Steal Your Competitor's Ads in 60 Seconds

You can steal your competitor's best-performing ads, decode why they're working, and recreate them with your brand in 60 seconds. Then make 100 more variations while you sleep.

You're about to get the complete 3-part automation I just built for my clients. The actual workflow files. Import them today, generate ads in an hour.

Last week, my team generated 2,847 ads using this exact system.

Total cost: $382
What it would've cost at standard creative rates: $213,525

I've been running growth for brands for almost 20 years. I've never seen a cost collapse like this.

The models that make this possible, Nano Banana Pro and Gemini 3 Pro, dropped 9 days ago. Almost nobody (outside of the tech/AI bubble) is using them yet.

Let’s dig in.

Shoutout to my sponsor The Rundown. Rowan is a really cool founder who blew up a few years ago with The Rundown AI. I’ve been following Rowan’s work for a while now. Definitely give him a sub or follow him on X. Thanks to Rowan and The Rundown AI for sponsoring me.

This is the easiest way for a busy person wanting to learn AI in as little time as possible:

Sign up to start learning.

Here's what's eating your growth team's bandwidth right now:

Meanwhile, the work that actually compounds (the systems, the plays, the leverage) gets whatever scraps of time are left.

Your best strategic minds are pixel-pushing.

And the math makes it worse. If you test 5 ad concepts per week and your competitor tests 50, you lose. No, they're not smarter, but they're feeding the algorithm more data.

Meta and TikTok don't care about your audience targeting anymore. They care about creative. Volume wins. Period.

Here's what creative testing actually costs:

Most brands can't afford to test at the volume the algorithms demand (and I don’t have clients who pay $15k just to test). So they test 5 concepts, pray one works, and wonder why acquisition costs keeps going up.

Your real loss isn't the money. It's the opportunity cost.

Every hour your team spends on creative production tests (I want to highlight the word tests, creative in general is critically important right now and you need a killer human team to manage that) is an hour not spent on the strategic work that compounds. The positioning. The offer architecture. The channel mix. The systems that make everything else work harder.

This automation shouldn’t replace your team. But it does free them to do the work that actually matters.

Your creative director still touches the winners. They refine, improve, add the human touch that makes good creative great. But what they're not doing is generating 200 variations from scratch. They're curating and elevating the best.

That's the unlock.

Until last week, AI-generated media had one fatal flaw: they looked like AI-generated slop.

But that just ended.

Nano Banana Pro is Google's new image model built on Gemini 3 Pro. Three breakthroughs matter:

Try making an ad in Midjourney that says "50% OFF SALE ENDS FRIDAY."

You'll get drunk-looking letters, crooked text, wrong fonts. You'll regenerate 15 times. Still wrong.

Nano Banana Pro gets it right 95% of the time on first generation.

Headlines, body copy, CTAs, all legible, all usable.

That is the difference between “cute AI experiment for Twitter” and “turn it on in Ads Manager today.”

Brand colors do not drift.

You give it hex codes, it matches them exactly. Your brand green is #10B981? It is #10B981 in every variation. Not “close enough green.” Exact.

The model respects physics too:

Your subconscious has spent your entire life learning how light and objects behave. When AI violates those rules, your brain quietly screams “fake.” Nano Banana Pro stops triggering that alarm.

9:16 for stories, 1:1 for feed, 4:5 for Instagram.

It does not just crop. It composes for that format:

You are not getting one image lazily cropped three ways. You are getting three natively designed assets.

Translation: for the first time, AI can generate ads that look like they came from a strong in-house design team, not a prompt toy.

Now let me show you what this looks like with a real brand.

Here’s what we did for a client last Tuesday.

Brand: DTC supplement company (sleep gummies)
Goal: New ad angles for Black Friday (last minute request)
Timeline: 2 hours start to finish

We didn't have time to guess.

Our system pulled competitor ads from Facebook Ad Library (shoutout Scrapecreators.com) and found one from Olly Sleep running 62 days with 18 variations.

If a brand has iterated 18 variations of the same concept for 2-plus months, they are not testing. They are scaling.

We fed that winner to Gemini 3 (the vision model).

It analyzed:

We are not copying their creative. We are stealing the psychology and the structure, then applying it to our brand.

We fed that psychological DNA to Nano Banana Pro with the client's brand rules:

About 30 seconds later, we had 5 on-brand variations:

Each one: correct spelling, exact brand colors, platform-optimized.

The client's team picked #2 and #5. We generated 3 more variations of each, our creative director made refinements, and deployed 6 ads by Wednesday.

Result after this weekend:

Ad #5 hit 2.3% CTR (their average was 0.9%). We killed the others, scaled #5 to 80% of budget.

Total API cost: $0.80
Time from brief to live: A day

The client’s comment:

The system works backward from money.

Most brands do this:

That is why most ads fail.

This system does the opposite:

Three parts.

The Watcher monitors competitor Facebook ads and scores them by effectiveness.

An ad running 45 days with 12 variations? That's a 36.1 score. Money printer.

An ad running 3 days with 1 variation? That's a 2.4. A test. Ignore it.

You are not drowning in every ad they launch. You are only studying proven winners.

The Psychologist analyzes the winners and extracts the underlying strategy.

Every great marketer has a swipe file. This is an automated swipe file that updates itself weekly.

If you want to analyze an ad without the automation this is a pretty good prompt (or use my gem)

The Generator creates your brand’s versions using the winning psychology.

Monday morning:

Total time: 2-3 hours
Total cost: $26 (200 images × $0.13)
Winners found: 3-5 typically

Your team goes from producing creative to directing it. The strategic work gets the bandwidth it deserves.

Monthly API cost for 1,000 ads: ~$150-170

Even if you only use 10% of what the system generates, you're paying $170 for 100 production-ready ads.

And the real ROI isn't the creative cost savings. It's what your team does with the time they get back.

I'm giving you all three workflow files. No catch.

Download the complete 3-part automation:

📦 Workflow 1: Ad Collector & Effectiveness Scorer
📦 Workflow 2: Gemini 3 Pro Psychology Analyzer
📦 Workflow 3: Nano Banana Pro Generator

Download everything here → Google Drive or Github

What's included:

Setup time: 30-60 minutes if you're technical, 2-4 hours if you're new to n8n.

Don't want to touch n8n?

Reply "WRAPPER" if you want me to build a front-end. Upload brand guidelines, paste 2 API keys, add competitor Page IDs, click "Generate." If I get 50+ replies, I'll build it next week.

Want my team to run it for you? (click below)

🎁 Get rewarded for sharing! I’ve scraped and distilled every lesson from the entire AppMafia course (and built a custom AI trained on it all). This means you get the exact frameworks, insights, and viral strategies behind their $36M empire.

Invite your friends to join, and when they sign up, you’ll unlock my App Mafia GPT.

{{rp_personalized_text}}

Copy and paste this link: {{rp_refer_url}}