Hooks: The First 1.5 Seconds Decide Everything

A viewer decides whether to keep watching before your first sentence ends. Not after your intro, not once your point lands — within roughly a second and a half, their thumb has voted.

That sounds brutal until you internalize what it means: the hook isn't the start of the video. The hook is the video. Everything after it is payoff for a promise the first moment made.

Why the window is this short

Swiping costs nothing and the next video might be better, so leaving is the default state. Recommendation systems watch that first moment closely — a video that loses most of its viewers in the first two seconds gets quietly buried, no matter how good second fifteen is.

The practical consequence: spend a disproportionate share of your effort on the first 1.5 seconds. A mediocre video with a great hook outperforms a great video with a mediocre hook, essentially every time. Unfair, but it's the arena.

Hook types that work

A hook operates on two channels at once — what's said (or shown as text) and what's on screen. You need both firing.

The direct callout. Name the viewer so precisely they feel caught.

• “If you run a service business and you're still not posting video, this is for you.”
• “POV: you built a great product and nobody knows it exists.”

The contrarian open. Attack something your audience currently does.

• “Stop posting product demos. They're killing your account.”
• “Your website is the reason your ads don't convert.”

Contrarian hooks earn attention through tension — the viewer stays to see whether you can back it up. Only open this way when you actually can.

Result-first. Show the end state before the process.

• “This entire video calendar came from one URL. Here's how.”
• Open on the finished thing — the dashboard, the transformation, the number you can stand behind — then rewind.

The curiosity gap. Withhold one specific thing.

• “There's a setting inside TikTok that most brands never touch.”

It works, but it's the most abused hook type in the feed. The gap has to close with a real payoff, or you're training viewers to distrust you.

The visual interrupt. Sometimes the hook isn't verbal at all: an unexpected first frame, something already mid-motion, text slightly too big for the frame. Feeds are pattern-recognition environments — being ten percent off-pattern visually buys the moment your spoken hook needs.

Patterns you can adapt

Steal these skeletons and load them with your niche:

• “Nobody tells [audience] this, but [uncomfortable truth].”
• “I tested [thing] for [time period] so you don't have to.”
• “The fastest way to [outcome] isn't [expected method].”
• “[Impressive artifact] — made in [surprisingly short time].”
• “Watch me [do the thing] in real time.”

A pattern isn't a script; it's a slot machine you load with specifics, and the more concrete the specifics, the better it performs. “Grow faster” is dead on arrival. “Fill a month of TikToks from your homepage” gives a thumb a reason to pause.

The mistakes that kill hooks

Starting with your logo. An animated logo intro is a skip button. Nobody is loyal to your brand at second zero.

Warming up. “Hey guys, welcome back, today I want to talk about...” is four seconds of nothing. Cut every video at the last possible moment before the interesting part.

Context before payoff. Founders love backstory — why they built the thing, the state of the industry. Viewers don't. Lead with the payoff and earn the right to give context later.

Overpromising. A hook the video can't cash gets you the view and costs you the viewer. It shows up in the data too: bait produces a wave of skips right after the hook, and that mid-video cliff is a quality signal pointing the wrong way.

One hook, thirty videos. Even your best hook fatigues. Rotate types deliberately — a callout today, result-first tomorrow — so the account doesn't become one repeated trick.

Treat hooks as their own production step

Don't write a script and bolt a hook onto it. Write hooks first, in batches — twenty in a sitting, before any scripting. Kill the weak ones while they're still one line of text, then script only for the survivors. And when you review the month, review hooks separately from videos: same topic with a different hook is often the highest-return retest you can run.

This ordering is baked into how Storymode generates scripts — hooks get written and varied first, before anything else about a video exists. Whatever your process, order it the same way. First 1.5 seconds first.