How to Write Ad Hooks That Stop the Scroll in 2026
The first 3 seconds determine whether your ad gets watched or skipped. Learn the four proven hook types, a step-by-step writing process, and real examples across six industries.
Mauricio Valdivia
·9 min read
Every day, the average person scrolls through 300 feet of social media content — roughly the height of the Statue of Liberty. In that endless feed, your ad has approximately 1-3 seconds to earn attention before a thumb swipe sends it into oblivion. The hook — your opening line, visual, or question — is the single most important element of any ad, yet most advertisers spend 90% of their time on product features and 10% on the one thing that determines whether anyone sees the rest.
Research from Meta shows that 65% of viewers who watch the first 3 seconds of a video ad will continue watching for at least 10 seconds, and those who reach 10 seconds are highly likely to watch the full ad. This means the hook isn't just about grabbing attention — it's a filter that determines your entire campaign's performance. A great hook with an average offer will outperform a great offer with a weak hook every single time. This guide teaches you how to write hooks that stop the scroll, organized by type, with a clear writing process and real examples across industries.
What Is a Hook and Why Does It Determine Ad Performance?
A hook is the opening element of your ad — typically the first 1-3 seconds of a video or the first line of ad copy — that either captures attention or loses it forever. In the context of video advertising, the hook includes the opening visual, the first words spoken, and any text overlay that appears immediately. Together, these elements create a split-second decision in the viewer's mind: 'Is this worth my attention?' The answer to that question determines whether your ad budget generates results or evaporates.
The data is unambiguous: 65% of viewers who watch the first 3 seconds will continue to 10 seconds, and viewers who reach 10 seconds are 3x more likely to convert. This cascade effect means your hook is the single largest lever you have over campaign performance. A 20% improvement in hook rate (the percentage of viewers who watch past 3 seconds) can reduce your cost per acquisition by 30-40% — because more people see your offer, your message, and your call-to-action. Every dollar spent on improving hooks delivers more ROI than any other creative optimization.
4 Proven Hook Types That Stop the Scroll
Provocative Question
Open with a question that challenges an assumption or creates instant curiosity. Questions engage the brain differently than statements — they trigger an automatic response where the viewer mentally tries to answer, which keeps them watching. Examples: 'What if everything you know about dieting is wrong?' or 'Why do 80% of new businesses fail in the first year?' The question must be relevant to your audience's lived experience to work.
Shocking Statistic
Lead with a surprising number that stops people mid-scroll. Statistics work because they provide concrete proof that something unexpected is happening. The key is specificity — '73.6% of businesses waste their ad budget' is more compelling than 'most businesses waste money on ads.' Pair the statistic with a brief visual representation or text overlay to reinforce the impact in those critical first seconds.
Relatable Problem
Start by describing a frustration your target audience experiences daily. When people see their own struggle reflected in an ad, they immediately feel understood and want to hear the solution. This hook type works especially well for UGC-style ads: 'I used to spend 6 hours editing one video ad...' or 'Does anyone else feel overwhelmed trying to run ads on five platforms?' Authenticity and specificity are what separate a good relatable hook from a generic one.
Counterintuitive Statement
Make a bold claim that contradicts conventional wisdom. Counterintuitive hooks work because they create cognitive dissonance — the viewer thinks 'that can't be right' and watches to find out more. Examples: 'Stop optimizing your ads — here's why' or 'The best ad creative takes 5 minutes, not 5 weeks.' The claim must be defensible and lead into a genuine insight; clickbait that doesn't deliver will destroy trust and tank engagement rates.
How to Write High-Converting Hooks: Step by Step
1. Identify your audience's #1 pain point
Before writing a single word, get crystal clear on the primary frustration, fear, or desire of your target audience. Mine customer reviews, support tickets, Reddit threads, and competitor comments for the exact language people use to describe their problems. The best hooks use the audience's own words, not marketing jargon. Write down the top 3 pain points and rank them by emotional intensity.
2. Choose your hook type based on the message
Match your hook type to your campaign objective and audience awareness level. Provocative questions work best for problem-aware audiences. Shocking statistics work for skeptical or data-driven audiences. Relatable problems work for UGC and testimonial-style ads. Counterintuitive statements work for audiences who think they already know the solution. Choose one type per ad — don't mix approaches in a single hook.
