Everything You Need to Know About Facebook Ads in 2026 transcript

This is how Facebook ads used to work, and this is how it works now. And this is one of the most successful brands actually executing it the way it needs to be done. Here's another brand doing it exactly right. So, in this video, I'm going to show you what changed in Facebook and how you need to adapt your entire approach in 2026. First off, let's start at the top. Old-school Facebook used to work in a way where you would select your audience, maybe interest groups, lookalikes, very classic audience selection.

And then you would create your ads for that audience. Sometimes you'd create the same ads for multiple audiences, sometimes you'd create individual ads for individual audiences. And then at the end of July in 2025, the Andromeda updates came around, and now that we have around 6 months of data to clearly understand what's actually happening, there has been a very clear shift. The shift is essentially that you're not creating those ads for the audience.

You are instead creating concepts that match to audiences. And it actually goes a bit further. See, most brands just think of a concept as a new ad idea, a vibe, something that looks a little bit different. But in reality, a concept is an avatar plus a template that creates the concept, and then that concept is matched to the audience based on the inputs of the avatar and the template. So, in this example here, we can clearly see my white avatar with black template creates a concept.

My white avatar with green template creates a concept. And as I scroll down, my white avatar with orange template creates a concept. My green avatar with black template creates a concept, and so on. So, here's an example of a fantastic brand that I always like to reference called Talentless. This is an example of them stacking concepts that are finding semi-unique audiences. You could look at this on any major brand right now, and you will see themes just like this.

My advice that I give to everyone is to literally screenshot all of your ads, throw them in a board, or even hang them on your wall. Play detective. And what you'll notice here in this example is we have groupings of ads that feel similar, that are often speaking to either the same template or similar avatars. I consider these to be concepts. You could be as rough or as strict as you want on your exact definition of a concept, but it's important to recognize that we are targeting specific avatars within each of these unique concepts.

So, what very likely happens in the Andromeda update is we have very different ranges for what each of these concepts and audiences actually drive in terms of efficiency and scale. Efficiency is a metric like ROAS or cost per acquisition while scale is a metric like spend, revenue, and eventually profit. So, what typically happens in scenarios like this are each of these concepts matching to different unique audiences or semi-unique audiences all drive different efficiency metrics.

For example, this concept might drive a 2x, this one might drive a 3x return on ad spend, this might drive a 4, and this might drive a 7. And what we're going to find is that instead of old-school Facebook mentality, which would be let's put all of our money into the 7x, and let's cut cut cut. That is old school. It does not function like that anymore because you will kill scale. We now see the ability to actually only spend certain amounts on certain audiences and certain concepts.

For example, starting at the bottom, our 7x concept might only be able to spend $100 per day. Our 4x concept might be able to spend $300 per day. Our 3x concept might be able to spend $500 per day. And then our 2x concept potentially could spend something like $1,500 per day. Now, each of these is completely acceptable and very incremental to the business when you stack them up, and as long as your 2x is over your efficiency metric that you're targeting.

Now, you could use ROAS, you could use CPA, or any other KPI that you care about. As long as these ads are profitable, you should not be cutting them from your account. Cuz like I mentioned before, old mentality pushes everything into the 7x and cuts everything else. Today, that crushes your scale and what almost always happens is your ROAS looks really good for two or three days after you cut a bunch of ads and then it tanks because all of these ads up here are building your top of funnel, which is the only way that this 7x ad can actually function.

Now, it obviously does not end there. I have a whole lot on this board and the first thing we're actually going to go through is how do we implement this into a structure that works for today. So, the old approach that we would take specifically on the prospecting side is we would launch a new packs with new creatives and we would take those creatives that are winning and we launch them into a scaling campaign. We'd have duplicates of multiple creatives and we would force spend to the best performing ads.

This no longer functions in the way that we want it to. So, we've now moved to a pack-based system with minimum budgets grouped by avatar and concept together. This is very different and very important to understand. This is the shift that we've made as a result of Andromeda. It took six plus months to lock in the strategy and I'm here to share it with you all. And by the way, if you want first access because I shared this strategy over two months ago inside my school community, then you can go to school.com/facebook.

There's a link down in the description. It is never going to be priced as cheap as it is today and you could literally ask me any question anytime inside the community where I myself will answer and you also have live Q&As. You can jump on the call. I do them every single week. Back into the structure here, the core differences that we have rolled out across all of the Moonlighters clients are really focused on two key pieces.

