5 Steps to a Video Ad That Actually Converts

In 2021 four versions of the same ad ran side by side with nothing changed except the first five seconds. One of those hooks produced 62% of the sales. One produced zero. A video ad that converts is five parts in a fixed order, and the first five seconds decide most of what you get.

The five parts, in the order they have to come

We've spent over $1.5 million on ads across the businesses behind this software, and the person who publishes this blog has sold or helped sell $25 million of online courses doing it. Every ad we've written that made money has these five pieces, in this order, because each part earns the right to the next one.

Part	Its one job	How long it gets
1. Hook	Make them feel a pain they already suspected they had	The first 3 to 5 seconds
2. Problem	Grow that pain with specifics, not adjectives	Most of the ad
3. Solution	Show the vehicle out of the pain	Late, and briefly
4. Objection	Answer the one doubt that stops the click	One or two lines
5. Call to action	Give one clear thing to do, with a reason to do it now	The last few seconds

The ad in the video above pointed at a free one hour workshop on selling courses. That training grew into Course Creator Pro, and it now comes with every Course Creator 360 account, so the teaching and the software that runs it sit in one login.

Step one: the first five seconds, where the money actually gets decided

If they don't make it past second five, nothing else you did matters. So in 2021 we ran that four hook test: same ad body, four different openings, side by side. In the first nine days hook D brought five sales, hook B brought two, hook A brought one, and hook C brought none. Sixty two percent of the sales came from one opening.

Here's the part worth sitting with. Hook C, the one that sold nothing, was the hook that had been scripted and planned to run by itself. The other three almost never got filmed, and the conclusion would have been that the ad didn't work, when what didn't work was five seconds of it.

The platform sets the rules for those seconds. On Facebook your ad starts on mute, so it has to stop a thumb before a word is heard. On YouTube you get about five unmuted seconds before the skip button, so the audio does the work. Same script, two different openings. Here's which of the two to start on if you only have budget for one.

Five hook types show up in ads that sell: painful or validating, value foreshadow, funny, pattern interrupt, and controversial. Almost everyone assumes theirs has to be funny. Comedy is the hardest to land, and the two that have made us the most money are the first two on that list.

Step two: make the problem specific enough to hurt

The best hook we've ever written opened by saying here's what professional filmmakers said were the biggest mistakes they made when they started, then listed ten of them. A mistake is painful, and everybody suspects they're making a few right now. Each of those ten confirmed a suspicion the viewer already had, so the pain grew for a minute and a half before anything got sold. That ad did close to a million dollars in sales for the course business we ran marketing for.

Blair Warren wrote a line in The One Sentence Persuasion Course that is worth coming back to constantly. People will do anything for those who encourage their dreams, justify their failures, allay their fears, confirm their suspicions, and help them throw rocks at their enemies. That's the job description for this part of the ad. An earlier ad did it with one line: you just bought a new camera, and now you're spending all day watching tutorials trying to figure out how to use it.

The mistake we see most is a problem that's too broad. One of our students ran an ad whose problem was that people don't know how to live their dreams. Everybody is trying to live their dreams, so that ad was aimed at everybody, and if you're appealing to everybody you're appealing to no one. The more specific the problem, the more unique your solution gets to be.

Specific problems come out of your customer's mouth, not your head. Before you write a hook, go get the actual words: sales calls, support messages, the questions people email in. If you don't have those yet, the free customer avatar template we built asks the same questions we do and fills in the pain lines you'd otherwise guess at.

Step three: hold your solution until the pain has landed

The solution is the vehicle. Somebody in pain wants out, and your product is the car. The beginner version is pitching from second one, and by the end the viewer feels like their ears got talked off.

In that ten mistakes ad the solution doesn't show up until one minute and twenty two seconds in. Ninety percent of it was pain, and the paid course never got mentioned once. Listing clients and showing off would have sold less.

People aren't dumb. They know they're watching an ad, so don't hide it. What you're earning is the ten seconds where you get to say what the thing is, and whatever you point them at has to exist and be good. Build the offer page before you write the ad, never after. If you don't have a page yet, our free AI website builder will generate a real one from a description of your business.

Step four: answer one objection, not fifteen

Every person who watches your ad has at least one doubt. Advertisers hear that and stuff six objection handlers in, which adds a minute and loses retention, so nobody reaches the call to action. One or two big objections belong in the ad. The other hundred belong on the landing page or in the emails that follow.

Objections also come almost last in the order of operations. Someone has to get problem aware, then solution aware, before they have a doubt worth answering. Answer a question they haven't thought to ask and you just planted it.

The smoothest way to handle one is inside the solution instead of as its own section. One of our ads listed weddings, luxury real estate, commercials, travel films and YouTube content in a single breath, because people kept emailing to ask whether it applied to their work. Nobody asked again.

