Copy This Meta Ads Testing Strategy for Better Results

0:00If you're running meta ads and putting time and money into testing, but your results still aren't improving, there's a good chance you're solving the wrong problem. Most advertisers now understand the importance of testing, [music] but they fail because they're testing the wrong things in the wrong order. Here's what I mean. So, I speak to meta advertisers all the time, literally dozens per week, probably more than 100 every month. And these advertisers are going to range from small advertisers with small budgets all the way up to advertisers spending much larger budgets working with with bigger brands. And it's a common story that I hear of like, "I'm testing. I'm trying things. I'm

0:30experimenting, but I'm just not making any progress. My results aren't getting any better." And when I delve a little bit deeper into those advertisers ad accounts, there's two problems, two main issues that I very commonly see. I'll explain what those are and then and then give you the solution to them in case you are suffering with the same issue. So, firstly is that a lot of those advertisers, even though they are testing, they're spending too much time

0:53and effort testing marginal variables. And what I mean by that is small things. Things that if your test succeeds, as in the new element that you tested against your old version outperformed the old version. That's basically what testing is, right? I've got new ad creative, I put it in an ad set alongside old ad creative, did it beat the old ad creative? If you test small things, small variations on what you were previously doing, or stuff that doesn't move the needle as much. We can think of primary text variations, even headlines, call to action buttons, lots of different elements, a background color in an image ad. Those are just some

1:30examples. There are There are many others. Even if that test succeeds, it's just not going to move the needle that much. You might see a small percentage improvement in results. Was it worth it? In many cases, no. And that's not to say don't test, it's to say test other things. I'm going to give you a specific order for for what I'd recommend uh when it comes to testing in

1:51a minute. We all only have so much time, so much money, so much mental bandwidth to put into testing experimentation. I would much rather work on the big things that can really move the needle if they work out as opposed to the small things where if they either fail or work, doesn't make a huge amount difference either way. Very often as part of my mentorship program, by the way, link in the video description in case you're interested, I will be speaking to a business owner or you know, maybe a marketing manager within the company running ads and they will ask me a question of should I go with

2:24option A or option B? And very often I will tell them, "I don't think it matters." Now, of course, there are scenarios where it does matter, but often I will tell them, "I don't think that matters. Instead, you should focus on this other thing because this other thing could get you a doubling in your return on ad spend. What you're talking about here in between A and B as something that you've decided is important might be a 10% difference in performance at most. Let's use our limited resources on the things

2:53that can can really matter. Now, a lot of the time when a business is just testing those small marginal variables, those little things, it's because that's within their comfort zone. They don't feel like they have the capacity or the experience to go and do something completely different. It's easy to test different copy variations

3:12on an ad, just change the text overlay. It's easy to swap out the background image on your, you know, product images that you're using as part of your ad creative. It might be much more difficult for some advertisers, some business owners to go, "You know what? We've only been operating with statics. We're going to go ahead and make video." Might be that you have to develop that skill as someone on camera or you need to work with people that have skills in the video editing arena or people to be on camera for you. That's going to require additional cost outlay, which again might be outside of your comfort

3:45zone. But those are the sorts of moves, we haven't tried that ad format before, that can make a massive difference um if done well. I've had many scenarios where I've recommended businesses work with um influencers or get UGC ads made. And they're like, "That's not really the style of how we've done things before. We're a bit nervous about how that might position our brand." Often, if someone's wanting significantly improved in results, and I will have that conversation and they'll say, "Ah, but

4:11we're not really sure about doing that. It's quite different to what we've done before." My answer is, "Well, yeah. It's supposed to be quite different to what you've done before. If you want significantly different results than what you're currently getting what you've got before, you are going to need to do something significantly different. So, I think getting out of comfort zone is a big one, being willing to experiment with new tools, and and take

4:29a risk. And I think that this leads into the the second point around testing and something that advertisers need to be aware of cuz it's really, really important. Most tests fail. So, most of the time when you're introducing a new ad or a new element, it's not going to perform as well as your existing. And the reason why is because the existing ad that is still alive and still running is probably a good performer for your business. Now, you might say, "Oh, it's not generating a

4:5510x return on ad spend or whatever." Yes, but relative to the other stuff that you've produced, it's probably a good performer. That's why it's still on. The ads that didn't perform as well, you probably turned those off a long time ago, depending on how long you've been you've been running ads, right? So, anytime you create something new to test against your best performers, most likely it's not going to perform as well as your best performers. And by introducing the new ad, budget is going to be spent on those. If they don't spend as efficiently as in they don't get as high a return on ad spend, as

