The hook is the very start of your ad, the first few seconds of a video or the first line and image of a static, and it is the part that decides whether someone scrolling past stops or keeps going.
When your ad shows up in a feed, the person looking at it is moving fast and has already scrolled past a dozen other things, so you are competing for a sliver of their attention against everything else on the screen. The hook is what you spend that sliver on. If it lands, it buys you the next few seconds and the chance to actually make your case; if it does not, the person is already gone, and the rest of the ad, the offer, the proof, the call to action, might as well not exist because nobody scrolled far enough to see it.
This is why so much of an ad's success or failure is settled right there at the opening. You can write a great offer and build a landing page that converts, but if the first moment does not stop the thumb, none of that work ever gets in front of anyone. That is also why experienced buyers put a large share of their creative effort into the first line and the first frame rather than spreading it evenly across the whole ad, because that opening is doing most of the heavy lifting.
A hook has to connect to what you are actually selling, not just grab attention for its own sake. A shocking or clickbait opener that has nothing to do with the product will pull in the wrong people, burn through impressions on an audience that was never going to buy, and can read as a bait and switch that costs you trust the moment they realize what the ad is really about. The hooks that work stop the right person by speaking to something they genuinely care about, and then the ad delivers on exactly that.
Stop the right person for the right reason, and the rest of the ad finally gets a chance to work.
Sources
- This page explains a creative practice used across platforms, not a single Meta setting, so there is no platform figure to source. Any example is illustrative.
Last checked 9th August 2026. Next check 15th August 2026.
