The Hook Formula I Use for Every Reel (With Real Examples)

I've written over 400 Reel hooks in the past year. About 60 of them went over 100K views. The rest flopped. Here's the pattern I noticed in the winners.

What a Hook Actually Is

A hook isn't your title. It's not your thumbnail. It's the first 1-3 seconds of your Reel that determines whether someone keeps watching or scrolls away. In those 3 seconds, the viewer makes an unconscious decision: is this worth my time?

My 4 Hook Templates That Work

Template 1: The "I Tried" Hook

Structure: "I [did something extreme] for [time period]. Here's what happened."

Example: "I posted a Reel every single day for 30 days. My growth actually slowed down."

Why it works: It promises a story with a counterintuitive ending. People stay to see the twist.

Template 2: The "Stop Doing" Hook

Structure: "Stop [common action]. Do this instead."

Example: "Stop using 30 hashtags. Use 3. Here's why."

Why it works: It challenges a common belief. People feel compelled to defend their position or learn something new.

Template 3: The "Number" Hook

Structure: "[Number] [things/tips/mistakes] that [result]."

Example: "The 3 lighting mistakes that make your Reels look amateur."

Why it works: Specific numbers feel actionable and scannable.

Template 4: The "Question" Hook

Structure: "What if [provocative scenario]?"

Example: "What if you could grow your Instagram without showing your face?"

Why it works: It creates an information gap. The viewer needs to know the answer.

The Hooks That Never Work

"Hey guys, welcome back to my channel." Nobody cares. Start with value, not introduction.

"Today I'm going to talk about..." Just talk about it. Don't announce it.

"So I was thinking..." Sounds hesitant. Confident hooks perform better.

How I Test Hooks Before Filming

I write 10 hooks for every Reel I plan. Then I read them out loud to myself. If I wouldn't keep scrolling to watch, I rewrite it.

The best test: show the hook to a friend. If they ask "and then what happened?" you've got a winner.

Related resources: Explore more at the Meta Business Help Center and Google Analytics.

The Anatomy of a Viral Hook

Every viral hook has three components: a pattern interrupt, a curiosity gap, and a promise of value. The pattern interrupt stops the scroll. The curiosity gap makes them need to know more. The promise of value justifies their time.

"I spent $2,400 testing creator tools so you do not have to." Pattern interrupt: high dollar amount. Curiosity gap: which tools? Promise of value: save money.

"The caption formula that took me from 2% to 5% engagement." Pattern interrupt: specific numbers. Curiosity gap: what formula? Promise of value: better engagement.

FAQ

Can I reuse the same hook template? Yes, with different angles. My "I tried X for Y days" template has generated 12 viral Reels with different topics.

How long should the hook be? Under 5 seconds. Every extra second costs you viewers.

Do hooks work for carousels too? Absolutely. The first slide of a carousel is the hook. Same psychology, different format.

Case Study: Testing 50 Hooks in 30 Days

I wrote 50 hooks for 10 Reels and tested them. Emotional hooks got 2.3x more views. Contrarian hooks got 1.8x more. The combination got 3.1x more.

My best hook ever: "I wasted $2,400 on creator tools so you do not have to." It reached 1.4 million people.

Hook Testing System

Write 10 hooks per Reel. Read each out loud. If it does not make you curious, delete it. Test the winner as a text post first. If it gets engagement, film it.

Hook Testing Protocol

I test hooks systematically. Phase one: write ten hooks per concept. Rate each on curiosity, emotion, and clarity. Only hooks scoring above seven of ten proceed.

Phase two: post the top three hooks as text stories. Track poll responses. The winner becomes the filmed hook.

Phase three: film with the winning hook. If performance exceeds average, add the formula to my reusable template library.

This protocol takes forty-five minutes per concept but increases average views by sixty percent.