00:00Last week, my marketing company shipped over 100 meta ad creatives across 10 different clients.
00:05We spent under two hours actually making them, and a year ago, this would have taken us
00:0920 hours or more. I'm going to show you the full stack, Claude, Nano Banana, and a couple of custom
00:14skills I built in VS Code, so you'll have everything you need to run it yourself by the
00:19end of this video. Let's get right into it. Meta ads today need volume to have success,
00:23and 20 creatives or more every single week is the new floor. It's no longer the ceiling.
00:27We spend over six figures every single month on meta ads for our clients, and this is the system
00:32that we actually use to generate winning creatives, not a theoretical one. Most AI creative tutorials
00:38or even different AI creative platforms or tools are really just basic. This is a super realistic
00:45workflow that can be used at scale, which is why it'll scale past 10 creatives in one flow.
00:50So just to break down exactly what this looks like, what we'll do is we have this four-step
00:55process. First, we feed client context into the model. We do this inside of VS Code with Claude
01:02with some custom skills that we've made. You can check those out by checking the link in the
01:06description, and I'll break those down for you here in a moment. From there, it actually generates
01:10the 20 angles first, not the creatives. We get the hook, the body, and the call to action. We
01:16separate it in this way with a few different styles as well, and I'll show you what that looks
01:20like. From there, we have Nano Banana render the static and background assets, and from there,
01:26if everything is good to go, we're good. But if we want to take it a step further, we'll use
01:30another
01:30app called Pleator, which actually takes those assets and takes these angles, takes them a step
01:36further, and can even generate them with other AI models, because then you end up having something
01:41like this, which is really cool, where we'll have new inputs here that we generated with Claude and
01:46VS Code. I have a new prompt that Gemini 3.1 will run through, and then from there, I can
01:52run it
01:53through as many different AI models as I want with a few different variations per image. So this one
01:59workflow here generated me 4, 5, 6, 7, 8, 9-ish, 10, 12, 13 creatives, give or take, and that's
02:11not even
02:11counting the ones that I did originally inside of Nano Banana. So I want to show you the system from
02:16start
02:16to finish. Let's get right into it. Okay, so I have everything set up inside of VS Code using Claude
02:23Code, and what this allows for is some really good file structure, especially if you have multiple
02:27offers or multiple clients that you're running this for. What I'm going to do is I'm just going to spin
02:32off a prompt down here, asking it to make me some new ad creative briefs, and then while it does
02:37that,
02:38let me show you what's actually inside these skill files. So these skills are pretty basic, but they're
02:43detailed. I'll basically have a few different options here. I have the primary skill here, which
02:48is this image prompt engine, and this pipeline, I make it very clear that, hey, you're not actually
02:53making the image, you're making the ad concept with visual direction for future AI to expand more so
02:59downstream. It can pull different ad copy briefs from different clients. It can check clients' brand
03:05guides if it has any. It can read something called a hook bank, which I think is a super powerful
03:10asset
03:10every single company should have. If you don't have like a Google Sheet or a Google Doc with a ton
03:16of
03:16hooks, just a ton of super simple one-liners as the call-out or as just the hook, you're missing
03:23out
03:23because you can be testing those hooks, iterating on them, and then actually tracking what's working.
03:28Then we have this ad structure guide that I put together where basically I tell that, hey, for a
03:32winning image ad, they usually consist of a call-out, a hook or the meet, and then some kind of
03:37CTA. So follow
03:38this structure every single time. I've got different hook formulas that I put in here. It's like how to
03:43achieve outcome without this hated thing. If you're this avatar doing this behavior, read this.
03:50You want this result in this time frame, so on and so forth. Then I actually gave it these four
03:55different hook sources based off of how I can actually do the copy, right? So pain points, dream
04:01outcomes, unique mechanisms, and points are in proofs. And then if the hook bank does have winners,
04:06riff off of those angles. So it's all about just building these skill files in a way where if you
04:11reverse engineer how you make winning ads or what the winning ads that you have, how they're
04:15structured, you can make a super easy skill file like this. Like I said, you can also click the link
04:20below and you'll get access to these, but I think it's good to make something that's based off of your
04:25winners. The last two skill files that I'll make is the actual angle to visual pairing. So I basically
04:33gave it instructions here that says, hey, once we have the add angle and all that kind of stuff,
04:38I want you to put it together in like a good looking visual approach so it actually looks okay.
04:43So if there's a pain, put before state imagery. If there's an outcome, show confidence, space control,
04:50achievement, clean tones, et cetera, right? There's comparison, split screen, side by side,
04:55all that kind of stuff. I'm just describing based off the angle, what should the visual kind of look
05:00like based off of that. Now what's really cool as well is I actually hooked all this up to work
05:04inside of Slack, which is really cool. So what I actually ended up doing is I made a bot in
05:11Slack
05:12where I'll put the client name, I'll tell it what I'm looking for, and then it'll make all these images
05:17based off of that same engine. So I technically don't even have to leave Slack. It's not showing
05:22the images here, but I can actually open up a Google Drive where I have everything housed and it'll
05:28actually store creatives for my clients or for my offers and it'll drop them right in here. Some of
05:33these crush it and do very, very well. This is a dating offer that we're running and it's doing
05:39great. So that's essentially that high level workflow inside of VS Code. Now let me show you
05:45what it looks like once we get that end result and then how to use it. Okay, so I gave
05:50some details
05:51of an offer and now my VS Code bot is basically giving me some prompts here. So what I'm going
05:58to
05:58do is I'm just going to pull a few of these. I'll copy this one and now we have a
06:03few options. We can
06:04just go directly into Nano Banana and we can actually have it generate these images because you can see
06:09like a really strong prompt here that it has. We'll send that through. It'll do its thing. The other
06:15option is we can actually jump into Pleator which just kind of maps this out visually and then we can
06:21actually map some of this out here. You could essentially do this same thing just directly in
06:26Nano Banana or even in VS Code and kind of build it out and string it together with Nano Banana.
06:33But Pleator
06:33is nice because you can visually see it. Then I've got a separate prompt in here that basically tells
06:38it how to handle the inputs that I'm giving it. And what's really cool is you could also give visual
06:42references too which is nice if you're trying to test a variation of a winner or something like
06:48that. So let's see this one's thinking here. Cool. This one's not bad. Super easy creative. Looks pretty
06:54clean and I'm sure this will work well for this dating offer that I'm talking about. And you could
06:59essentially run this whole thing on a loop once you have more than enough creatives to get you these
07:0420 angles. Like I said if you're doing it in Pleator or something similar or even if you just stitch
07:08this
07:09up inside a cloud code through Slack you could generate a ton of angles. Like this Slack workflow
07:14that I showed you every time I prompted it and gave it all the information it needs it would generate
07:20I think 10 or 5 concepts and then 8 images. So 4 concepts, 2 versions per concept. And it would
07:30work
07:31really really great. We'd get all those feedbacks and all those creatives and it would crush it. And this is
07:35a really easy way to just take the idea and the concept and turn it into a strong ad creative
07:42and just test. Meta is so big on testing nowadays you have to be pumping out volume and even if
07:48the
07:48ad creatives are not perfect they're going to help you guide your direction on what you should be making
07:52next to help you get the best results in your paid ads. So if you like this video let me
07:57know down
07:57below. Let me know if there's anything else I can cover to make this better. I'm not selling a course
08:01or anything on this but I do have a community that gives you more resources to make this easier for
08:06you to follow. So if you're interested click the first link in the description, join us in there,
08:10start making more money from your paid ads and getting more success all while working less.
08:15I'll see you in the next one.
Comments