First, as we can see here, we're rolling out 7-day minimum budgets. What this means is at the ad set level, we are selecting ad set minimum budget for only 7 days. After those 7 days, we're pausing down that ad set minimum budget and we're allowing it to run on its own. So, we're just pulsing in, pushing some spend and then retracting afterwards, either allowing the ad to scale up naturally on its own or decline. The reason we do this is actually quite simple.

Think about it like this. Most of your ads, when you launch them for the very first time, they get very, very little spend, and then eventually they start to pick up, and then they plateau. What we want to do with our ads is something entirely different. We want to short-line the first 7 days, so we actually force some spend, and yes, waste some spend for the first 7 days. And then if the ad is successful, it's going to spike much quicker until it finds its plateau again, just like here.

Or it's going to immediately decline and not be worth it. Now, if the ad is successful, what you are paying for and what you're getting in return is disproportionate. For example, this is where your testing budget is going, marked in red. This is very small compared to your clear upside, marked in green. Everything in green is the cash you are losing as a result of not using minimum budgets. Because what we have seen specifically in Andromeda is less ads get adequate amounts of spend.

So, we're launching new ads, and when we don't set ads at minimums, we're literally seeing them get zero spend or just a few pennies, a few dollars. Now, I love the idea if you've been here for a while, if you've watched my videos for a long time, you know I like CBO, you know that I'm cool with even 100 packs inside of the same campaign. But, in reality, when things start to break, we have to shift our strategy. And this right here is the shift we're doing at the structural campaign level.

Now, we've touched on creative, but right after we lock in our structure, we need to build our creative flywheel. That creative flywheel means we're launching new ads, we're waiting around 7 to 14 days, we're analyzing those ads using incremental attribution. This is important, incremental analysis. We're then taking those winners, and we're developing new concepts based on those winners, and then we're injecting ads from our competitors.

This could be anything that we think is cool that our competitors do. So, back to our creative board, this is some examples of creative we built, just static ads for some of our clients. The objective that we need to focus on now is very different than it used to be. People are telling you things like you only should make new concepts, iterations don't matter. Throw that out of the window, it is not true. Look at any successful brand that is growing, they're increasing the total volume of their ads, period.

What we want to focus on is nailing concepts that work for us. And then once we find a concept that actually works for us, what we're doing is we're taking that ad and we're creating a bunch of iterations. We want to go as deep as possible on successful concepts. It is completely okay to have a lot of iterations of similar concepts. Now, iterations don't look the same as they used to look. Iterations look very different.

For example, here's an iteration of an ad concept. These feel different. One is a little funny, another one's a little bit more functional, but both of these target similar avatars and similar templates. And by the way, if you think that quality of ads doesn't matter and you just need to throw a bunch of ads at the ads manager, you're completely wrong. Quality is most important. Quality concepts will always trump a volume of crappy concepts.

Here's how it works. This is your pixel right here. When you have good concepts targeting the correct avatars being targeted in a CBO system where they're allowed to breathe, what winds up happening is you get more good purchases, good purchases represented by green. When the Facebook pixel sees good purchasers and good purchases, it will then optimize more to the green. Now, green does a couple things. They have higher AOV, higher LTV, you know they're a better customers, less returns, etc.

But when you start to target the wrong people, for example, people that don't actually purchase, don't convert, or if they even do convert, they're buying less of your item, or in a very bad case, you're actually potentially training your pixel for low-quality customers. Then what winds up happening is you start to get a couple yellows. And eventually, you start to get a lot of reds. Reds are one-time purchases, never come back, never interact with your business, high return rates.

Unfortunately, over a 28-day cycle, if your pixel is trained in the wrong way, then the greens start to get overwritten by reds. It is so hard to get out of the red and start to get greens again. Now, if you do think you're in this position, the one way out is high-quality creative. Now, we could do the same thing for new customers versus existing customers. So, let's say you're mixing all your campaigns together, you don't have clear swim lanes, like the swim lanes we always want to operate in here, which are prospecting, retargeting, and retention.

Without these swim lanes, this is basically what's going to happen to your pixel. Your pixel is going to see a mix of new and existing customer purchases. So, let's say there's a few new in here, and then there's a few existing in here. What your pixel is going to realize is it's way easier for it to convert existing customers, of course, cuz they've already purchased from you. And Facebook is agnostic to where the purchase comes from.

