I Studied 100 Viral Hooks, These 6 Will Make You Go Viral

Kallaway

Abstract

Twenty-two minutes on hooks, structured as psychology, then six archetypes, then the component that decides whether any of them work. The mechanism is contrast — set an expectation, introduce an alternative, and the distance between them is the pull — and the six formats are six ways of generating that distance.

• Any single idea can be framed through all six archetypes, so the format is a choice rather than a constraint.
• Every hook has four components: the spoken line, the visual, the on-screen text and the audio.
• Alignment between those four is a multiplier, not an addition — it is the difference between a few hundred views and hundreds of thousands.
• Viewers take a video in as visual, then audio, then visual again, which is why what gets shown matters more than what gets said.
• With no strong visual, the correct move is often to question the video rather than to write harder.
• The decision rule ends on throwing the video out, which is what makes the rest enforceable.

Everything below follows the video in order, from the premise through the six formats, into the four components and the alignment argument, out through the five-step build and two worked examples — one that won and one that failed. For anyone building a personal brand the useful part is the inversion: the hook is the image, and the sentence is support.

Chapter summaries

00:00 - The hook decides whether a short video wins, and it is the highest-return thing to improve

The hook is the single biggest determinant of a short-form video winning or losing, and the highest-return thing to improve.

"the biggest difference between winning and losing on social media is the hook this is the"

— 00:00

Of everything that can be changed about a short video, this is the one with the largest effect and the lowest cost. That makes it the highest-return place to spend attention, which is the premise the rest of the breakdown rests on.

00:38 - A hook has one job, and it is opening a loop nobody can leave

A hook has one job: to open a curiosity loop the viewer cannot comfortably leave unclosed.

"first before anything else we need to understand the psychology behind how hooks work this is"

— 00:38

Stated that narrowly, most openings are revealed to be doing something else — introducing, contextualising, greeting. None of the three is wrong in itself, and none of them opens a loop, which is the only job on offer here.

00:57 - Curiosity is manufactured by contrast, and the distance is the pull

Curiosity is manufactured by contrast: set an expectation (A), introduce an alternative (B); the distance between them is the pull, and bigger distance means a deeper hook.

"video now the best way to create that Curiosity Loop is with contrast the viewer starts"

— 00:57

Set an expectation, then introduce an alternative. The distance between them is the pull, and a bigger distance makes a deeper hook. Contrast is the manufacturing process rather than a stylistic choice.

01:26 - The fortune-teller hook sets the present against a changed future

Fortune-teller hook: place the present against a changed future to provoke curiosity about what is coming.

"all right now the first hook format is called the fortune teller and it's all about"

— 01:26

The present set against a changed future. It provokes curiosity about what is coming, and it works because the viewer already lives in the first half of the comparison.

02:30 - The experimenter hook shows a demo peer to peer

Experimenter hook: show a demo or experiment peer-to-peer, as one person showing another what they found.

"in a minute okay now the second hook format is called the experimentor and this one"

— 02:30

A friend-to-friend look what I tried reads as low-risk and invites watching through to the result. Openings where the device or method being tested is itself the on-screen subject carry it — best for product demos, new methods, and showing how a tool works.

04:41 - The teacher hook extracts the method from something learned

Teacher hook: frame a lesson teacher-to-student, extracting the method from something you learned or observed.

"right let's keep going now the third hook format is called the teacher and this one"

— 04:41

Positioning as the one who has already solved it promises the viewer a shortcut past the failure. Openings that promise a set of lessons drawn from a case that worked are the shape — best for building authority in a category in order to sell that expertise.

05:55 - The magician hook forces attention onto something outlandish

Magician hook: a visual or verbal stun gun that forces attention onto something outlandish, usable alone or stacked before another format.

"sell that expertise okay now the fourth hook framework is a little bit different than the"

— 05:54

A visual or verbal stun gun forcing attention onto something outlandish. It works alone and it stacks — placed in front of another format it buys the attention that format then spends.

08:00 - The investigator hook builds contrast against something not yet known

Investigator hook: build contrast against a secret or finding the viewer does not yet know.

