I Turned Claude Into a $880/Day Motion Graphics Skill (No After Effects)

0:00People are making over $500 a day selling motion design like this. And their clients have no idea that it was made with Claude without opening After Effects and without a single key frame. Every animation that you're watching is actually code that Claude wrote. And it takes only one day to make, so you can take on way more clients than any motion designer out there ever could. [music] At the end of this video, I'll show you exactly how much profit I kept from this single project. But ultimately, is it actually as good as a human pro? To find out, I hired a real motion designer and gave them the exact same brief, files,

0:37and budget to see just how big the gap between us really is. Now, he asked for 5 days and Claude got one. And only one of us actually made the deadline. So [snorts] in this video, I'm going to show you everything. How I found a client, the exact message that got me hired, how Claude built the video, and ultimately how I got paid. So without

0:57further ado, let's get into it. So step one is to find a client. Now, there's a lot of motion design styles out there and not every client is willing to pay well. So it's important that if you want to earn more, you need to do the style that pays the most. Now, today that style is SaaS or tech launch motion videos. For just 30 seconds of this kind of motion design, clients pay

1:23anywhere from 1,500 to $10,000. And finding clients here is a lot easier than my niche scene. You can do it with no experience and an empty portfolio. So here's exactly how to get it done. I go to Upwork and type SaaS launch video into the search. And as you can see, there's a ton of jobs here at all kinds

1:44of budgets. And here's the first tip. Make sure to use different searches to get more jobs to choose from. You always want to try things like product launch video, UI animation, app promo video, and SaaS explainer. But, the most important thing is picking the right jobs out of all of these. Now, most people would scroll and literally apply to everything. That's exactly how you burn your connects and ultimately get

2:09zero replies, which means zero money. So, here's three things that I always check before I apply. First, you want the client to say payment verified, 10K plus spent, and a higher rate that's above 50%. That means that ultimately they actually hire and they actually pay. So, you're not wasting connects on somebody who's going to ghost you. Second, freshness. Now, you want to sort by newest and always keep refreshing. I go for posts that are less than 30 minutes old, which means that the client just posted the job, so they're probably

2:41still online watching proposals come in. [music] You want to apply in that window and you'll end up being one of the first people that they see. Third, the competition. You want to scroll down to activity on this job. There you can actually see how many people already applied and how many of the client is already interviewing. Fewer than 15 proposals and nobody interviewed yet means that they haven't picked anybody and your message is actually going to get read. Fourth, the money. You want to open their website and check whether they actually just raised a round. A freshly funded startup about to launch has both the budget and the deadline, so

3:18you know they're going to be paying well and fast. Now, if all four of these things check out, [music] then apply. If not, skip it and move on. Now, another insider tip, and yes, it does break my first rule, you're going to see jobs for $10 or $5 an hour from clients with zero hires. Most freelancers are going to skip them instantly, but a lot of the time that's just a brand new client who doesn't know what to put in the budget field yet. So, make sure to check out their website. If the company is real, then maybe it's worth applying anyways and quoting your real price. There's

3:51almost no competition and it's a client who's actively looking. And here's the most hardcore tip of all. Always add cover letter and use the exact keywords from their post. Now, if they wrote product launch video and Figma, then those words need to be right in your first lines. That way the client instantly sees you actually read the post. My structure here is always the same. Their task, my solution, and a clear deadline. So, that way the client knows exactly what they're getting and when they're getting [music] it. I'm going to show you the exact template in just a second. Now, quick pause. If this is useful, please make sure to

4:29subscribe. This channel is all about making real money with AI tools. The workflows, the prompts, and the experiments just like this one. And for everything first, follow me on Instagram. That's where the newest stuff drops before YouTube. Okay, let's get back to it. Now, you found the job, so let's get hired. Okay, I've picked five jobs that look promising. Now, I'm going to send each of them a message just like this. Feel

4:59free to pause the video and read it. This message is written so that the client picks you even if you don't have a portfolio. Now, why like this? Well, first, it's because the opening line is actually about them. On Upwork, the client only sees the first two lines in the preview. And almost every single proposal out there starts with something like, "Hi, I'm a motion designer with X years of experience." And honestly, that's boring. I'm pretty sure that none of you would click on a message like that. Now, the next reason is that the concept replaces the portfolio. A portfolio shows what you did for somebody else. A ready-made [music]

5:36concept shows what you'll do for that. The client can already picture their video and you're the only one who showed it to them. And most importantly, the small first milestone. Because of it, the client takes almost no risk. With a message like this, clients will start choosing you over your competitors. Because literally a couple hours later, three out of the five clients replied,

