How To Find Winning Facebook Ad Creatives Without Wasting $10K In Testing

A three-step system for testing Facebook, Instagram, and TikTok ad creatives without overspending: research trends using the Facebook Ad Library and Creatify, generate 50 headlines and build simple ads around them, then test with either a CBO campaign ($50-$100 daily across eight creatives) or a scientific method ($20-$30 per ad set for precise control).
Chapters
Coming up with brand new ad creatives and finding winning ad creatives is one of the most crucial elements when it comes to scaling your paid ad campaigns on Facebook, Instagram, and really across any paid ad platforms. So in this video, I wanna give you the 3-step process that we've used to find winning ad creatives without blowing your ad budget, without blowing budget period. So let's just get right into it. I'm gonna draw some stuff on the screen, give you guys a breakdown of what the actual strategy looks like, and then I'll take you guys inside of an ad account and show you what it actually looks like in action. So first things first, when it comes to finding winning ad creatives without blowing your budget out is research, researching and figuring out what is working in the industry, how things are working, what angles are working, hooks, copywriting, the framing of the videos, the framing of the pictures is the most important thing. And it's really the big foundation of actually making sure that this works. So where can you research and how can you make this happen? Well, there's a few areas that I really like to go to. One is easy, is the Facebook ad library. You can see what people in your industry or in your niche are actually running, how long they've been running it, what the ad looks like. And it's really just going to get your creative juices flowing and actually allow you to figure out, okay, what are some ad ideas that I could use here? The next one is something that I don't see people talking enough about, and that's just social media generally. So you could do things like Reddit. Um, you could do things like YouTube, which are huge, uh, Facebook forums, Facebook groups, and school communities as well. These are all really, really powerful platforms. Whereas if you can go on these platforms and research what's working, what's not working, what people are looking for, what questions they have, those are gonna be ad angles that will directly influence the hooks or the creative inside of your ad accounts as well. Use ChatGPT as well. Use the tools that you have, do some research and find some winning creatives or winning creative concepts that you can use Now, the last element that I'll talk about, which is a cool tool that I've been using recently, and it is called Creatify. Creatify is really cool. I'm not sponsored by them at all, but it's essentially an ad spy tool and it works in a really cool way because it allows you to find market trends based on certain keywords. And all it's doing is it's pulling stuff from the Meta and TikTok ad libraries anyway. So you could find this stuff if you did it yourself, but the search engine around it is really cool. And then what you can do is if you find some ads you like, so for example, I found some from Tony Robbins. You can save those ads, you can view all of them here, and then you can actually view the insights of the ad as well. So you can see the creative insight, you can see the selling points around that certain ad, the target audience, you can even read the copy, et cetera, et cetera. What's really cool for video ads as well is you can actually grab I'll just go here real quick and I'm just gonna click on a random video. It'll tell you how strong the hook is. It'll tell you what the hook is and the different ways that they're selling points. And you can see all the different hooks here. So really, really cool stuff. It allows you to get a ton of ideas for your different ads based on competitors or based on other industries. Another really cool thing in Creatify is it'll give you an AI summary of the brand. Selling points, hooks they're using, ad copy, and headlines. So this gives you more than enough information to be able to take and use into actually making solid ad creative. So for research, Creatify, Reddit, all social media platforms, Reddit is a good one, any kind of forums, and then the Facebook ad library. So after that point, what actually happens next? Well, then we actually gotta create some stuff and I cannot spell. And when it comes to creating things, it's, actually a lot more simpler than you, than you need to make it be. Um, 'cause some people overcomplicate this a lot. And when it comes to creating something, the structure that I like to follow is first off, coming up with a ton of hooks or headlines. And I think a really easy way to do this is to write up just on a piece of paper or in a Google Doc, 50 different hooks or headlines that you could use in your ads. Now, you may not necessarily use them as the whole ad. It'll be maybe good for an image ad. It may be the beginning part of a video ad that you do, but coming up with these hooks and headlines make up 80%, 90% of the success inside of your creative. If you're doing a video ad and you have 50 hooks, you've got 50 split tests right there. You just gotta film that main chunk of the video. That's always the same. If you have 50 headlines, you can use those in an image creative, same layout, and simply just test a new headline. Creatify, looking at and having proper research is one of the best things you can do to actually make good creative and make it easy to make good creative too. And that's an important thing. It's not just about making good creative. Anyone can kind of spend a lot of time, energy, and just put a lot into making a really, really good creative. But the thing is, how easy are you making it for yourself or your team or your company, agency you're hiring, whatever it is? How easy do they actually make good creatives? Is it a— Once-off thing here and there that kind of works? Or is it actually repeatable, and you're making creatives based off of a system that are working? So making ad creatives, that's exactly how I like to go about it, is by making new hooks, new headlines, and using those to kind of as the baseline for my creative. Now, the last piece in making this whole thing work is actually the structure. And this is the structure inside of the ad account. I'll kind of break this down really quickly, but there's 2 ways that I like to test creatives. There's the optimized structure is the scientific way of making it, or like the specific way of doing it. And the difference between these 2 different methods is based on really how much you're willing to spend and how you wanna structure everything as well. Both of them will work. Like I said, it's just really based on how much you're actually spending. So let me go through both of these first. Here I'll draw them out and then we'll actually go into the Ads Manager