You have everything you need to make a video ad. You have a phone that shoots in 4K. You have an editing app that used to cost a studio thousands of dollars. You have AI tools that will spin up footage, voiceovers, even a whole talking person, for the price of a coffee. So you sit down, you make the thing, you launch it, and it does nothing. It just sits there, spending money, while you quietly wonder what you did wrong.
Here is what you did wrong. You treated the ad like something you shoot, when it is really something you write. The camera, the app, the AI, none of that is the hard part anymore, and none of it is where the ad is won or lost. The ad is won or lost in the script, the actual sequence of words and moments that carries a stranger from a thumb-flick to a purchase. The most experienced performance advertisers, the kind who have spent hundreds of millions of dollars figuring this out, will tell you the same thing: something like 80 percent of an ad's performance is decided before a single frame gets shot. It is decided on the page.
The good news is that a selling script is not a bolt of inspiration that either strikes you or doesn't. It has a shape. It is closer to a recipe than a painting, and once you can see the recipe, you can write to it on purpose, every time, instead of staring at a blank timeline hoping something clever arrives. That is what this is: the shape of a video ad that sells, beat by beat, so you can build one whether you shoot it yourself or hand the script to someone else.
Before You Write a Word, Find the One Nerve
The single biggest mistake owners make is starting with themselves. Their product, their features, their story. The ad that sells starts with the customer, and specifically with the one thing that is already bothering them before they ever see you.
So before you write, you research, and the research is not about your product at all. It is about finding the exact ache your customer is walking around with, in the exact words they use to describe it. You do not have to guess at those words. Your customers have already written them down for you, in a dozen places. Read the reviews on your own products and your competitors'. Read the Reddit threads and the Facebook groups where your kind of buyer complains. Read your own customer-service emails, the ones where someone explains, in frustrated detail, what they were struggling with. That is your script's raw material, and it is free.
As you gather it, you will notice something. All the different complaints tend to trace back to one root problem. A hundred symptoms, one nerve. Your job is to find that nerve and build the entire ad around it, not around the twelve features you are proud of. Pick the one problem your customer feels most, say it back to them more clearly than they could say it themselves, and you have their attention before you have shown them a thing.
Every Ad That Sells Has the Same Five Beats
Once you know the nerve, you need a structure to hit it. Here is the one to memorize, because it works across almost every product and price point: Hook, Villain, Hero, Proof, Ask.
Grab attention. Name the enemy. Introduce your product as the thing that beats it. Prove the claim. Tell them what to do next. That is the whole spine. Everything else is detail hung on those five bones. Walk through them in order and you have a script; skip one and you usually have an ad that limps. Let's take them one at a time.
Beat One, the Hook: Earn the Stop
The first two or three seconds are a separate battle you have to win before any of the rest matters. If the opening does not stop the scroll, nobody sees your villain, your hero, or your offer. They are already gone.
A good hook works on two channels at once: what people see and what they hear or read. Something visually a little unexpected, paired with a line that opens a loop their brain wants closed. Curiosity is the most reliable lever there is. Tell someone not to picture a pink elephant and they picture a pink elephant. A hook that makes a viewer think "wait, what is this?" buys you the next few seconds.
But there is a trap here, and it is the one that fools people with good numbers. You can write a wild hook that stops everyone and has nothing to do with your product, and your hook rate will look fantastic while your sales stay flat. That is the tell: if lots of people stop and almost nobody sticks around, your hook was a cheap trick, not a relevant one. The stop only counts if it sets up the thing you are actually selling. (This beat is the whole game in miniature, and we have written a piece just on it: Your Ad Has Three Seconds. Most Waste All Three.)
Beat Two: Give the Problem a Face
Now you have their attention. The instinct is to spend it introducing your product. Don't, not yet. First, name the villain.
Every good story has one, and your ad is a tiny story. The villain is whatever is causing the pain you found in your research: a bad habit, the old way of doing things, a whole category of competitor products that quietly let people down. You give that problem a face, and then you make it a little bigger and worse than the viewer had let themselves admit. Not by lying, but by connecting the dots they had not connected. If you sell something that stops a nagging problem, spell out what that problem is actually costing them, day after day.
