Creative and hooks guide
Viral hooks: real hook ideas from Videos that beat their Creator's usual reach
By Jackson Gathondu. Published 1 November 2026. Updated .
Most lists of hook ideas are written from memory. The hooks on this page come from real TikTok Videos that got many times more views than their Creator usually gets, and each one is quoted exactly as the Video opened.
We pulled them from 12 Research Runs on our own account on 1 November 2026, then Scored the Videos against each Creator's median views. Below you get the hooks, the templates our analysis wrote from them, what the wider benchmarks say about hook types, and the steps to find hooks like these in your own niche.
What counts as a viral hook here
A hook is the first line and the first picture of a short Video: the part that decides whether a viewer keeps watching. Millions of views alone do not make a hook viral, since a large account gets millions of views on an ordinary post.
So every hook below passed a stricter test. Its Video beat its own Creator's median views, measured by a Score. A Score of 50x means the Video got 50 times the views of that Creator's typical recent Video. That way a breakout on a 600-follower account counts as much as one on a large account.
The hook type, the opening line and the visual hook come from the Video analysis. When a Video had no speech, the opening line is the caption or the words on screen, quoted as written.
15 TikTok hooks that beat their Creator's usual reach
Each row is one TikTok Video from our Research Runs. Views against the Creator's median come from a Score taken the same day. Very large multiples usually belong to small accounts, where one Video can go far past a median of a few hundred views.
| Hook, word for word | Hook type | Niche and Creator | Views against the Creator's median |
|---|---|---|---|
| In what world is this an extra small? | Question | Fashion, @dhayaspamz on TikTok | 1,440x (839,586 views, median 583) |
| it only takes $27.397 a day to spend $10,000 a year. let that sink in. | Curiosity gap | Personal finance, @budgetwithsam on TikTok | 1,341x (6,500,000 views, median 4,848) |
| Wait. Is he pointing at me? | Question | Pet products, @katieandgus on TikTok | 1,194x (4,000,000 views, median 3,349) |
| The first time I saved up $10,000, it felt like I had the world in my palm. | Story | Personal finance, @itsnotyaboygerad on TikTok | 1,112x (1,147,394 views, median 1,032) |
| Welcome back to understanding the skin, and today we're talking about eye creams. | Direct address | Skincare, @freshfacety on TikTok | 827x (1,100,000 views, median 1,330) |
| I will literally die on this hill. That skin toner is one of the most underrated skincare products that you can put in your routine, not to mention usually one of the most affordable, too. | Bold claim | Skincare, @izzythepa on TikTok | 560x (2,800,000 views, median 4,999) |
| it's 9:30pm and I'm craving a coffee so. badly. i wish there was a nighttime drink i loved as much as coffee | Problem | Coffee, @paving.my.way_ on TikTok | 494x (520,240 views, median 1,054) |
| 9 Things Your Dog Wishes You Knew | Listicle | Pet products, @r8vr7eak9bi2 on TikTok | 465x (198,151 views, median 426) |
| Here's how to invest your first $1,000, coming from a millennial with over $1 million invested at 37. | Social proof | Personal finance, @morgan.on.money on TikTok | 370x (2,100,000 views, median 5,676) |
| These are my top home purchases, not from 2026, but from the past decade. | Curiosity gap | Home goods, @em__sarah__ on TikTok | 348x (460,500 views, median 1,322) |
| The person who invented this really deserves a raise. | Bold claim | Pet products, @wyliekqjfks on TikTok | 287x (279,198 views, median 973) |
| As a pharmacologist gym rat, I want you to know that not all protein is equally effective for building muscle. | Problem | Supplements, @pharmgymrat on TikTok | 202x (179,900 views, median 892) |
| This is just your reminder that the beans you use in your Breville espresso, absolutely matter and let me show you why. | Problem | Coffee, @kayyfrankk on TikTok | 108x (49,300 views, median 455) |
| How to look more expensive. | Direct address | Fashion, @oliviasara_23 on TikTok | 11.8x (177,800 views, median 15,050) |
| If you want to find your personal style, stop looking at clothes. | Bold claim | Fashion, @margarita.muradova on TikTok | 7.8x (110,153 views, median 14,100) |
What these hooks have in common
Most of them name a specific person, number or object in the first sentence: an extra small that does not fit, $27.397 a day, a pharmacologist, Breville beans. A viewer knows within one line who the Video is for.