and I'll show you how to set these up as well. So optimized structure, we'll start here at this top level and this will be your campaign. In an optimized structure, you're gonna have a CBO budget. So a campaign level budget. From there, you're gonna have 4 different ad sets. So those are gonna be these right here and these will be your ad sets. Within those ad sets, each 4. So if we had 1, 1, 2, 3, 4, each of those ad sets are gonna have 2 ads in them, each testing different angles. So let's say you had came up with 50 hooks in this previous step that we spoke about, and you can make 50 image creatives. You're gonna have a lot to work with, but at this point you just really need 8 creatives to start because 2 ads per ad set, and you'll have 1, 2, 3, 4, 5, 6, 7, 8 creatives. So you'll end with 8 creatives. And the reason why you don't need as much budget for this is the campaign budget here is gonna distribute the spend across the 8 creatives. Now it may not do it evenly, and this is a big thing. Understand that Facebook will not necessarily test to all 8 creatives. You're not necessarily gonna get a good idea if ad 1 is actually a winner compared to ad 7, um, or et cetera. But the idea here is to find something that works quickly without really spending too much budget. I'd recommend starting off the CBO budget anywhere between $50 to $100 a day. And that's kind of a broad range, but it really just depends on what you're optimizing for. So if you are optimizing for, um, a booked call or for a sale, if I can get get my thing to move down. That's really what the difference is gonna be. So like booked calls, you want those to be around like $100 a call. So I'd probably set the budget to around $100, $150. But if you're looking for sales and you want your sales to be around $30 to $40, I think like a $50, um, even a $60 a day spend is smart here. So just depends on what exactly it is that you're testing. Is how you're gonna structure this thing. So that's really the best way to set this thing up is let me just kind of go over this again. First step here is to research, use tools like Creatify, Reddit, YouTube, Facebook groups, school groups, different forums to figure out what people actually want, what problems people are having, and what solutions people are looking for. You can use the Facebook ad library as well. Then it comes time to create based off of all the research that you did, the problems, the current situation, desired situation that people are actually looking to achieve. Let's actually make some stuff and start simple. Start with short hooks, short headlines that you can use and you can take and using the research that you did and some of the creative ideas that you like, model those with the creatives and the hooks and the headlines that you have kind of come up with or have in your head. Once you have those creatives, start with just 8. From there you can do either an optimized structure or scientific, which I'll get into in a moment. But for the optimized structure, you'll take your 8 creatives, you'll do a CBO campaign. 4 ad sets, 2 ads per ad set, test those, but just understand that you may not get a very, very specific result on if ad 7 is actually a worthwhile winner because Facebook may spend all of the budget in ads 1 through 3, but nothing on the rest of them. That's fine. Facebook is pretty smart. Facebook's algorithm is really smart. So let it do its thing. And then you can always switch to the scientific structure, which I'll get into now. If you want something a little bit more specific and you have a little bit more budget to use. Okay, so now talking about the scientific structure of testing. Scientific. My handwriting's not great here. My pencil's being all weird, but with the scientific structure, it's actually very, very similar. You're gonna have your one campaign. You're then gonna have multiple ad sets, and this is where you determine your budget. If you're, I would recommend having around $20 to $30 per day. Per ad set and per ad. So depending on what you're willing to spend is how many ad sets you'll have and how test— quickly you want to test and find a winner. That's where this will come into play. So say I'm only willing to spend about $60 a day. I'll make 3 ad sets, 3 ad sets, with each one of these spending $30 a day. Each one will spend $30 a day. We've got these 3 different ad sets. And then inside those 3 ad sets, we're going to have 3 different ads. So ad 1, ad 2, ad 3. Now the difference between these 2 strategies is that this one, you're controlling the spend and you're controlling what creative gets to spend. Each creative will spend about $30 a day. So you'll know for sure if ad 1, 2, or 3 is actually working. And then obviously you can obviously test a lot more than 3. I'd recommend testing kind of similar to the optimized structure here. 4 ad sets, 1 ad per ad set. You're testing 4 ads at a time. If you want to spend more, you could spend more and test more creative and figure out what works. Now, a really cool way to kind of mash these 2 together is you could start with the optimized setup, figure out what works, figure out what Facebook is kind of drawing more spend to, take those top winners from here. So let's say, let's say ad 1, let's say ad 3 and ad 7 were working really, really well. Well, what you could do from here is you could take those 3 ads and then set up a new campaign where you're basically testing a more scientific testing strategy where you've got your one campaign, you've got ad 1, you've got ad 3, you've got ad 7, each in their own ad set, each spending $30 a day across all of them, which is a really, really cool way to kind of merge those 2 together. So this is exactly what we like to do when we're testing ad creative is do the research, figure out what actually people want, come up with some headlines, and then test it in those 2 structures. Really easy way to do it. I don't think I need to show it in the Ads Manager, made it pretty simple. When you're setting these up though, for a campaign budget optimization, make sure you choose that here before you kind of get to the ad set level, obviously. So you can choose campaign budget, Just a normal daily budget for whatever it is that you want to spend. And that's pretty much it. If you enjoyed this video, let me know. You can click below for Creatify if you're interested in it. Um, I don't know if I have an affiliate link. If I do, I'll put it there, which is cool, but it's been a really useful app. I'm not sponsored by them at all. Um, it's been really, really cool and I've been using it in my company. So I think you should check it out as well. And if you're looking for more ads help, you want us to audit your ads, just click below, reach out to me. Glad to check out your ads if you're a good fit. And if you're actually spending money, and have a good offer on your paid ads, we can definitely help you. I hope you enjoyed this video, and we'll see you in the next one.










