How to test 5 hooks from one shoot

01What is a hook test?

A hook test runs several versions of the same ad that differ only in their opening. Everything after the hook stays the same: the demo, the claims and the call to action. When one version holds viewers longer or converts better, you know the hook caused it, because nothing else changed.

That separation is the point. If you change the hook, the demo and the ending at once, a winner tells you almost nothing about why it won.

02Why test hooks instead of whole ads?

Because the opening is where a viewer decides whether to keep watching. Someone scrolling gives an ad a moment's attention, and the opening has to earn the rest. A strong demo can't help if no one stays to see it.

Testing hooks also costs less than testing whole ads. You film the demo once, and each extra hook is one more short take, not another shoot.

03How do you film five hooks in one shoot?

Plan the shoot as the hooks first, then everything that follows them.

1. Write five genuinely different hooks. Each one should test a different idea, not different wording of the same idea. A useful set covers several angles:

   Angles a hook can take, and what each opener does

   Angle	What the opener does
   The problem	Names the frustration your viewer already has
   The result	Shows or states the outcome before explaining how
   The question	Asks the question your buyer is already asking
   The call-out	Says exactly who the product is for
   The objection	Answers the doubt that stops people buying

2. Film the hooks back to back. Say each hook to camera as its own take, while your energy and light match. A hook can also be a reaction or a reveal with no words; each take needs to be at least 1 second long. Film a second take of any line you stumble on. The hook is the whole opening moment, so what's on screen can change too: hold the product, point at the problem, show the result. What follows it can't.
3. Film the demo once, so it works after any hook. Show the product while you explain it: an unboxing, a walkthrough or a screen recording. A silent screen recording works too. Never refer back to one opener ("like I just said"); start the demo with a line that follows any of them. Mashcut takes up to 10 demo clips.
4. Film one call to action that fits every hook. One clear ending line, said to camera.

Keep the whole set, meaning every hook, the demo and the call to action, to 10 minutes of footage or less.

04How do you turn one shoot into five ads?

By hand, it's five edits: trim each hook, cut the pauses and retakes out of the demo, add captions, and assemble each version. That's where the hours go.

Mashcut does that editing for you. You pick the clips on your iPhone, and Mashcut's servers do the cutting:

1. Pick your hooks. Choose up to 5 hook takes from your camera roll.
2. Pick your demo. Choose the demo clips, and your call to action if you filmed one.
3. Tap Generate. Mashcut keeps one take of each line you repeated, cuts the ums and pauses, and adds captions from your own words. It makes one ad per hook: each opens with a different hook, followed by parts of your demo and your call to action.
4. Review and export. Watch each ad and fix any caption. Each export is a 9:16 video at 1080×1920, up to 35 seconds long, ready for TikTok, Reels or Shorts.

Mashcut edits your footage; it never generates new video, voices or claims. Every word in the ad is one you said on camera or typed yourself.

05How do you run the test and pick a winner?

Use your ad platform's own testing tool, change only the hook, and judge every version on the same metric.

• Use the platform's test tool, not on-and-off switching.
  • Meta: create an A/B test in Ads Manager. Meta advises (opens in a new tab) keeping the ad sets identical except for the variable you're testing, and not using the test's audience for any other campaign you're running.
  • TikTok: its split testing (opens in a new tab) guide lists video hooks, the opening 2–3 seconds, as a creative variable you can test.
• Run it long enough. Meta recommends (opens in a new tab) a minimum of 7 days for an A/B test, and its tests run for 30 days at most. Early numbers swing.
• Give every version the same budget.
• Judge on how many people watch past the opening.
  • Meta: divide 3-second video plays (opens in a new tab) by impressions. Many advertisers call this the hook rate.
  • TikTok: do the same with 2-second video views (opens in a new tab) and 6-second video views (opens in a new tab).
  • Decide before you start which result, conversions or cost per result, breaks a tie.
• Iterate on the idea, not just the ad. When a hook wins, film your next batch as variations of it.

06How do you deliver hook tests to a brand?

Deliver the ads as a labelled set, and price the variants as their own line.

• Label every version. Mashcut puts each ad's number in its exported file's name, so you and the brand can tell the versions apart.
• Quote hook variants as a line item. Each extra version costs you one more take on set and one credit, not another edit. Put the number of hook variants in the brief, and bill for them.

07What mistakes break a hook test?

• Changing more than the hook. A new demo or a new ending in one version means you can't tell what caused the win.
• A demo that only fits one hook. "As I just said" breaks every other version.
• Testing near-duplicates. Five rewordings of one opener test the wording, not the idea.
• Calling it too early. An early lead often disappears once more people see the ads.
• Editing each version from scratch. Rebuilding the same demo five times is where the time goes. Reuse it.