UGC hooks that stop the scroll: 25 examples

A hook is the first one to three seconds of a short-form video — the opening line, the opening frame, and the on-screen text that land before a viewer decides whether to keep scrolling. Its only job is to earn the next three seconds; it is not the place to explain your product. It is also the highest-leverage variable in UGC, because the same footage re-cut with a stronger opening can move from a few hundred views to a few hundred thousand.

What does a hook actually have to do?

Every upload gets an audition. The platform shows a new video to a small test audience, watches what they do in the first seconds, and decides from that whether the video graduates to a bigger batch — the full mechanism is in how short-form reach works. A hook that fails doesn't underperform; it ends distribution before the rest of the video is ever seen. That is why a retention cliff in the first two seconds is the most common cause of TikToks getting no views.

Three layers fire at once, and all three are the hook:

• Spoken: the first sentence out of the creator's mouth, ideally started mid-thought with no greeting in front of it.
• Visual: what is on screen in frame one — a face, motion, a mess, a result. Not a logo, not a slow zoom, not a title card.
• Text: the caption burned onto the video, short enough to read at a glance, roughly eight words or fewer.

The rules underneath every good hook are the same. Be specific — a number, a name, a real situation beats an adjective. Create tension the viewer wants resolved. Cut the runway: no "hey guys," no intro, no throat-clearing. And never open on the product; the product arrives once the story has.

25 UGC hook templates that stop the scroll

Each of these is a fill-in-the-blank pattern with one concrete example. They are written for a study-app style product so the examples stay comparable — swap the nouns for your category and the shape still holds.

Curiosity gap — open a loop the viewer has to close.

• "Nobody tells you ___." → "Nobody tells you the first month of nursing school is the one that filters everyone out."
• "I was today years old when I found out ___." → "I was today years old when I found out my phone had a study timer built in."
• "The reason ___ isn't what you think." → "The reason your practice scores stalled isn't what you think."

Problem callout — name the pain out loud in second one.

• "If you ___, stop." → "If you study with your notes open, stop — you're rereading, not recalling."
• "Still ___? It's probably ___." → "Still failing practice exams? It's probably your review method, not your memory."
• "You don't have a ___ problem, you have a ___ problem." → "You don't have a motivation problem, you have a plan problem."

POV and relatable — put the viewer inside a moment they've lived.

• "POV: ___" → "POV: it's 11pm, the exam is Friday, and you haven't opened the study guide."
• "Me explaining to ___ why ___" → "Me explaining to my roommate why I have 47 tabs open."
• "That moment when ___." → "That moment when the practice test asks the one thing you skipped."

Contrarian — take a side the audience can argue with.

• "Stop ___. Do this instead." → "Stop highlighting your textbook. Do this instead."
• "Everyone says ___. They're wrong." → "Everyone says you need eight hours of studying a day. You don't."
• "Unpopular opinion: ___." → "Unpopular opinion: flashcards are the most overrated study tool there is."

Listicle — a number promises a finite, skimmable payoff.

• "___ things I wish I knew before ___." → "Three things I wish I knew before my first clinical rotation."
• "___ ___ under $___." → "Four study tools under $20 that actually did something."
• "Ranking ___ from worst to best." → "Ranking every study method I tried this semester, worst to best."

Social proof and result — lead with the outcome, not the process.

• "I did ___ for ___ days. Here's what happened." → "I used this every day for 30 days. Here's what happened."
• "___ people asked how I ___, so here it is." → "Forty people asked how I went from a 62 to an 84, so here it is."
• "My ___ went from ___ to ___." → "My practice score went from 62 to 84 in three weeks."

Demo tease — promise a thing they have to watch to believe.

• "Watch what happens when I ___." → "Watch what happens when I run my notes through this."
• "You can ___ in ___ seconds." → "You can turn a 40-page PDF into a quiz in about ten seconds."
• "I didn't know ___ could do this." → "I didn't know my phone could do this."

Warning — a cost the viewer is about to pay.

• "Do not ___ until you ___." → "Do not buy another study guide until you watch this."
• "If you're about to ___, watch this first." → "If you're about to start your first clinical, watch this first."

Before and after — show the gap in frame one, explain later.

• "This is ___ before. This is after." → "This is what my notes looked like before. This is after."
• "Day 1 versus day ___." → "Day 1 versus day 30 of studying like this."

Why can the same video 10x with a new hook?

Because the hook is what gets graded. The body of a video only affects performance for people who stayed, and the hook decides how many of those there are — so re-cutting the first two seconds changes the input to the algorithm's most important test while leaving everything else identical. That makes re-hooking the cheapest edit in short-form: an hour spent putting three new openings on footage you already have will usually beat an hour spent filming something new.

It also makes the hook the wrong thing to improvise. Treating it as an afterthought is one of the more expensive UGC mistakes, and writing the hook options into the brief up front is how you avoid it — our UGC content brief template asks creators for three hook variants per concept before anything is filmed.

How do you test hooks at volume?

One video tells you nothing. Hook testing only becomes real data when you run enough variants close enough together that platform conditions are comparable. The method:

• Change one thing. Same body footage, same audio, same length — only the opening line, frame, and text move.
• Run three to five variants per concept, published within the same week.
• Tag every post by hook type — curiosity, contrarian, POV, warning — so you're scoring patterns, not individual videos.
• Judge on retention first. The 3-second retention rate tells you whether the hook worked; views tell you what happened after it did.
• Give it a floor. Two posts is an anecdote. Ten posts in a hook category is a signal worth acting on.

Volume is the whole unlock, which is why this works better as a program than as a one-off shoot. Our Medceptor campaign ran 10 creators to 300 unique videos published 1,200 times in 30 days — that's a sample size where hook patterns separate from luck, and it produced 4.1M views, 224.4K engagements, and a 38% revenue lift (numbers in the case studies). The same logic explains the Memo result: 145 posts pulled 4.29M views from an account with 2,672 followers, because each video earned its own audience at the hook.

The weekly hook loop we run

Same five steps every week, for every client:

• Monday — score. Pull last week's posts, sort by 3-second retention, group by hook type.
• Tuesday — sort winners from noise. Anything above the account's median retention with enough posts behind it goes in the winners pile; the bottom quartile gets cut, not fixed.
• Wednesday — brief. Winning hook patterns become next week's templates, each with three variants so the test continues.
• Thursday and Friday — re-hook. Proven footage gets new openings and reposts; nothing that worked gets published only once.
• Every week — widen. One deliberately new hook type per creator, so the library keeps growing instead of converging on the same three formats.

Run that for a month and hooks stop being a creative guess and become a ranked list your creators write from. It's the same loop underneath everything in what actually drives virality on TikTok — and it's the reason volume compounds instead of just accumulating. If you'd rather see the whole engine than build it, how it works walks through it end to end, and UGC examples shows what these hooks look like once they're on screen.