sentence payoff of your video.
- Ask for ten openings, two from each of five hook types you name, each under 20 words.
- Ask it to mark which ones overpromise compared with the payoff you gave it.
- Pick the best two, then rewrite them in your own words, out loud.
The last step matters. AI hooks tend to share tells: “Here’s the thing,” “Let’s dive in,” “game-changer,” and a rhythm of three adjectives. Viewers have heard them thousands of times. If a line sounds like it could open anyone’s video, it isn’t a hook yet.
Frequently asked questions
How long should a hook be?
The core promise should land in the first 5 seconds, which is roughly 10 to 15 spoken words. The full opening, including the loop and the payoff promise, usually fits in 20 to 30 seconds.
Can AI write my hooks?
AI is good at generating many variations quickly, which is useful for step 1 of the testing routine. It’s weaker at knowing which one is true to your video and sounds like you. Use it for options, then choose and edit yourself.
Should I put my channel intro after the hook?
If you have one at all, keep it under 3 seconds and place it after the payoff promise. Many successful channels have dropped branded intros completely.
Do hooks matter for search traffic?
Yes, although a bit less. Search viewers already know what they want, so the hook’s main job is confirming they found it: “Here’s exactly how to [what they searched], step by step.” Hooks 1, 2 and 36 work well here.
My hook is good, but people still leave at 30 seconds. Why?
Usually because the first real piece of value takes too long to arrive. After the hook, deliver something useful within the first minute, even if it’s small, to prove the video is worth the rest of their time.
Is using hooks manipulative?
Not if the video delivers. A hook is a clear, honest summary of why the video is worth watching, said quickly. It becomes manipulative when it promises something the video doesn’t contain, and that tends to punish itself: viewers leave, comment and stop trusting your titles.
Do I need a different hook for each language or country?
If you publish in several languages, yes. Numbers, references and even what counts as “surprising” differ between audiences. Translate the idea, not the sentence, and test it the same way.
Keep reading
Last checked September 25, 2026. YouTube Studio features and labels change over time. Sources: YouTube Help, “Measure key moments for audience retention”; Humble and Brag (March 2026); PrePublish. Psychology references: George Loewenstein’s information-gap theory of curiosity (1994); Daniel Kahneman and Amos Tversky’s work on loss aversion.
Was this article useful?
Your feedback helps us improve these guides.