Credibility works the same way. We've run ads that made millions without a name on screen once, because the footage was the proof. Testimonials do that job too. Showing beats saying.

The objections you leave out still need somewhere to go. That's the follow up sequence, and it's the part beginners skip. In Course Creator 360 the emails come out of the same account as the funnel, deliverability and DNS get set up for you on your onboarding call, and sending is unlimited on Premium and Elite, so nobody has to wire an ad tool to an email tool with a third program in the middle.

Step five: one call to action, and a reason to click it now

Your call to action does two things: it creates a cliffhanger, and it is lucidly clear. The cliffhanger is why you binge Netflix. Every episode ends by hinting at what you'd miss, and that thought beats the desire to sleep. V-Shred does it in their ads: one food that raises your testosterone 52%, then, near the end, four common foods wrecking it, and you only find out which by clicking. Value promised, value delivered, biggest piece behind the click.

Clarity is the other half, and friction is the enemy. The cardinal sin is more than one call to action. Listen to the average YouTube outro: like, subscribe, hit the bell, check the merch, follow on Facebook, here's my Instagram, click the next video. Seven asks in ten seconds. People have a hard enough time making their bed. In an ad you're paying for, every extra ask is money burned.

Then the click has to keep the promise. If the ad says click below for a free one hour training, there had better be one behind it, with a big obvious button and nothing in the way. No pop up, no forced intro video, no scrolling past three promos. Here's what has to exist behind the click for an ad to make money instead of traffic.

What changed in five years, and what didn't

In 2026 we spent about $10,000 testing UGC creators for Course Creator 360 ads. Seven creators, roughly $2,000 for the raw footage, same brief for each: one ad body, five hook variations. Most sent back ten or more. Five years after that split test, we're still buying hooks by the handful and ad bodies one at a time.

What changed is the packaging. Cinematic ads ran next to a creator with a phone in a kitchen, and the phone held its own. The anatomy didn't change. We judge every ad on hook rate, then hold rate, then cost per click, then cost per lead, then cost per trial, because that's the order the five parts fail in. If the hook rate is bad, nothing downstream is diagnosable. Most of the money we've watched people waste, and plenty of the million we put through Facebook, went to ads whose problem was five seconds long.

If you've never run a video ad, start here

Write the call to action first, because everything upstream exists to make that one thing feel urgent. Then write one ad body of forty five to sixty seconds using parts two through four. Then write five hooks for it, not one, and be suspicious of your favorite, because the favorite in that four hook test sold nothing. Film them in the same session and run them against each other on a small budget until one pulls away.

The part that stops most people isn't the script. An ad needs a page to land on, a checkout that works, an email sequence that follows up and a place for the contact to live, and stitching four subscriptions together to get that is where the project dies. That's the whole reason Course Creator 360 exists: the funnel templates, the checkout, the emails, the CRM and the course sit in one account, and the templates came out of ads that actually ran. Take the 30 day trial, build the page your first ad points at, then go write five hooks.

Frequently asked questions

How long should a video ad be?

Long enough for the pain to land and no longer. Most of ours run forty five seconds to two minutes, and the ten mistakes ad spent its first eighty two seconds on the problem. Short ads fail when they skip from hook straight to pitch.

What makes a good hook for a video ad?

Pain the viewer already suspected, in their words, in under five seconds. The two that work best are the painful or validating hook, which names a mistake they're afraid they're making, and the value foreshadow hook, which promises something useful and then delivers it inside the ad. Funny is the hardest to pull off and the most attempted.

How many hooks should I test?

Five per ad body, minimum. One hook out of four produced 62% of the sales in that split test, and the one scripted to run alone produced none. Change only the first five seconds between versions so you know what caused the difference.

Should I run ads on Facebook or YouTube first?

Start where your first five seconds are strongest. Facebook plays muted, so the opening has to be visual. YouTube gives you about five seconds of sound before the skip button, so the opening has to be verbal. We started on Facebook and moved to YouTube two years in, and doing the research before the move is why it worked. You don't need a crew either: in our 2026 test a creator with a phone competed fine against cinematic footage.

What do I need built before I run my first ad?

A page that keeps the promise your call to action makes, a way to take the email or the payment, and a follow up sequence for every objection you left out of the ad. That's a weekend inside Course Creator 360 with the templates, and if you'd rather see it built before you commit, book a demo and we'll walk your funnel end to end.

Related

• Facebook Ads vs YouTube Ads for Course Creators: Which Wins?
• What $1 Million in Facebook Ads Taught Us About Selling Courses
• How to Scale Ads to Seven Figures: My 7-Step Loop