5:24good a cost per conversion, your overall result, when it comes to return on ad spend, cost per conversion, across that ad set, that campaign, your whole ad account, is going to decrease whilst the test is running. And a lot of Meta advertisers, particularly ones that are spending a huge amount of time in their ad account and they're also the owner of the business, they get very emotionally invested in the short-term results that their Facebook and Instagram ads are generating. I'm sure there are people who are watching this video who will be like, "Yep, yep, that's me." And they're checking the results constantly, and it's really important to them. It might

5:57be their most important customer acquisition channel, sometimes even their only customer acquisition channel. And it's easy to get wrapped up in how I feel about my business is dictated by the return on ad spend number this morning, yesterday. I've I've seen that hundreds, if not thousands, of times people be in that scenario. That sets you up to fail when it comes to testing

6:19in a number of different ways. So, it means that you could become reluctant to test in the first place. So, maybe you go through a few rounds of testing, but then you're like, "Oh, every time I test it seems to get worse results, so I'm just going to not." Because of the situation just described, most tests cause a temporary drop in

6:34performance. And it might mean that you're far less willing to take big swings like I've just talked about, and and go for things that are riskier. If you do something really different to what you are currently doing from say an ad creative standpoint, it's likely to produce really different results. That could be significantly worse or significantly better than what you're currently getting. Whereas, if you just test those marginal little things, likely it's not going to make a huge difference either way. You're going to get slightly worse, slightly better, the same result. It feels safer. And if you are too emotionally invested in what was my return on ad spend yesterday, what's my

7:08return on ad spend so far today, you could be reluctant to to make those those gambles. And all advertising to some extent is a gamble. We try and, you know, have house odds and and weight it in our favor and do all these things to help improve results, but there's always an element of we're going to throw this out there, and we're going to see how

7:26this performs. The other problem that that causes when it comes to testing is not giving things long enough. So, you run a test, you see performance drop off, you're like, "Oh, I can't have a drop in ROAS. I can't cope with that." And then you kill the test. You don't give it enough time to fully know how it's going to perform and

7:44how things are are going to shape out. If you make sure it got through the learning phase and give that a proper crack in comparison to what you're comparing it to. Because the big upside to consistent testing is if we do something really different to what we're currently doing, there is a chance. Now, it's a small percentage chance for each new test you run, but there is a chance that we find something that massively outperforms what we're currently doing. And you're only going

8:11to find that out if you take that risk. So, the way to think about it in terms of testing is almost like how a venture capital firm would think about investing. They're going to invest small amounts of money in lots of different startups, most of which are going to fail, but they know that the one that does succeed is going to pay for all the

8:31failures 100 times over. Exactly the same applies to testing. We're going to test lots of different things. A lot of them aren't going to work out. We're going to get worse results than our control, our existing ad. But, if we do try that UGC ad as an example, and our average return on ad spend in our ad account from Statics is a 2.3, and this UGC ad absolutely crushes it and gets a 5.7, that's a complete game changer. It was actually worth testing a load of ads that were coming in at a 1.9 and a 1.2 and didn't look like working at all in order to find that one cuz then, once

9:06you find the winner, you can spend so much more on that that it covers all the failed experiments 100 times over. Okay, so what's the solution to these problems? So, let me give you a a testing strategy and a framework that actually works. So, we're going to test the big things first. Already talked about that. The first thing we're going to test, the most important thing, is the offer. Offer being not just your product or service, but everything that

9:28comes with it. Things like guarantees. If you can add a guarantee of results to a service, for example, that's a massive way to offer offer the same service, but massively improve the offer. Guaranteed to achieve a certain result or they get their money back. Guaranteed to achieve a certain result or you will keep working with them for free until they

9:47achieve it. Um some sort of other guarantee involved. It's going to very much depend on on your business, the products and services that you offer, but a guarantee could be part of it. Urgency, scarcity. Do they need to buy before the end of the month, otherwise this particular thing is no longer available. Have you only got three slots left to work with you as a personal trainer. You know, I'm just coming up with examples here of of ways to think about offers and how you improve them. Could be discounts, buy one get one free, all sorts of different

10:10things, guaranteed installation times. Loads of different ways to to improve an offer. And for more information on that, by the way, my recommendation would be Alex Hormozi's 100 Billion Dollar Offers book. Absolutely fantastic. It's like a foundational educational resource. It's just brilliant for marketers, for meta advertisers, um to help you improve your offer. Cuz a lot of people think, "Well, my offer is just the product service that I sell. Maybe I put in a discount." You're like, "No, no, no. There's a lot more to it than that." And that book will will walk you through it. But the offer is the single most important part. You can have a poor

