Meta's learning phase requires 50 optimization events per week to exit. If your ad set is stuck, your budget is likely too low (a $20 CPA needs $1,000/week), you have too many fragmented ad sets, or you're editing too often. Consolidate campaigns, use higher-funnel events like add-to-cart, and avoid major changes for 3-7 days.
If you have ads stuck in the dreaded learning limited phase inside of your Facebook ad account, I wanna show you exactly how to fix it in this video. It's a very big thing that's misunderstood by people that run paid ads, whether you're doing it full-time, you're working with a lot of people, or you're doing it yourself. And I'm gonna show you exactly how to get out of the learning phase fast inside of your ad account. And first off, this is pretty much what happens with people is they'll launch their ads Everything looks great. The offer is solid and they'll run it for a few days and then learning shows up or learning limited shows up with that yellow dot and people freak out about it. Costs will swing left and right day over day, and it doesn't feel like things are improving and people really don't know what to do. So to set the stage, the learning phase is just Meta figuring out who to show your ads to. Every new ad set will start in the learning phase with the green dot where it's just like pulsating and it says learning. And the algorithm is testing a few things. First, it's testing who to target, when to show the ads, and then which placements actually work best. The easiest way to actually get out of the learning phase is to get 50 optimization points per week to exit learning and to actually stabilize. And I'm gonna show you exactly how to do this, 'cause some people think 50 events equals spending more, and sometimes it does, but sometimes it makes— what really needs to happen is just having the right setup. And patience. So let me show you exactly how to do that. First things first is learning limited, your ads stuck, and here's how to kind of get out of it. So this is exactly what's happening. You've got low budget, fewer events, bad results, less spend, and then it just keeps going in this vicious cycle. And you need to break the cycle, and here's how you do it. Most people are staying stuck because of 3 mistakes that they're making. First, their budget's too low to actually generate 50 events a week. Second, they have too many ad sets and their structure is not set up correctly. And 3, they're constantly editing, and every significant change basically resets everything and puts the learning phase back to zero. At the end of the day, your budget needs to match the 50 event target, and you just have to do the math to figure out what that is for you. If you're getting cost per conversions of around $20 and you need to get 50 of them a week, you need to be spending around $1,000 a week, which is $143 a day. And a lot of people are setting budgets based on comfort, not math. And that's the problem. And if you can't afford, you're seeing this, you're like, oh shoot, $20 CPA is amazing. My CPA is even higher. I can't even afford to spend $50 a day. Then you need to change your optimization to something else. So purchases or whatever your main funnel event isn't happening fast enough, you have to go higher up in the funnel. So if you're starting by optimizing down here, You should actually, and you can't maybe afford to spend that, or it's too nerve-wracking, go higher up in the funnel to an add to cart or a page view and start optimizing that. Get 50 conversions a week of add to carts. You're going to get some purchases here or there. And what's even better, what I think is even a bit smarter is let's say you are optimizing for purchases and you're not getting events fast enough, lower the price of the product just temporarily to get those 50 conversions a week. So Facebook knows who the heck to optimize for in the first place. And then you can go from there. 'Cause the more events you get, regardless of what they are, means faster learning and more stable delivery. And I really think that people should have optimizations happening for every event, regardless of even if they're optimizing for it. So you could have an event for add to cart, initiate checkout, but you're just optimizing for purchase. That's fine, 'cause all these events are gonna feed the entire pixel, which is just gonna get you out of learning limited faster. It's weird, a lot of people when they start advertising on Meta, they think it's just a big creative game and good ads and good copy, which is partly the case for sure, but a lot of it's based on data as well. You've gotta lean into the math and you have to let Facebook do what it's so good at, which is finding the right people and targeting them right for you. Next is actually structure. So fewer ad sets usually means more budget per ad set and a faster exit from learning. And you can see there's kind of 2 examples here. If you split the budget across 4 ad sets, each one can get a little bit too little to learn if you're not having high enough budgets. If you're not willing to spend this much, it's gonna be a little bit harder. And each ad set, even at $50 a day here, if you're only getting 14 conversions 'cause you're not spending enough, it's stuck. Versus if you just do $200 a day across 4 ad sets, maybe using a CBO campaign, or something else, you're going to hit those 56 conversions a week with the exact same amount of spend across that one campaign or that one ad set, which is going to give you a lot better results. So try that out and see how that works. Now, every major edit will send you back to zero. So you have to stop touching things. And these are 3 big things that will reset your learning phase. First is changing your budget by more than 20%. If you're doing big budget swings, it's not going to help you. If you're constantly swapping out your audiences or you're adding new ads to an existing ad set, that can also cause discomfort in the algorithm. So instead, duplicate the campaign or duplicate the ad set if you have everything in a campaign and make same change, make those changes with fresh copy, fresh audiences, or a different budget inside of there. So that way you have your winner that's running, and then you have a new ad which you can keep going on the back burner as well. At the end of the day, you just wanna set yourself up to exit the learning phase before you even hit publish. Upload all the creatives that you want in there, set your daily budget using the math, use broad targeting, and don't touch anything for 3 to 7 days. Rotate creatives by turning off underperformers, turning on some new ones, and then actually adding them not at the ad level, but add them at the ad set level or a new ad set to actually prevent you kind of screwing everything up that you've maybe built so far. And this is a little cheat sheet that you can use to exit the learning phase, uh, kind of going over everything there. And that's essentially everything. Learning phase can be misunderstood. I hope this super tactical, very direct video will help you get out of the learning phase faster. And like and subscribe for more videos like this, and I'll see you in the next one.
