Every menu vendor claims their thing lifts sales. Here are the numbers that hold up, the psychology behind why video works better than photos, and an honest look at when it won't help you.
Short answer: yes, when the video is on the menu where people order — not on social media. But the useful version of that answer needs the actual numbers and the caveats.
That last point is the one to anchor on. The photo-versus-no-photo gap is well established across every major ordering platform. Video is the same mechanism, turned up: it shows texture, steam, motion and portion size in a way a single flat image can't.
It answers the question a photo leaves open. A picture of a burrito doesn't tell you how big it is. A two-second clip of someone picking it up does. Portion uncertainty is a real reason people don't order unfamiliar dishes.
It rescues food that photographs badly. Curries, stews, ramen, biryani, dals — brown food in a bowl. These are frequently the best things on the menu and the worst-selling items online, purely because they look flat as a still. Motion is what makes them read as delicious: the steam, the spoon pulling through, the sauce moving.
It stops the scroll. Guests skim menus fast. Movement is the one thing the human eye can't ignore, so whatever you film gets attention — which is why what you film matters as much as whether you film.
Video moves orders toward whatever you filmed. That's the entire mechanism — and it cuts both ways.
If you film your cheapest, lowest-margin item, you'll sell more of your cheapest, lowest-margin item. Your order count goes up and your profit doesn't. This is the single most common way restaurants waste a video menu.
Film your highest-margin dishes first. Then your catering or family packs, because high-ticket group orders are the hardest sell from text alone and benefit most from seeing the whole spread.
Being straight about this:
Don't film the whole menu and hope. Pick three high-margin dishes, add video to only those, and leave the rest of the menu unchanged. Then compare those three items' order share for 30 days before and after.
You're looking for a shift in mix — those items taking a bigger share of orders — plus movement in average order value. That's a real read, not a vibe.
The pattern is consistent: the lift concentrates in the first few dishes you film, catering and family-size items respond most strongly, and food that photographs poorly gains the most. You do not need a filmed menu of 60 items. Three to six good clips carry most of the result.
If you want to see it working on a real restaurant rather than a mockup, our partner Chef's Kebab runs one live. And if you'd rather not hold the camera yourself, the AIOrders Video Menu includes the filming.