Many open on tension and leave it unresolved. The question hooks ask something the Video then answers, and the problem hooks state a pain the viewer shares (craving coffee at 9:30pm, protein that does not work) before showing a fix.
None of them open on a brand name. Only one says who the Creator is, the pharmacologist, and it does so to back up the claim in the same sentence. In the rest, the first words are about the viewer or the thing being shown.
Hook templates written from real hooks
When a Research finishes, its Digest lists the hooks that stood out and turns each one into a template with blanks. These are quoted from those Digests, blanks and all, with the hook they came from. Swap in your own product, audience and numbers.
| Template | Hook type | From this hook |
|---|---|---|
| it only takes [number] a day to spend [number] a year. let that sink in. | Curiosity gap | it only takes $27.397 a day to spend $10,000 a year. let that sink in. (TikTok, personal finance) |
| In what world is this an [size]? | Question | In what world is this an extra small? (TikTok, fashion) |
| Why does this [product] look [mistake]? The [product] is fine, but nothing feels connected. Let's fix it with [number] simple changes. | Question | Why does this room look cheap? The furniture is fine, but nothing feels connected. Let's fix it with six simple changes. (TikTok, home goods) |
| Stop buying expensive [product]... | Problem | Stop buying expensive dog treats... (TikTok, pet products) |
| Girl to girl, I tried [product] to [outcome], but it looks [mistake] than before. | Problem | Girl to girl, I tried pimple patches to calm the bump, but it looks redder than before. (TikTok, skincare) |
| If you're trying to [outcome], stop [mistake]. | Problem | If you're trying to get in shape, stop overcomplicating it. (Instagram Reels, fitness) |
| If your [project] is still sitting at [negative state], I think I know why, and it's probably not because you need a better [solution]. | Problem | If your SaaS is still sitting at zero MRR, I think I know why, and it's probably not because you need a better product. (YouTube, B2B SaaS) |
| If you want to find your [outcome], stop looking at [mistake]. | Bold claim | If you want to find your personal style, stop looking at clothes. (TikTok, fashion) |
| There are two types of [professionals], ones who think [task] is easy, and ones who've actually done it. | Bold claim | There are two types of marketers, ones who think B2B marketing is easy, and ones who've actually done it. (YouTube, B2B SaaS) |
| Here's how to [action] your first [number], coming from a [audience] with over [number] [outcome] at [number]. | Social proof | Here's how to invest your first $1,000, coming from a millennial with over $1 million invested at 37. (TikTok, personal finance) |
| [Audience] just starting a [product] listen up. This will take [number] of your week. | Direct address | Ladies just starting a fitness journey listen up. This will take 5% of your week. (Instagram Reels, fitness) |
| [number] [product] on the App store that deserve a spot in your daily [outcome] rotation. | Listicle | 4 games on the App store that deserve a spot in your daily gaming rotation. (TikTok, mobile games) |
Which hook types get the most views
The TikTok hook benchmarks measure every hook type across 480 analysed TikTok Videos from 10 Research Runs, captured the same day. There, a Video's views are compared with the median Video in its own niche rather than with its Creator.
Problem hooks had the highest median views of the common hook types, at 1.97x the niche median across 61 Videos, and 49% of them reached at least 2x. Curiosity gaps came next at 1.48x and beat their niche's median in 7 of 10 Runs, the most consistent hook type across niches. Demonstrations were the most common opening (28.5% of Videos) but sat at 0.85x, and Videos with no hook did worst at 0.40x.
Opening lines were short. 71.5% of the TikTok Videos had a spoken or written opening line, the median line was 10 words, and the middle half ran 7 to 17 words.
Treat these as starting points for tests. Research picks Videos that are doing well for the brief, so this is a sample of what works in each niche, not a random sample of TikTok.
How to find more viral hooks in your niche
Ask your Assistant for each step in plain language, or call the REST API. Paid steps can quote their price first if you ask for an estimate.
