Transcription
Andromeda has changed everything. And most strategies just don't work anymore. Costs are up and profits have tanked. Everyone running ads who figured out what was working has had a massive unfair advantage.
In this video, I'll show you step-by-step the new best Facebook ad strategy and how to set it up so you can stop day trading your ad account and get back to scaling your business. I'm opening up my ad account where I spend my money. I'm going to share my screen real quick and show you the exact structure and more importantly, why we structure it this way.
This is the modern version of the one campaign method which we now call Andromeda one. Ad set one is the control. This is our benchmark, a small set of proven ads and ideally there's a good mix of images and videos. These aren't winners forever, they're just the most stable ads that are good enough to scale from and rely on.
Ad sets two and three are tests. Each one contains a single 322 ad and these two ad sets should be thought of as an AB test. And these ads have one specific job, to improve or even replace the worst ad inside the control ad set. Also, in a little bit I'm going to show you exactly how to identify the worst ad because I think most people pick the wrong one. We're not testing to find the next winner to scale and abuse. Remember, Facebook ads are a team sport. We want to find the worst player on the team and replace them so that we can win more often.
Now let's talk about how to judge a test. This is the part most experts get wrong and this is the most important part of scaling any account. We're not judging the test by any individual ad metric, CPA, ROAS, CTR, hook rate, none of that. Those are diagnostics. The goal is money. We only need to ask one stupid simple question. Did the campaign's profit volume go up? And by that I mean total revenue minus total ad spend. If yes, we just created more money to reinvest so we can scale the budget. If no, the test is a loser. That's it.
Real quick, if you like this video, please go ahead and like this video and subscribe so you don't miss anything else and share it with all of your friends. YouTube loves that and that's how we help put back the fiction so that everyone can see more success and less stress.
Also, check out Disruptor Academy. We just launched a brand new course on Andromeda one. Literally soup to nuts everything you need from setup to scale. Automated rules for low budgets, high budgets, literally everything you need to see more success and less stress. The link is right there. All right, that being said, let's get back to it.
Budget behavior is the signal and the three-step testing process I'm about to show you is where Andromeda one gets really powerful. After we launch the test ads, we wait. And then, step one, delivery. Did the ads earn spend? If the answer is no, the ads aren't competitive and the test is basically dead on arrival and we start over.
Step two, evaluation. What happens to the campaign if the answer is yes? If the test earns spend and the profit volume goes up, the test is winning regardless of what any individual ad metric says. And if the test earns spend and the profit volume drops, it's a loser and we start over.
And then step three, scale. If profit volume went up, we have more money to invest so we increase the budget. And you continue to scale until profit volume stops improving and then you're back to testing. It doesn't matter if any of your new ads are winners or if your ROAS goes from two to 20. If profit didn't go up, it's a loss. Only the business outcome matters because it's not about any individual ad, it's about the whole team. And if we can't scale the budget, we need to make the team better.
Now, a little earlier I promised I would show you exactly how to identify the worst ad and it's also stupid simple. When I evaluate ads, I don't start with ROAS. ROAS is a ratio. It can look amazing while you make less money. We want to make more money. So what we care about is gross profit per transaction. I call it GPT and yes, you can build it as a custom metric in your dashboard. Here's how to do it. Here's why this matters.
Ad A could be a 4x ROAS on a $10 CPA with a $40 AOV. That's a $30 gross profit. Ad B could be a 2x ROAS on a $60 CPA with a $120 AOV. That's a $60 gross profit. Half the ROAS, double the profit. So when we evaluate an ad, the stupid simple question is, does the ad have a GPT above campaign average? If not, that's when we want to try to replace.
This stupid simple structure gives you something most advertisers never get. Certainty. Because we know exactly what changed and when it changed and what it did to the bottom line of the business. We can say things like, "That new hook increased profit volume." Or, "That angle brought in a higher AOV customer." No guessing, no emotional attachment, no subjective view based on attribution.
And here's the hidden upside, the cross-channel effect. When you improve the quality of who you bring in and keep your traffic consistent, everything downstream gets better. Let's take a step back to look at what's really happening. Our creative testing upgrades the customer quality coming into the store. And because all spend runs through one system, Meta's Andromeda gets better at matching the right ad to the right buyer. And because we're not launching hundreds of random ads, all of our customers begin to look and feel and behave the same. Then we scale the budget to buy more of those higher quality customers. And all of that means that our site converts better, search and email pick up more revenue and even returning customer revenue also climbs. And that's the point. We're making the business better by buying better customers. That's why we judge success by the bank account, not just by platform ROAS. I don't really give a about ROAS.
And this is the difference between optimizing ads like a performance marketer and being an operator scaling a business. I think about it like this, hunters versus farmers. Hunters lived in caves for 100,000 years living and dying by their ability to feed themselves today. Hunters measure success by ROAS and daily CPA. They ask, "Did this ad work today? What's my cost right now? Did it hit immediately?" They live and die by short-term swings.
Once humanity learned how to farm, we went from living in caves to building empires and putting a man on the moon. We are in the empire building business. If you want to scale to the moon, you have to start thinking like a farmer. Farmers focus on feeding as many people as possible. One corn plant can produce three or four pounds of food and you can plant 10 other corn plants from what you don't eat. So for the same amount of work, over time you go from four pounds of food to 400 because you're focusing on building a better system and that's what farming is. Farmers focus on total profit and growth. Their job isn't to chase win after win, it's to build a system that works and to protect it and to scale it with the resources that system builds for itself.
And that's what Andromeda one is actually optimizing for, similarity to produce scale. We're increasing the volume of high quality data flowing into the platform so it gets better at matching the right message to the right buyer and finding more of your best customers.
Now, before you copy this structure, we have to talk about budget. Because a lower budget changes the rules just a little bit. If your budget isn't high enough to reliably get more than one ad set out of the learning phase, then everything lives in one ad set. Your control and your tests are just ads inside of the same ad set. Then you follow a simple ladder. One ad set until you can afford to get one testing and one control ad set. And then you scale until you can afford two testing ad sets. And even when I'm spending a million a month like I showed you before, I don't have more than two testing ad sets. Once you get there, you're done. You're ready to scale to the moon.
So now that you understand what Andromeda one really is, it's not a tactic or a hack and it's definitely not a structure you copy and paste from a screenshot. It's a decision to stop being a hunter and to start being a farmer. It's choosing a system you can run every week, not a lucky ad you hope doesn't die.
Now that you understand that, next we need to talk about the ad itself. I got to show you the exact ad format I use, how to build it and how to feed the algorithm the kind of data it needs to actually scale because the best structure in the world won't scale if your ads aren't teaching Meta to win and that is exactly what this next video is all about.
>> [music]