Hook Intelligence

Ad Hook Library

The first three seconds decide everything. 24 hook types ranked by their share of 1,671 decoded winning ads — with the psychology, verbatim openers and writing rules behind each one.

1,671 ads decoded24 hook types6 hook familiesFree — no login

Key takeaways

  • 24 distinct hook types identified across 1,671 decoded winning video ads, grouped into 6 psychological families.

  • Curiosity Spike leads at 8.9% — but no single hook exceeds 9%, which means there is no dominant opening and no safe default.

  • Competence Restoration is the single largest psychological driver at 25.5% of decoded ads, ahead of Loss Aversion at 21.3%.

  • The strongest predictor of a hook failing is a mismatch between hook family and audience awareness — not the quality of the line itself.

Rankings

The 5 hooks that win the most

Share of decoded winning ads opening on each hook type, across all categories.

The full table

All 24 hook types, ranked

Every opening classified in the decoded set, ordered by frequency. No hook exceeds 9% — the distribution is long-tailed, which is why testing across families beats optimising inside one.

  1. 1Curiosity Spike
    8.9%
  2. 2Open Loop Statement
    7.6%
  3. 3Identity Callout
    6.4%
  4. 4Data Point Start
    6.1%
  5. 5Problem Agitation
    5.8%
  6. 6Contradiction Hook
    5.4%
  7. 7Visual Shock
    5.2%
  8. 8Results Flash
    4.7%
  9. 9Story Start
    4.6%
  10. 10Loss Aversion Warning
    4.3%
  11. 11Us vs. Them
    4.1%
  12. 12Sound-Off Text Hook
    3.9%
  13. 13Mistake Callout
    3.6%
  14. 14Authority Opener
    3.4%
  15. 15Secret Reveal
    3.3%
  16. 16Day-in-the-Life
    3.3%
  17. 17Persona Mirror
    2.9%
  18. 18Unexpected Action
    2.9%
  19. 19Social Proof Stack
    2.8%
  20. 20Before-After Tease
    2.6%
  21. 21Urgency Trigger
    2.1%
  22. 22Negative Hook
    2.1%
  23. 23Confession Opener
    2.1%
  24. 24Belonging Signal
    1.9%

Browse

The six hook families

Each family shares one psychological mechanism. Pick the family that matches your audience's awareness level first, then pick the hook inside it.

By category & persona

Category-specific ad hooks

The same 24 hooks rank completely differently depending on what you sell and who is running the ads. 20 product verticals and 10 audience personas, each with its own ranking, psychology and swipe file.

View all 30 topics

Psychology

What actually drives the opening

Hook types are surface executions. Underneath them sit six mechanisms that do the persuasion — and knowing which one you're firing matters more than which line you write.

Competence Restoration

25.5%

The viewer is shown a gap in their own capability and offered a way to close it. The largest single driver in the decoded set, and the reason mistake callouts and how-to framing travel so well.

Loss Aversion

21.3%

An existing cost is made visible. Because a loss is felt roughly twice as intensely as an equivalent gain, this is the cheapest way to convert inertia into motion.

Closure Delivery

17.9%

An open loop is created and then paid off. The satisfaction of resolution is what retains viewers through the middle of an ad, where most drop-off happens.

Identity Signaling

14.2%

The purchase is attached to who the viewer is or is becoming. Slower to convert than fear, but far more durable against price comparison.

Novelty Reward

11.8%

Something genuinely unseen is presented. Fires pre-cognitively and reliably, but decays faster than any other mechanism as the feed adapts.

Social Consensus

9.3%

Evidence that people like the viewer already decided. Works best late in an ad, where it resolves the skepticism Delivery creates.

Method

How to pick a hook

Most hook testing fails before it starts, because it tests the wrong variable. Running five versions of a curiosity hook against each other tells you which sentence performs best inside a family you have not yet validated — an answer that does not transfer to your next product or even your next campaign.

Start one level up. Pick one hook from three different families and run those against each other. The result tells you which mechanism your audience actually responds to, and that finding survives everything: new creative, new offers, new products in the same category.

The second most common error is a temperature mismatch. Cold audiences have no problem to be reminded of, so fear and loss hooks land as manufactured. Problem-aware audiences already know what hurts, so curiosity hooks feel like a delay. Match the family to what the viewer already knows, and the specific line matters far less than you would expect.

Unaware

Pattern Interrupt or Story

The viewer has no problem in mind. Buy attention visually or narratively before you attempt to argue anything.

Problem-aware

Fear & Loss or Curiosity

They know it hurts but not why. Name the cost, or open the gap that explains it.

Solution-aware

Proof or Identity

They are comparing options. Lead with evidence or with a callout that says this one is for you.

Most aware

Proof, then a hard offer

They need a reason to act now. Skip the setup and put the strongest evidence first.

At a glance

The library in numbers

Ads decoded

1,671

Hook types

24

Hook families

6

Top hook share

8.9%

Frequently asked questions

What is an ad hook?

An ad hook is the opening move of a video ad — the first one to three seconds whose only job is to buy the next three. It is not the same as the ad's message or its offer. A hook can be a line of dialogue, an on-screen text card, a visual, or an action, and the same underlying product can be sold behind any of the 24 hook types in this library.

Which ad hook works best?

Curiosity Spike opens the largest share of decoded winning ads at 8.9%, but "most common" is not the same as "best for you". The right hook depends on audience temperature and category. Cold audiences reward curiosity and pattern interrupts; problem-aware audiences reward fear and loss framing; research-driven buyers reward proof. The strongest single predictor of a hook failing is a mismatch between the hook family and the audience's awareness level, not the hook itself.

How many hooks should I test?

Test hook families before you test hook variants. Running five curiosity hooks against each other tells you which sentence is best inside a family you haven't validated. Running one hook from each of three families tells you which mechanism your audience responds to — a far more useful result, and one that survives being applied to your next product.

Does the hook matter more than the rest of the ad?

It matters first, not most. A weak hook caps how many people see the rest of the ad, so it sets the ceiling. But a strong hook attached to a weak middle produces high three-second view rates and no conversions, which is the single most common misdiagnosis in creative testing. If view rate is high and conversion is low, the problem is Delivery, not the hook.

How long should a hook be?

One to three seconds, carrying exactly one idea. The most common structural error in the decoded set is a hook that tries to establish the brand, the product and the claim simultaneously. Every additional idea in the opening frame reduces the odds that any of them is retained.

Where does this hook data come from?

These shares come from 1,671 high-performing video ads decoded by the Klip Kanvas research team, sampled from public ad libraries and creator platforms across eight D2C verticals. Each opening is classified into one of 24 hook types and one of six psychological families. The figures are editorial research intended to guide creative decisions, not audited platform metrics.

Explore more

Ready to put this into practice?

Create your first AI UGC video ad in minutes — no filming, no actors, no editing.

Try Klip Kanvas free

Ready to make ads like these?

Paste a product link and Klip Kanvas writes the script, casts the creator and renders the ad — no filming, no actors, no editing.