6:00already ready to start working with me. But this project looks the most interesting to me. It's a startup building an AI-powered video recording tool, [music] and they need a launch video before they go live. Now, they already have a rough vision that matches mine, and their UI is already done in Figma. So, I quoted one day and a thousand dollars for a 30-second motion video. Now, that's actually not that much, since you remembered how much videos like this go for on the market. We messaged back and forth a bit more, and you can actually pause the video and read it, and that's when they ultimately agreed. Now, we're

6:34ready to start generating. But before we generate anything, [music] remember the challenge from the start of the video. So, we need to find a motion designer. Because I want to know whether AI can actually replace a pro. So, a real motion designer is going to get the exact same task, same brief, same Figma files. Now, I'm going to go to Fiverr

6:59and search for a SaaS motion designer. And I'll skip everyone who doesn't have UI work in their portfolio. Character animation and UI motion are actually completely different skills. I found this guy Alex. He has 47 reviews and clean UI work, and he quoted a thousand dollars for 30 seconds in five days. So, I sent him the brief. And a few minutes later, he replies. [music] Okay, so the deal is done. The pro gets five days and a thousand dollars. Claude

7:30gets one day and the exact same brief. Now, let's see what Claude can actually do with this. Now, if you think you can just drop a prompt into Claude and get a masterpiece out of it right away, then you're dead wrong. It never works on the first try, but once you've set up the process, you can actually turn it into a custom skill. Which is to say, it'll become a permanent instruction set that Claude will follow every single time for this kind of work. After that, then you can generate everything you need from a single prompt. And by the way, I've attached my entire motion design skill

8:07in the blog post link below so that you can download it and use it on all of your own projects. But what's actually inside of it? Basically, it's everything that a senior motion designer carries around in their head. First, the story. Every launch video that I make follows the same exact path. Grab attention, show the problem, show how the product fixes it, and finish with the payoff. Now, if a scene isn't serving one of those four needs, then it's out. Second, timing. Anything important stays on screen for about a second and a half. It's long enough to read, but short enough that nobody's

8:45getting bored. Third, the motion rules. And this is actually where most of the magic happens. Nothing ever moves linearly. Things need to come in over about half a second with a slight overshoot, and they leave faster than they came in. Every click gets a tiny press, and the cursor never moves in a straight line. It moves like an arc as a real hand would. But the golden rule of smooth animation is this. Nothing should ever fully stop. In the skill, I tell Claude to work the way that a motion designer works in After Effects. Group related elements under a parent, basically a null object, and move the

9:25whole group instead of every piece separately. Then, you want to offset the keyframes, so each element starts a couple of frames after the one before it. And start the next move before the last one has ever fully settled. When the keys overlap like that, then every movement flows into the next one. Nothing freezes, nothing snaps, and it all feels like one continuous, coherent motion. That's the difference between looks like a website and looks like a real launch video. And Claude does all of this in code, so you never have to

9:59set a single keyframe yourself. [music] And fourth, the tech specs. 1920 pixels wide, 1080 pixels high, 60 frames per second. Plus safe zones, so when you adapt the video to vertical, everything important stays inside the frame and nothing gets cut off. Now, right now, I'm just dropping it into the chat, but before we hit generate, there's two more strict rules that you always need to keep in mind. Rule number one,

10:26storyboard first, then animation. Without a good storyboard, your output is going to be absolute garbage. Now, in a bit, I'll show you the same exact video generated with and without a storyboard, so you can see the difference in quality yourself. Rule number two, if it's motion, you have to teach it. You have to feed your UI components, define your easing curves, and give it a solid baseline. And this next part is important. Before this project, I spent more than 10 hours testing different setups, and this is actually the one that worked for me, so it'll work for you, too. So, here's everything that I gave Claude. The brief

11:04from the client, the storyboard scene by scene, breaking out what's on screen, how long it stays, and where the camera goes. The brand colors and fonts, the background images, and the UI straight from Figma, exported as SVG, not screenshots. Now, the part that matters, when the camera zooms into a button, an SVG stays razor sharp, and a screenshot turns into a blurry mess. Then, one prompt. Using the motion skill, build this storyboard as a 30-second launch video scene by scene. Claude starts writing. Every scene becomes its own block of code. The tasks pile up, the copy-paste grind, the agents kicking in, the payoff, and the logo. Then, the

11:47whole thing renders frame by frame into a video. So, I ended up dropping it all into Claude Opus 5.5, and right away, it broke. Just look at this. The text is overlapping, the camera pan is way too fast. Also, the whole layout just explodes off the canvas, the font at some points fell back to Times New Roman, the task counter jumped to 200 in a single frame, and all of the agents' cards pop in at once attached on top of each other. It's total chaos. Now, this is where most people quit and say, "AI motion doesn't work." But, almost every single one of these bugs has a simple