3. Write 10 hook variations in 15 minutes
Set a timer and write 10 different hooks without self-editing. Quantity produces quality — your first 3-4 hooks will be obvious and generic, but hooks 7-10 often surprise you with unexpected angles. Write each hook as the exact first words of your video or the first line of your ad copy. Keep each hook under 10 words for video and under 15 words for text ads. Speed prevents overthinking.
4. Filter through the 3-second test
Read each hook aloud and ask: 'If I saw this while mindlessly scrolling at 11pm, would I stop?' Eliminate any hook that requires context to make sense, uses industry jargon, or sounds like an ad. The best hooks feel like the beginning of a conversation, not a sales pitch. Narrow your 10 hooks down to 3-5 finalists based on this filter. Get a second opinion from someone in your target audience if possible.
5. Test your top hooks in live campaigns
Create identical ad variations that differ only in the hook — same offer, same CTA, same targeting. Run them simultaneously with equal budget for 3-5 days. Measure hook rate (3-second video views / impressions), watch-through rate, and ultimately CPA. The winning hook will usually be obvious within 48-72 hours with sufficient budget. Scale the winner and repeat the process every 2-4 weeks to combat creative fatigue.
Test 10 hooks in one afternoon
Create AI video ads nowHook Examples by Industry
| Industry | Hook Example | Why It Works | Expected CTR |
|---|---|---|---|
| E-commerce | I found a $12 product that replaced my $200 routine | Specific price contrast creates instant curiosity and perceived value | 2.5–4.0% |
| Fitness | I lost 15 lbs without giving up a single food I love | Counterintuitive — contradicts the sacrifice narrative of fitness | 3.0–5.0% |
| Education | This 10-minute habit made me fluent in Spanish in 6 months | Specific timeframe + surprising speed creates believability | 2.0–3.5% |
| SaaS | We deleted 4 tools from our stack and grew 3x faster | Counterintuitive — less tools = more growth challenges assumptions | 1.8–3.0% |
| Restaurants | The secret menu item 90% of our customers don't know about | Exclusivity + curiosity — viewers feel they're missing out | 3.5–5.5% |
| Beauty | My dermatologist said stop using this popular ingredient | Authority + warning — triggers loss aversion and concern | 2.8–4.5% |
Use AI to Test Hooks at Scale
The biggest advantage you can have in hook writing isn't creative talent — it's testing velocity. The business that tests 20 hooks per month will always outperform the one that tests 2, because advertising performance is fundamentally a numbers game. Most marketers know this, but the production bottleneck of creating new video creative for each hook variation makes high-velocity testing impractical with traditional methods.
AI video generation eliminates this bottleneck entirely. You can write 10 hook variations in 15 minutes, generate a unique video for each one in under an hour, and have all 10 running in live campaigns by the afternoon. This means you can find your winning hook in days instead of months, and when creative fatigue hits, you can refresh with new variations immediately. The combination of systematic hook writing and AI-powered video production is the most powerful creative testing workflow available to advertisers in 2026.
Frequently asked questions
What makes a good hook for a video ad?
A good hook stops the scroll in 1-3 seconds by creating immediate curiosity, emotional resonance, or cognitive dissonance. It must be relevant to the target audience's experience, specific rather than generic, and make the viewer need to know what comes next. The best hooks use the audience's own language and address their primary pain point directly without sounding like an advertisement.
How many hooks should I test per campaign?
Test 3-5 hook variations per campaign cycle, with each variation getting equal budget for at least 3-5 days. This gives you enough data to identify a statistically significant winner while keeping production manageable. Top advertisers test 10-20 hooks per month across all their campaigns. The more hooks you test, the faster you find winners — aim to increase your testing velocity over time.
How do I know if my hook is working?
Track hook rate — the percentage of viewers who watch past 3 seconds. On Meta, this appears as 'ThruPlay rate' or '3-second video views.' A hook rate above 30% is good, above 45% is excellent. Also monitor the relationship between hook rate and downstream metrics like CPA: a hook that gets high viewership but low conversions may be attracting the wrong audience through misleading promises.
Should my hook mention the product or brand name?
Not in the first 3 seconds. Leading with your brand name or product is the most common hook mistake — viewers don't care about your brand yet; they care about their own problems. Save brand mentions for after you've earned attention. The exception is well-known brands with strong positive associations, but even then, leading with a compelling hook and introducing the brand at second 3-5 typically outperforms.
Sources
Related articles
Find your winning hook with AI video
Generate dozens of hook variations as video ads in minutes — test at scale and find what stops the scroll for your audience.
Start free