10:40campaign, ad creative can be average, but if the offer's good enough, it can work. The vice versa is not true. A really good campaign is not going to make up for a bad offer. And when I look at ad campaigns that are just generating no leads, no sales, it's nearly always an offer problem. That's the most common issue, and that's what you you need to fix first. We need to get an offer and test different offers until we find one that people are interested in, and you'll see that because that's when you'll generate leads and sales. And we need those leads and sales to then be

11:06able to test the other things because we need to assess the results of those other things based on return on ad spend or cost per lead or cost per sale. Um and we need those events to be happening in order to be able to make those to to run those future tests on different things. The next thing we test is going to be angle. So, angle refers to the main reason why someone might be interested in buying your product or service. Now, most products and services have multiple angles, and what I'd recommend you do is not try and include multiple angles within any one ad. So,

11:39you want to separate those out for testing purposes. That could be one ad per angle or a few different ads per angle, and then you run those and you quickly see, "Okay, people are most interested in this angle versus another one." So, let me give you an example. Let me give you an example of our our done-for-you Facebook advertising services, right? As as as an agency. Why might, let's say, a business owner Obviously, we work with like marketing managers within corporations, but let's use the example of a business owner. Why might And this is, by the way, just a side note, is why it's so important to

12:06know the specific customer avatar you're going after. Why might a business owner be interested in buying our Facebook ads done-for-you services? Okay? It might be that they want to improve their return on ad spend. So, they want better results. Pretty obvious angle. It might be that they want to scale, don't feel like they have the expertise or the capacity to do so. So, maybe they're not actually interested in improving ROAS, but they just want to maintain ROAS while spending a lot more

12:31and building a large customer base. And it might be that they want to save time. If they're managing their own ads, it's taking up more and more of their time. They have other things to focus on their business. That's another potential angle. And it might be to alleviate anxiety and overcome imposter syndrome. This is common if a business owner has, say, started really small, it's gone well, they've worked it out, they've scaled up to a point where they're now spending a good amount of money every month, and they're thinking, "I don't have the expertise for this. I'd rather hire a professional agency to handle this for

12:58me." Another example could be status. Now, the way that would work with our service is, let's say, a business owner had a bunch of other business owner friends that were all sort of into marketing, and they knew I was. Well, that person, if they hire our Facebook ads agency, could then go to their mates and say, "Oh, I've just hired Ben Heath's Facebook ads agency to run my Facebook ads for me." And that would provide status, assuming the people they're saying that to know who I am, which I appreciate will be not that many business owners, but will apply to some who who operate within this within this

13:27community. So, those are five different angles. So, we'd create different ads and work out which one is most important, and each ad would just focus on the one, and where do we get the better response rates? Now, having done those tests, different people obviously respond to different ones, but the single most important is improving results. It's return on ad spend. So, I know that out of those five different angles I just listed for our done-for-you Facebook ad services, return on ad spend is the most important. So, a lot of our marketing is return on ad spend led. So, we have screenshots and testimonials quoting improvement in return on ad spend

14:00numbers throughout our marketing. We demonstrate on my Instagram stories. You can follow me on Instagram to to see them. Um we do what we call result of the days um multiple times per week, right? So, we'll take a screenshot from a client ad account and show what performance is like, often with before and afters and things like that. So, because we've done the testing, we know that's important for our service, therefore that's the angle that we lead with. There's going to be different ones depending on different businesses. Um some of those ones I mentioned are going to apply across the board. Other ones could be value, could

14:27be saving. Um e- either it's cheap in comparison to alternatives, or it's just good value for what it is. It could be you've got better products and features than alternative competitors. Like, really think about why would someone want to buy ours instead of someone else's, and list out those angles, and then go ahead and test those. Once you've got an idea of which angles uh perform best, then you would want to move on to different styles. And this is one that really trips people up, advertiser, because they often get comfortable with a particular style of ad, and trying to get them to create other styles is really difficult. I'm

15:00not necessarily talking about format here. I know I kind of alluded to that earlier on in the video, but really with Andromeda, we kind of want multiple ad formats running simultaneously all the time. So, within an ad set, you're going to have statics and carousels and videos, for example, because different ones are going to speak to different people. Meta's also being really smart nowadays in terms of they might use your video ads kind of in a top-of-funnel capacity to generate awareness. You can get more explaining done in a video, but then retarget when someone's watched that video with the statics because it's just really quick and easy and they can