There is a specific version of this that punches above its weight: sell against the alternatives they have already tried. Your buyer is almost never seeing this problem for the first time. They have bought other things that promised to fix it and didn't. If you ignore that history and just pitch your thing, they file you under "more of the same" and scroll on. Name the other options, acknowledge why they fell short, and then step in. One caution, because it is easy to get wrong: the goal is to name a real enemy the customer already feels, not to manufacture a fake crisis or scare people with something invented. Cheap fear buys a flinch and costs you trust. Real problems, told honestly, are more than enough.
Beat Three and Four: Be the Hero, Then Prove It
Here is where your product finally arrives, and it arrives as the answer to the villain you just built up. That framing matters. A product introduced cold is a pitch. The same product introduced as the thing that ends a pain the viewer is now feeling is a relief. Same features, completely different reception.
But a claim on its own is just words, and people have been burned by words before. So the moment you make a claim, you back it. This is the proof beat, and it is the one amateurs skip. Show the thing working. A side-by-side test, a demonstration, the product doing on camera exactly what you just said it does. If you cannot demo it directly, borrow trust: a real customer's words, a genuine before-and-after, a number that is specific enough to picture. "Rated 4.8 by twelve thousand customers" lands in a way "loved by customers" never will.
While you are here, handle the one objection you know is coming. Every product has it, the thought that makes people hesitate right at the end. Too expensive, too complicated, "will it work for someone like me." You already know what yours is, because it is sitting in those customer emails. Answer it inside the ad, before they can use it as a reason to leave.
Beat Five: Tell Them Exactly What to Do
You have earned attention, named the enemy, shown the hero, and proved it. Do not fumble the last step by getting vague. The ask can be simple, and simple is better: where to go, what it costs, and the offer if you have a good one. That is it. People will not take an action you did not clearly ask them to take. An ad that builds a beautiful case and then trails off without a clear next step is a sales pitch with no close, and it leaves money on the table every time.
Keep Them Watching Past the Hook
One more thing the five beats assume but do not say out loud: people have to stay through all of it. Stopping the scroll is the start, not the finish. The middle is where a lot of ads quietly lose people.
Two habits fix most of it. First, keep the pace up. Paid social moves faster than any other kind of video; cuts that would feel rushed in a normal video feel right here. Keep it moving and you keep them watching. Second, re-hook along the way. Viewers slip back into scroll-mode partway through even a good ad, so give them a small jolt every so often, a new visual, a turn, a fresh line, to pull them back in. And plan for the fact that every ad wears out. The pros do not shoot one hook and hope; they shoot the same script with ten or twelve different openings, so when the first one tires, they roll out the next and the ad keeps living for months instead of a week.
You Don't Need a Studio to Shoot This
Notice what this whole thing has not required: a production budget. The script is the expensive part, and the script is free. Once you have it, a phone, some decent b-roll, and formatting that looks native to the feed instead of like a commercial will out-perform a glossy studio spot built on a weak script almost every time. There are plenty of ways to make video ads now, some of them very cheap, and it is worth knowing when a still image beats a video entirely, which we have covered in You Don't Need a Video. You Need a Static Built to Win.
The tools will keep getting better and cheaper. AI will write you a script if you ask it to. But it will write an average one, because it is reaching for the average of everything that has been written before, and it does not know your customer's specific nerve. The owners who win with this are the ones who learn the craft first, so that when they do use AI, they can steer it, because you can only guide a tool as well as you understand the thing you are asking it to make.
That understanding, knowing why an ad works and being able to build one on purpose, is the real skill now that everything else has gotten easy. It is the same reason a scrappy ad often beats a polished one, which is a story in itself: Anyone Can Make an Ad Now. Almost Nobody Knows Why One Actually Sells. It is also most of what we do at BrandRocket: we write and direct video ads built to sell, not just to look nice, for owners who would rather run their business than learn to script hooks all evening. If that sounds better than guessing batch after batch, that is the conversation to have. Either way, the next time you make a video ad, do not start with the camera. Start with the page.