Run a Research on your niche
One Research searches the Platforms you pick, ranks Videos by how far they beat their Creator's usual reach, analyses the shortlist and writes a Digest. It costs 1,036 credits for up to 50 analysed Videos, and reading it again later is free.
Research hooks in TikTok skincare videos for women in their 20s, then show me the Digest hooks.Read the Digest hooks and templates
Each Digest hook comes with its verbatim text, a template with blanks, the hook type and a short reason it works, plus the lift of that hook type in the Research. Reading the Digest is free.
List every hook in that Digest with its template, and mark the hook types that beat the Research median.Pull every opening line
The Research Videos list carries each Video's analysed hook: the type, the opening line, the visual hook and why it works. Listing them is free, so you can sort and filter as often as you like.
From that Research, list every Video's opening line and hook type, sorted by views.Score the Videos against their Creators
A Score compares a Video's views with its Creator's median, which tells a breakout hook apart from a big account. A TikTok Score costs 9 credits per Video. For Instagram Reels under 30 days old, a Creator lookup returns Scores on each Reel for 6 credits.
Score the top ten TikToks from that Research and keep the ones over 10x their Creator's median.Read the full scripts
The transcript shows what the Video says after the hook and how it answers it. A transcript on its own costs 10 credits, and a full Video analysis 23 credits.
Get the transcripts of the five highest-Scored Videos and show how each one pays off its hook.Write your own hooks
Ask for new hooks built on the templates that worked, each one naming the source Video, so you can check the evidence before you film.
Write ten hooks for my vitamin C serum from the templates of the Videos over 10x, and link the Video each one borrows from.
Real output: one hook and its Score
Both samples come from our own account on the day this guide was written, trimmed to the fields that matter here. The first is one Video from the skincare Research, with its analysed hook. The second is the Score for the same Video, taken about 35 minutes later, so its view count differs slightly.
TikTok, via REST API
POST /v1/research with intent for TikTok skincare Videos, platforms tiktok, max_videos 50, then GET /v1/research/{id}/videos
{
"items": [
{
"id": "7696670183980567821",
"platform": "tiktok",
"creator": { "handle": "izzythepa", "follower_count": 2801 },
"posted_at": "2026-10-14T23:31:37.000Z",
"snapshot": { "views": 2766950, ... },
"intelligence": {
"hook": {
"type": "BOLD_CLAIM",
"rationale": "A strong, passionate statement right away grabs the viewer's attention and sparks curiosity.",
"visual_hook": "woman speaking directly to camera with glowing skin",
"opening_line": "I will literally die on this hill. That skin toner is one of the most underrated skincare products that you can put in your routine, not to mention usually one of the most affordable, too."
},
"source": "speech",
"visual": { "format": "TALKING_HEAD", ... },
...
}
},
...
]
}TikTok, via REST API
POST /v1/videos/tiktok/7696670183980567821/score
{
"run": { "kind": "video_score", "status": "completed", "charge_credits": 9, ... },
"result": {
"video": {
"id": "7696670183980567821",
"snapshot": { "views": 2800000, "likes": 208200, "saves": 52719, ... },
"scores": {
"outlier": 560.11,
"outlier_p90": 116.67,
"percentile": 100,
"velocity": { "window": "snapshot", "views_per_hour": 2196.94 },
"baseline": {
"sample": 16,
"median_views": 4999,
"p90_views": 24000,
"confidence": "high",
"posts_per_week": 4.41,
...
}
},
...
},
...
}
}How to read that output
The outlier field is the Video's views divided by the Creator's median views over their recent Videos, here 16 of them. 560.11 means this Video got about 560 times what the Creator usually gets. The outlier_p90 field does the same against the Creator's 90th percentile, a stricter bar, and it is still 116.67.
The baseline confidence is high because enough recent Videos stood behind the median. With low confidence, read the multiple as a rough guide.
The source field says what the analysis had to work with. speech means it heard the opening line; captions_only means the line was taken from the caption or the text on screen.
Data limits
Data limits for hook research
- score_video, which Scores one Video on demand, is TikTok only. Instagram Reels under 30 days old get Scores through a Creator lookup; older Reels come back aged out.