"all right now the fifth hook format is called the investigator and this one is super"

— 07:59

The gap between not-knowing and knowing is itself the loop, and revealing closes it. Openings tee up a hidden thing, a little-known place, an unexplained result — best for creators positioning at the leading edge or in a research capacity.

09:11 - The contrarian hook states a belief against the conventional wisdom

Contrarian hook: state directly a belief that runs against conventional wisdom.

"people don't know okay and the sixth and final hook format is called the contrarian and"

— 09:10

A belief stated directly against the conventional wisdom, without hedging. The hedge is what kills it, and the format only works when the position is genuinely held — because the contrast comes from the reader's existing belief rather than from anything the sentence supplies. A manufactured contrarian line produces the shape without the pull.

10:26 - Any single idea can be framed through all six archetypes

You do not have to match a format to a topic: any single idea can be framed through all six archetypes, so the format is a choice, not a constraint.

"contrarian now I'm going to tell you a secret you don't actually have to to figure"

— 10:26

The contrast can be placed against the future, a demo, a lesson, a stun, a secret or a belief, for the same subject. One product gets worked through all six openings in turn, which is what makes the archetypes a menu rather than a diagnosis.

11:51 - Every hook is made of four components

Every hook is made of four components: the spoken line, the visual, the on-screen text, and the audio.

"right now all hooks are made up of four components the spoken hook which is what"

— 11:51

These are the four channels through which the opening reaches the viewer at once. They are named as the parts to be engineered separately and then matched, rather than written as one line.

12:06 - The multiplier is alignment between those four

The multiplier is alignment: the gap between a few hundred and hundreds of thousands of views is the four components pointing at the same thing.

"big secret right here the difference between 500 and 500,000 views is unlocking Max alignment between"

— 12:06

Any mismatch between what is said, shown, written and heard splits the viewer's understanding across two versions of the same moment. Even a slight difference between the spoken and the visual hook is said to reduce effectiveness.

12:27 - Alignment matters because a viewer who does not understand cannot be hooked

Alignment matters because of comprehension: a viewer who does not understand what they see and hear cannot be hooked.

"perfectly and the reason this matters is because because of comprehension comprehension is the degree to"

— 12:27

A viewer who does not understand what they are seeing and hearing cannot be hooked by it. Comprehension is the precondition, and misalignment costs comprehension before it costs interest.

13:14 - A video gets taken in as a visual, then audio, then visual again

Viewers take a video in as a visual-audio-visual sandwich: they see the visual and text first, then hear the spoken line, then look back at the visual to confirm it.

"going to tell you this is what all viewers do the first thing they do is"

— 13:14

The visual and text first, then the spoken line, then a look back at the visual to confirm it. Eyes are said to process ten to a hundred times more information per second than ears, which is why the order runs that way.

13:55 - The most important part of a hook is what gets shown

Because the visual leads, the most important part of the hook is what is shown, not what is said — the "key visual".

"what they heard okay so tactically tying all this together what does this actually mean and"

— 13:55

Because the visual leads, what gets shown matters more than what gets said. The key visual is the hook and the sentence is support — which inverts how almost everybody writes them.

14:39 - A five-step build starts from the visuals that already exist

A five-step build: inventory the visuals you have, find the biggest-contrast angle and let it pick the format, write the spoken hook, add supporting text, then watch it back and gut-check comprehension.

"it again one more time but more organized so the very first step is you need"

— 14:39

Each step resolves one component in the order the viewer will process it, which is what makes the sequence matter rather than the list. It is restated a second time in a tightened, ordered form.

14:48 - With no strong visual, the video itself is in question

If you have no strong visual, either manufacture one with motion graphics or stock footage, or seriously question whether to make the video at all.

"will be to set up the contrast back to the visual if you don't have a"

— 14:48

A hook built on a weak visual starts behind, because the leading channel is empty before a word is written. The question is posed as a gate at step one, ahead of any writing.

15:39 - Step five is a comprehension check rather than a polish pass

Step five is a comprehension gut-check: if the visual-speech-visual pass is not clearly aligned, stop and rework the hook rather than ship it.

"someone hears the speech and then step five I'm going to watch it back and gut"

— 15:38

If the visual-speech-visual pass is not clearly aligned, the hook gets reworked rather than shipped. It is a stop condition rather than a polish step, which is the difference that makes it worth naming as a step.

