The same video with several openings, so the test finds the one that stops the scroll.
Make me [3 / 5] different hooks for this video, for an A/B test. The video: [a direct link] The product: [what it is] (photo attached if you have one) Who it is for: [the audience] Each hook a different mechanism: the result first, a question, a pattern break, a bold claim, a close-up with a sound. The product or the promise is on screen in the first two seconds of every hook, and everything after the hook stays identical. Write me the hooks first. Show me the plan and one itemized price first, and start only after I approve.
seedance-25: Keep the exact product from the reference images. Same colorway, logo placement, silhouette and materials. Do not invent a sibling SKU. Do not beautify the print into a different design. Extreme close-up, the finished result fills the frame for one second, then a fast pull back reveals the product in a hand, one continuous move, hard daylight, crisp detail, the product readable by the second second.
Use it when a video works in the middle and at the end, but people scroll past in the first seconds. An A/B test measures one thing at a time, so here only the first seconds change and everything after them stays identical. The test then tells you which opening stops the scroll, and nothing else.
How it works: Cadabra first reads the video, finds its shots and its strongest moment. Then it writes three to five hooks, each a different mechanism: the result first, a question, a pattern break, a bold claim or a close-up with a sound, and shows them to you before anything costs money.
There are three ways to build a hook: open on a strong moment already inside the video, which is the cheapest; film a new opening shot, trimmed to its first two or three seconds and followed by the original video; or replace only the first line of text on screen.
A tip: the product or the promise belongs in the first two seconds of every hook. A curious opening that hides what is being sold stops the scroll but does not sell. And run every version to the same audience, so the only difference between them is the hook.
The preset builds three to five. Three are enough for a first test, and five let you try every mechanism. Each version needs enough views for the result to be clear.
When the opening shots are new, one version is quoted and each of the others costs the same. The price is shown before anything runs, and writing the hooks is free.
That works. A product photo is added only if you have one, and then it is held in every new shot. Without it, the promise itself has to be on screen in the first two seconds.
עמוד הבית · בלוג · וידאו AI · איך עושים סרטון AI · מוזיקה AI · תמונות AI · קול AI · תמונה לסרטון · כתוביות בעברית · כמה זה עולה · אפקטים לעריכה · פריסטים · כל המנועים · קרדיטים ולא מנוי · AI בעברית · קורס מתחילים · מדריכים · נגישות
Cadabra AI · סלע מייטל (עוסק פטור) · Meital Sela · רחובות