12:24cause. And once you know that cause, then the fix is just a single second. The layout exploded because the Figma screens were designed 1440 pixels wide, and the video we asked for is 1920. So, I need to tell Claude, "Lock the canvas to 1920 by 1080, and position everything relative to the center, not the edges." The font, on the other hand, broke because it wasn't loaded before the first frame. I just need to say, "Load the font before rendering anything." The counter, "Make the count up last about a second. Long enough to read, but quick

12:58enough that it doesn't drag." [music] The agent cards, bring them in one by one a few frames apart, and let each card land before its check mark appears. The camera, slow down the pan and give it an easy ease, so it starts and stops softly. I spent the next hour going back and forth just like this, debugging it line by line. The first few versions

13:22were all about fixing what's broken. After that, it became about taste. The task counter needed a soft red glow as it climbs. The cursor needed to hesitate between tabs for a second as if you're losing track of what you were copying. Each of those is one sentence to Claude and just a few seconds of waiting. And every one of those fixes is now actually baked into the skill in the blog post below. So, when you use it, you skip all

13:50of this pain of the first hour. And then it clicked. Just look at that. It's a buttery smooth 60 FPS animation rendered straight from code. No timeline stretching, no missing plugins, and I didn't set a single keyframe by hand. Now, do you remember our first rule? Here's what happens without a storyboard. Same assets, same skill, same prompt. [music] Simply no storyboard. Claude will start

14:18guessing. [music] Scenes come in in a totally random order. Everything moves at the same time and there's just no coherent story. It feels like just a bunch of nice-looking UI flying around the screen. With a storyboard, it's problem, pain, fix, payoff. Now, the same tool but a completely different output. And here's why this whole thing takes just a single day. Say the client comes back and asks to change the task counter from 200 to

14:48999. In After Effects, that means opening the project, finding the layer, fixing the keyframes, and waiting for a re-render. Here, that's just one line of code. I patch it up in seconds and the new version is ready in just a couple of minutes. So, the total time is under a day. Total cost is a $20 Claude subscription. Meanwhile, Alex was still

15:13working on his style frames. [music] Before we watch the full generation, I checked in with the motion designer and they said that they'd make the deadline. So, I'll be back in a couple of days to show you both results. Now, in the meantime, I sent what we've made to the client and here's what they replied. The same day I sent the final files, they approved them and the payment got

15:39released. A full 7 days have passed and we're back. The main reason for the delay is that the motion designer missed their deadline. On the day when they were supposed to send the final video, they sent me this message. And the next day, this one. But, today they finally delivered all of the videos, so we're

16:01ready to watch both versions. All right, [music] let's look at Alex's video first.

16:29Honestly, it's solid. It hits the brief, the UI looks clean, and the animation is quite smooth. If I were the client, then I'd be totally fine with this. But, there's two catches. We waited seven full days for it because they kept pushing the deadline and it cost a thousand dollars. Now, here's our version built entirely in Cloudopus 5.5

16:50using code.

17:13Look at the camera moves, the photo real background behind the UI, and how every click has weight. So, which one do you think turned out better? Ours or Alex's? Let me know in the comments below.

17:30Okay, let's talk money because that's probably why you're still watching. Now, the client paid me $1,000. Claude cost me 20 bucks for the subscription and I spent less than a day on it. And most of that day actually went into the storyboard and not the actual generation. After Upwork's fees and the subscription, I kept $880 for just a single day of work. And here's a little hack that I didn't use this time, but you should be aware. If the client is solid and you explain it the right way, then you can ask them to actually cover the Upwork fee. Just be up front with them. My price is $1,000

18:06and Upwork's fee goes on top. A client who's happy with your work will usually say yes without thinking twice. That's another hundred dollars straight into your pocket. Now, flip it. Imagine that you're the client. Your launch is in 2 weeks and you've got two options. Option one is Alex. Nothing against this guy. His video is obviously good, but if you're looking at a full week and you're waiting on your prize, going back and forth on revisions, and hoping that it lands before your launch date. Option two is somebody like [music] me. It's the same level of quality but delivered the very next day. And the revisions

18:45take 1 hour, not days. Who are you hiring? Exactly. And here's the part that most people miss. A SaaS company doesn't need just one video. They need a launch video, then a feature update, then an ad, and then the following launch. So, if I'm the guy who can deliver fast, whose content looks good, and I never miss a deadline, then I'm not just getting one $1,000 job, I'm becoming their go-to motion guy. [music] And one client like that can quickly become an anchor source of income. And remember, don't treat this as a magic button. The storyboard and the skill working hand-in-hand are what make this work,

19:23and they're both in the blog post linked below. So, here's the bottom line. Clients aren't buying this because they care about code or they care about keyframes. They buy it because they get a launch-ready video by the very next day. Everything that I used for this, the motion skill, the storyboard layout, and the exact pitch messaging that landed this job is in the blog post linked below. Go grab it, run a quick test on your own, and let me know how it goes. As always, I'll see you guys in

19:54the next one.