16:00 - A winning example carried the contrast inside the visual itself

A worked winning example: a key visual carrying embedded contrast — something most viewers have never seen — needs almost no verbal contrast to work.

"right now really quick I just want to walk through a couple examples of my own"

— 16:00

The key visual carried the contrast inside it — something most viewers had never seen — so almost no verbal contrast was needed. When the image does the work the writing gets easier, which is the practical payoff of leading with the visual.

17:44 - The winning hook was a stun tag, a description and a future framing

The winning hook was written as a stun tag, then a context-lean line describing exactly what is shown, then a future framing — perfect alignment of visual and speech.

"had to actually write the hook so context lean contrast contrarian take so I literally said"

— 17:44

A stun tag, then a context-lean line describing exactly what is on screen, then a future framing. The description of the visual is doing the alignment work, and it reads as redundant right up until it is missing.

18:37 - The failing example promised something the visuals could not depict

A worked failing example: an abstract promise the visuals cannot depict, worsened by jargon, produces comprehension loss and far fewer views.

"views now before we end the video Let's quickly go through one bad example where I"

— 18:37

An abstract promise the visuals could not depict, made worse by jargon. Comprehension collapsed and the views followed. The failure is diagnosed at the alignment rather than at the writing.

20:13 - The fix was a more concrete framing, and the right move was to kill it

The fix was a more concrete, showable framing; lacking the b-roll to support it, the correct move was to kill the idea rather than post it.

"looking back maybe what I should have done has gone more explicit so these rooms are"

— 20:13

A hook that cannot be made clear is signalling a video that should not be made. The abstract promise gets reworked into a concrete one, and the conclusion is still that it should have been canned — there are always more ideas, so abandoning one costs little.

20:33 - The decision rule ends with throwing the video out

The decision rule for any video: identify the visual, pick the format, write the hook, then test for clarity and alignment — and throw the video out if it is not there.

"what you need to decide when you're going through this process what is the visual what"

— 20:33

The rule ends on the discard, which is the part that makes the rest enforceable.

Personal Branding Lessons

An episode about openings, with an unusually strict test at the end. The moves below are the working method.

Manufacture curiosity through contrast

Set an expectation, then introduce an alternative. The distance between the two is the pull, and a bigger distance produces a deeper hook. That makes curiosity something built to specification rather than something stumbled into on a good day. 00:57

Frame one idea through all six archetypes before choosing

The format is not dictated by the topic — any single idea can be run through all six. Trying them all before committing costs a few minutes and almost nobody does it, which is why most openings end up in whichever format the creator defaults to. 10:26

Align the spoken line, the visual, the text and the audio

Four components, and alignment between them is a multiplier rather than an addition. A strong line inside a misaligned hook underperforms a mediocre line inside an aligned one, which is why writing alone rarely fixes a hook that is not working. 12:06

Start from the visual rather than from the sentence

Viewers take a video in as visual, then audio, then visual again, and eyes process far more per second than ears. What gets shown therefore matters more than what gets said — an inversion of how almost everybody writes an opening. 13:55

Inventory the visuals you already have first

The build starts with an inventory rather than an idea: what footage exists, what can be shown. The biggest-contrast angle then picks the format, and the writing comes after both. Reversing that order is what produces hooks nothing can illustrate. 14:39

Question the video when there is no strong visual

Either manufacture one with graphics or stock footage, or seriously ask whether the video should exist. The second option is offered as a real answer rather than as rhetoric, which is rare in advice of this kind. 14:48

Gut-check comprehension before shipping

Watch it back and check the visual-speech-visual pass is clearly aligned. If it is not, the hook gets reworked rather than published. It is a stop condition rather than a polish step, and that distinction is what makes it worth naming. 15:39

Throw the video out when the alignment is not there

In the failing example the fix was a more concrete framing — and without the footage to support it, the correct move was to kill the idea rather than post it. The rule ends on the discard, which is the part that makes every step before it enforceable. 20:33

Questions

Each answer ends at the moment in the recording where it is given.

What actually decides whether a short-form video succeeds or fails?