- YouTube Creator data covers Shorts only, not long-form uploads. Scores come back on a channel's Shorts through get_creator; score_video does not work on YouTube. Video analysis takes YouTube Videos up to 10 minutes long.
- Instagram Keyword search is best effort, so an Instagram Research can return fewer Videos than you asked for.
- Facebook Video search is opt-in and best effort. Facebook Videos under 30 days old get Scores through a Page lookup, the same as other Platforms.
- Refreshing a Research's view counts is TikTok only.
- When a Video has no speech, its opening line comes from the caption or the text on screen. In the benchmark sample that was 208 of 480 TikTok Videos.
What it costs
Each request that does work is a Run, paid from a prepaid Balance of Credits at 1,000 credits per US dollar. There is no subscription. A Research costs 1,036 credits for up to 50 analysed Videos, a TikTok Score 9 credits and a transcript 10 credits. New accounts get 1,000 credits free, valid for 90 days. A Run that fails costs nothing, and a partial one is charged only for what finished. Every price is on the pricing page.
To read the script of one Video without an account, paste its link into the free video transcript generator.
Connect your Assistant
The same server works in every Assistant below. None of them needs an API key: you sign in to ShortsIntel and choose which organization pays.
Claude.ai
Web and desktop. No API key: Claude signs in to ShortsIntel for you.
- Open Claude's add connector page with ShortsIntel filled in, and choose Add. Or go to Settings, Connectors, Add custom connector, and paste the endpoint below.
https://www.shortsintel.com/api/mcp - Sign in to ShortsIntel, approve the connection and choose the organization whose Balance pays for Runs.
- In a chat, turn the ShortsIntel connector on and ask your question.
ChatGPT
Web and desktop, in Developer mode. No API key: ChatGPT signs in over OAuth.
- Open ChatGPT's connector settings and switch Developer mode on.
- Create a connector named ShortsIntel, paste the endpoint and choose OAuth.
https://www.shortsintel.com/api/mcp - Sign in to ShortsIntel and choose your organization.
- In a new chat, choose the + next to the message box, then Developer mode, and turn ShortsIntel on.
Claude Code
One command, then sign in from Claude Code.
- Add the server in your terminal.
claude mcp add --transport http --scope user shortsintel https://www.shortsintel.com/api/mcp - Run /mcp inside Claude Code, choose shortsintel and sign in.
Codex
One command; Codex signs in on first use.
- Add the server in your terminal.
codex mcp add shortsintel --url https://www.shortsintel.com/api/mcp - If no browser window opens to sign in, start it yourself.
codex mcp login shortsintel
Cursor and VS Code
One-click install links; each asks you to confirm.
- In Cursor, open the Cursor install link.
- In VS Code, open the VS Code install link.
- Sign in to ShortsIntel when the editor asks.
Questions
- What is a viral hook?
- A viral hook is the opening line and first shot of a short Video that keeps people watching long enough for the Video to spread. On this page a hook only counts if its Video beat its own Creator's median views, so large accounts do not win by default.
- What are good hook ideas for TikTok?
- In 480 analysed TikTok Videos, problem hooks and curiosity gaps beat the niche median most often, and Videos with no hook did worst. The 15 hooks on this page, such as "In what world is this an extra small?", each got far more views than their Creator usually gets.
- How long should a hook be?
- In the benchmark sample the median TikTok opening line was 10 words, and the middle half ran 7 to 17 words. Several hooks on this page are under 10 words.
- Are these hooks from TikTok only?
- The 15 hooks with Scores are all TikTok, because one-Video Scores are TikTok only. The templates also include hooks from Instagram Reels and YouTube.
- Can I copy these hooks?
- Use the templates rather than the exact words. Each template keeps the shape of a hook that worked and leaves blanks for your own product, audience and numbers.
- How do I find viral hooks in my own niche?
- Run a Research on your niche, read the hooks and templates in its Digest, Score the top Videos against their Creators, and read the transcripts of the ones that beat their Creator's median by the widest margin. An Assistant connected to ShortsIntel can do each step from a plain-language request.
Try it in your Assistant
Connect ShortsIntel to the Assistant you already use and ask the question this guide answers, or try the calls in the Playground first.