So, Facebook is going to literally believe that your existing customers are easier to get. And then what's going to happen is your pixel is going to be completely trained on existing customers. And that means that when you go broad, your pixel is just going to keep finding existing customers or people really close to existing all the way down the funnel. That's how you eliminate the ability to scale. Just to prove this, here's a chart of 10 brands that we worked with at the beginning of 2024, and we continue to work with into 2026.

This is a year plus of data where they previously were not implementing the M4 structure, and then we implemented the structure in full throughout the entirety of the year. First off, the three lines. We have yellow, prospecting, that's new customers. We have blue, engaged customers. And we have red, existing customers. Now, what we noticed for all three of these brands before they started to work with us was that they were spending way too much on engaged and existing.

In fact, at one point, engaged customers almost lapsed prospecting, pure new customers. And if you look at once we took over the management of these accounts, you'll notice that the average prospecting spend increases slightly month by month, excluding Q4, and excluding Q4, we have slight increases in prospecting spend, and we see a massive decline in engaging existing customers. And that's because we're actually able to spend less on engaging existing, push that money to prospecting, make more profit for these businesses by actually spending less money.

And you might also be wondering how do we scale these businesses quickly? How do we shift that cash from engaging existing into new? Or how do we just scale a new prospecting campaign in general? Well, it really comes down to being not afraid of the learning phase. Take this example here. The learning phase is quite simple. I've gone through this before. You learn, you wait some time, your ad's going to active, you increase spend, you go back into learning, and you're flagged that either active, learning, or learning limited most of the time.

What happens with accounts is you scale very, very, very slowly. You are always in this position of tiny increase, wait. Tiny increase, wait. Tiny increase, wait. The reality of what you want to happen, and you probably see many times over and over, is how your competitors are spending so much money so quickly. The idea here is that the learning phase is not something that needs to be respected. See, when you spend more money, you are just adding more people to your impression list.

You're just telling more people about your business that haven't heard of you before. So, people have an average time to buy. It's represented here by this blue and purple line. When you add more spending, represented by these three people here, they're just going to take more time to convert. You most likely just need to be patient. So, for many businesses that we work with, doubling spend is common practice. We don't worry about increasing by 10 or 12 or 15%.

We're very aggressive when we know that revenue and profit is for the taking. Every business's core objective is as much money as possible at the end of the month in the bank account, not arbitrary metrics in the Facebook Ads Manager. We're talking real profit in the bank account. And this is where it gets a little bit advanced. I won't lie to you. This is serious stuff now. At this point, we want to take things a step further.

We call this M3 of our M4 method. M3 is advanced breakdowns. We want to look at everything under the hood. This is where we're looking at day of week analysis and we have breakdowns for all of this stuff. It takes a lot of time. This is fully accessible to every single business and we also provide this tracker inside of our school community. We do age and gender breakdowns so we actually understand where we should be spending our money.

I love the day of week analysis the most because most businesses spend the same amount of money every single day. But as you can see in just this example right here, on Sundays, Fridays, and Saturdays, we spend over 30% more in total ad spend compared to Monday through Thursday where we spend significantly less because our ROAS and cost per acquisitions are worse on those days. You want to spend more on the better performing days.

Why would we give Meta the same amount of money every single day? And we see that directly correlate to actually the same ROI that we get on the Shopify side as well. And then finally, M4, this is the icing on the cake for everything. This is the final destination for most brands. It's cost caps. Cost caps are the bridge from highest volume to controlled scale. There's literally a setting in your account called the cost per result goal or the bid cap goal and it says it on the label.

Meta will aim to spend your entire budget and get the most purchases using the highest volume bid strategy. That's what everyone's on. If keeping the average cost per result around a certain amount is important, enter a cost per result goal. Now, I'm not going to go too deep on this because this is for brands that are only spending $100,000 plus. If you're not at that level, do not think about cost caps yet. Do M1, two, and three and then graduate yourself to actual cost caps.

The key here is you could either give Meta permission to spend everything, aka your daily budget, or tell it only buy when you could actually hit my target. That is everything that you need to do to run Facebook ads successfully, profitably in 2026 and probably well beyond. This has been the most turbulent year and insane shifts have already occurred. So, if you want to truly be ahead of the curve, go to school.com/facebook to join our community or if you want someone to just take care of this for you, go to the moonlighters.com/apply.

All the links are down in the description as always and we'll take care of the ads for you so you could actually work on your business and your product. That is all for today. I hope everyone got a ton of value out of this one. Let me know about questions and comments below and I'll see you soon.