The single biggest lever named is the hook — the opening seconds — described as the 80/20 of the whole thing. The argument is that attention is decided before the body is reached, so a stronger opening lifts everything after it and a weak one caps the result no matter how good the rest is. It is offered as a pattern seen across the best performers rather than a proven law, but the practical takeaway is to move most of your effort to the front, where what the platform rewards is set. 00:00

How does a hook create curiosity in the first place?

Through contrast. You set an expectation, then introduce something that does not match it, and the gap between the two is what the viewer's attention fixes on — a curiosity loop it feels a pull to close, like a rabbit hole. A wider gap is described as a deeper hook. The catch is time: the contrast has to register in roughly five seconds or it does not take hold, so the mismatch must be immediate and legible rather than slow-burning or clever. 00:57

What are the different types of hooks I can choose from?

Six recurring shapes are named: a future-teasing opening, a peer-to-peer demo, a teacher's lesson, a scroll-stopping visual or verbal stun, a hidden-secret reveal, and a flat counter-belief. Each is simply a different place to put the same contrast — against a teased future, a demonstration, a lesson, a shock, a secret, or a stated disagreement. Each is tied to the kind of creator it suits, which is one way to stand out in a crowded space. 01:26

Do I have to match a particular hook type to my topic?

No. The playbook explicitly dissolves the idea that a shape belongs to a topic: any single idea can be poured into all six openings in turn, shown by running one ordinary product through every one. That makes the archetype a choice rather than a constraint the material forces on you. The useful question therefore moves downstream — not which shapes exist, but which framing carries the biggest honest contrast for the idea in front of you right now. 10:26

Why do two similar videos get wildly different view counts?

Alignment. A hook is broken into four channels — what you say, what you show, what you write on screen, and what you hear — and the difference between a small result and a large one is named as those four pointing at the same thing. A mismatch, even a slight one between the spoken line and the image, splits comprehension and weakens the hold. The figures used are a way of stating the size of the gap, not measured outcomes from a specific test. 12:06

What matters more in a hook, what I say or what I show?

What you show. The account given is that a viewer takes a clip in as a sandwich: they see the image first, then hear the words, then look back to confirm the two match — because sight registers faster than sound, offered as a rough ratio rather than a measured figure. Because the visual lands first, it carries the most weight, so the opening should be built around a chosen "key visual" before any line is written. 13:14

Is there a repeatable process for writing a hook?

Yes — a five-step build. Inventory the visuals you have; let the biggest-contrast angle pick the format; write a two-to-four-line spoken hook that describes the image; add supporting on-screen text; then watch it back and gut-check comprehension. Each step resolves one channel in the order a viewer will meet it. Treated as a repeatable system rather than inspiration, it makes the blank page a sequence of small decisions instead of one large one. 14:39

How do I know whether my hook is good before I post it?

The final step is a comprehension gut-check. You watch the opening back and ask whether, going visual to speech to visual, a viewer would be clear on what the video is about and whether that matches where the rest of it goes. If the pass is not clearly aligned, the instruction is to stop and rework the hook rather than move forward. The review exists precisely to catch misalignment before it ships, not to approve what you already made. 15:39

When should I abandon a video idea instead of posting it?

When the hook cannot be made clear and aligned. One of the creator's own clips stalled because it promised an abstract idea the footage could not depict and leaned on unfamiliar terms; the honest conclusion was that the right move had been to abandon the idea rather than rescue the hook. An unshowable opening is treated as a signal about the whole video. Because a scrapped post is a reversible decision and ideas are not the scarce resource, cutting it costs little. 20:33

Sources

I Studied 100 Viral Hooks, These 6 Will Make You Go Viral

The sections above follow the episode's own order. Each timestamp is the point in the recording where the idea is discussed, so the post reads as a map of the conversation rather than a rearrangement of it.

Quotations come from the episode transcript with spoken filler ("okay", "right", "you know", "um", "actually", "kind of", "sort of") and false starts removed. No words are added: square brackets mark anything inserted for sense, and an ellipsis marks any cut. British spellings are restored, and speech-to-text errors were repaired against context only where the intended word is unambiguous. Where a line appears both in a cold-open teaser and in its place in the conversation, the timestamp cites the second — the point at which it is actually said.

Quotes are stamped with the moment they are said rather than with a speaker's name.