How many hooks should I test at once?

4 to 6, if your budget allows. With more, each hook gets too little budget for a clear result. On a small budget, test 2 or 3. You may miss a strong one, but the method is the same.

Which metric should I use to judge a hook?

The 3-second view rate first, then cost per sale for the best 2 or 3. Cost per sale alone mixes the hook with the rest of the ad.

Can I test new hooks on an ad that already works?

Yes, and it is the best place to start. The rest of the ad already sells, so you only test the opening.

Is the first frame part of the hook?

Yes. People see the first image before they hear a word, so treat it as part of what you test.

What if all the hooks perform the same?

Then the hook is probably not the problem. Look at the offer or the audience next.

Can I film the main part again between versions?

No. A new recording is a different performance, so you no longer know what caused the difference. Use one recording for every version.

Guide

How to test hooks cheaply: one video, several openings

Updated Sep 22, 2026

Film 4 to 6 different openings for the same video and run each version as its own ad. Judge them first on who keeps watching, then on cost per sale.

In short

The cheapest way to test hooks is to film 4 to 6 openings for one video and run each version as an ad.
Everything after the first 3 seconds stays the same, so any difference in results comes from the hook.
Stop the hooks that the fewest people watch past 3 seconds. That shows up much sooner than sales.
Give the best 2 or 3 hooks more budget and pick the winner on cost per sale.
Most of the cost is ad spend. In Cutroom, one edit plus six 30-second versions costs about $2.56.

Smart woman with mobile in hand listening attentively to interlocutor and looking down while sitting in front of modern computer — Photo by www.kaboompics.com on Pexels

Openings to test for one video: 4-6
The part of the ad that changes between versions: 3 sec
Cutroom credits for one edit plus six 30-second versions (160 credits): $2.56

What is a hook test?

A hook test runs versions of one ad that differ only in the first few seconds.

The hook is what people see and hear first: the opening line, the first image and any headline on screen.
Every version has the same middle and the same ending. Only the opening changes.
That way, a difference in results can only come from the opening.

What changes between the versions

Part of the ad	Between versions
Opening (first 3 seconds)	Different
Main part	Same recording
Proof	Same
Call to action	Same
Captions and music	Same

Part of the ad: Opening (first 3 seconds)
Between versions
Different
Part of the ad: Main part
Between versions
Same recording
Part of the ad: Proof
Between versions
Same
Part of the ad: Call to action
Between versions
Same
Part of the ad: Captions and music
Between versions
Same

How to test hooks step by step

You film once and run several versions of the same ad.

Pick clearly different ideas, such as naming the price, asking a question or showing the result first.
Keep the openings about the same length, or you end up testing length instead of words.
Film all the openings in the same sitting, with the same light and background as the rest.

A hook test, step by step

Write 4 to 6 openings
Different ideas, not rewordings of one
Film them in one sitting
The main part once, then each opening
Make one version per opening
Same middle and ending in every version
Run them as separate ads
Same audience and budget for each
Stop the weakest hooks
Lowest 3-second view rate
Pick the winner
Lowest cost per sale among the best 2 or 3

This is the whole editor

Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.

How to judge a hook test

Judge in two rounds: first on attention, then on sales.

Round 1 uses the 3-second view rate, the share of people who watched at least 3 seconds. Many advertisers call it the hook rate.
In Meta Ads Manager, you get it by dividing 3-second video plays by impressions.
Round 2 uses cost per sale. Give the best 2 or 3 hooks more budget until each one has sales.
A hook that wins round 1 but loses round 2 is getting attention from the wrong people.
Never pick a winner on views alone. You end up with an ad people watch but do not buy from.

How much does a hook test cost?

Most of the money goes on ad spend, because every hook needs enough viewers to compare.

In Cutroom, one edit plus six 30-second versions uses 160 credits, about $2.56 on a paid plan.
The rest is ad spend. Give every version the same budget in round 1, so six versions cost six times what you give one.
Judging round 1 on 3-second views saves the most, because weak hooks are stopped before they use much budget.
Testing on an ad that already sells is cheapest. The rest of the ad is proven, so you only pay to test the opening.
Test 4 to 6 openings if your budget allows. On a small budget, test 2 or 3. You may miss a strong one, but the method is the same.

Common hook test mistakes

Each of these can waste a whole test, and each one is easy to avoid.

Hook test mistakes and how to avoid them

The main part changes
The mistake
One version gets a new recording
How to avoid it
Use the same main recording in every version
Hooks too similar
The mistake
Six rewordings of one idea
How to avoid it
Four to six clearly different ideas
No end point
The mistake
The test runs on with no limit
How to avoid it
Set the budget or end date before launch
Judged on views only
The mistake
The most-watched hook wins
How to avoid it
Confirm the winner on cost per sale

	The mistake	How to avoid it
The main part changes	One version gets a new recording	Use the same main recording in every version
Hooks too similar	Six rewordings of one idea	Four to six clearly different ideas
No end point	The test runs on with no limit	Set the budget or end date before launch
Judged on views only	The most-watched hook wins	Confirm the winner on cost per sale

Where Cutroom fits

Cutroom makes the versions from one upload. You film the main part and all the openings in one video, up to 3 minutes long.

For each version, delete the other openings in the transcript and export. Undo brings them back for the next version.
Cutroom drafts an on-screen headline for the opening, or you write your own.
One edit uses 100 credits and each 30-second version uses 10 more to export. An edit plus six versions is 160 credits, about $2.56.
Lite is $19.99 a month for 1,250 credits, and Basic is $39.99 a month for 2,500.
Your first video is free, with no card, and you have 7 days to make it. It has a Cutroom watermark until you upgrade.

Slab hook style — Opening headline styles in Cutroom
Keep one headline style for every version, so only the words change.

Punch hook style — Opening headline styles in Cutroom
Keep one headline style for every version, so only the words change.

Questions people ask

How many hooks should I test at once?: 4 to 6, if your budget allows. With more, each hook gets too little budget for a clear result. On a small budget, test 2 or 3. You may miss a strong one, but the method is the same.
Which metric should I use to judge a hook?: The 3-second view rate first, then cost per sale for the best 2 or 3. Cost per sale alone mixes the hook with the rest of the ad.
Can I test new hooks on an ad that already works?: Yes, and it is the best place to start. The rest of the ad already sells, so you only test the opening.
Is the first frame part of the hook?: Yes. People see the first image before they hear a word, so treat it as part of what you test.
What if all the hooks perform the same?: Then the hook is probably not the problem. Look at the offer or the audience next.
Can I film the main part again between versions?: No. A new recording is a different performance, so you no longer know what caused the difference. Use one recording for every version.

Film one video with several openings and let Cutroom make the versions. Your first video is free.

Try for freeYour first video is free, no card, and you have 7 days to make it. It carries a Cutroom mark; Lite at $19.99/month removes it