15:29see it and they go, "Yeah, yeah, no, I'm actually interested in that. I'm going to go ahead and check it out, buy it, become a lead, etc." Metal do all that within one asset. Don't need to have different campaign stages all within one asset. I've got other videos um talking about that. So, I'm not saying it's all about ad format here. I'm talking really about um styles of creative. This is going to apply more to to video ads. And if you aren't making video ads,

15:48definitely start. Definitely test that. If you are making video ads, good chance you're probably making a type of video ad, maybe two, and you're missing out on the others. So, what could you create instead? Obviously, there's UGC. I would separate that from influencer. Those to me would be two different styles. So, UGC, user-generated content, I'm sure you're familiar with that type of ad creative. Um that person doesn't necessarily have influence or a following. You can create very similar type of um ad creative using an influencer that is typically going to be more difficult to get, more expensive to get cuz you have to pay the influencer,

16:18but has more clout cuz it has more scroll-stopping power. People recognize the influencer and their recommendation carries weight. So, two different styles there. You could have founder-led videos. It's fairly self-explanatory what that's going to be. You could have demonstration videos on the product or service side. So, literally showing people, "This is how our products work." or "How we make our products." "This is how our service works." or or the end result. Um if you think about businesses that are portfolio-based, they often do really well with demonstration-based videos. You can think of like an interior designer, for example. You've got client testimonial videos. So, um people talking about how fantastic a

16:54certain thing is. You can have client testimonial videos that feature multiple client testimonials. You can mix and match these. So, you could have like a founder-led video that ends in a client testimonial or a few. You could weave a client testimonial into demonstration. But just I wanted to give that list there of the obvious categories that we think of um when we're going through, "Okay, have we tested these different styles?" You can also do things like um

17:15um, animation, animated type videos. Again, a totally different style. Won't work for a lot of businesses, will work really well for some. So, really have a think about that and try and get out of your comfort zone and test different options. Again, you might be surprised by which style works best. And then assuming you are using video ads, which I hope you will be, um, next thing we would look to test would be hooks. So, hook being the opening line and scene of a video ad, first few seconds. Changing that is one of the highest leverage activities, or should I say, testing that is one of the highest

17:44leverage activities you could do as a Meta advertiser. Because the hook is what convinces people to either actually watch it or not. And if they don't watch it, they don't hear about what it is that you're advertising, they aren't going to take the action you want them to take. If they do decide to watch it because of the hook, make each difference. And you can see certain ads that have a hook rate of, let's say, 8%. So, 8% of people made it past the first 3 seconds. You change out that hook and all of a sudden the hook rate is 33% and just think about how many more

18:13people the same amount of people, you know, I'd been put in front of, but that what's that going to eat? Just over four times more people are actually hearing about your thing. It's a huge leverage in the hook rate and that'll be something we look to test. There's too much fear, quick side note, about Andromeda and not being able to test, um, different hooks because Andromeda will see the ads as the same thing. I'll show you how to get around that in a second, but don't worry

18:33about that. Those are the most important things to test. After that, if we want to, and often we won't get to that cuz I'd rather keep testing those big things, then, but it but if we really feel like we've got those nailed, then we will start to testing smaller variables like primary text, headlines, um, small tweaks to static images like background colors at CTA buttons, um, and things along those lines. So, I mentioned a second ago about testing different hooks and Andromeda and the fears that you won't be able to do that cuz Andromeda post-Andromeda Meta will see ads with different hooks as the same ad and only enter one into the auction,

19:05which can happen. Here's how you get around that. So, I'm in an example Meta ad account, just going to create a quick example sales campaign and I'm just going to jump to the ad level. And when you own the ad level, if you scroll down beneath this ad creative section, you've got this setup test and creative testing option. So, if you go ahead and click on that, you can test two to five different ads. So, let's go ahead and put in five. Uh how decide how much of your daily budget, so how much of the budget from within this ad set you want to allocate to this test. This

19:32is defaulted to £11.89. I don't know why out of a £25 a day daily budget within this ad set, that's fine. How long should you run the test? 7 days, etc. Just a very quick thing on creative testing. Uh make sure you change how do you want to compare performance from cost per post engagement to whatever it is you're optimizing for. So, it could be for example cost per lead. Um this is a sales campaign, so it'd actually be cost per purchase or return on ad spend, but you know, make sure it's just the end

19:56thing that you want to to optimize for. I've made this change. Meta's saying that uh we might not be able to do this, so you might need to extend your time frame or spend more money because we might not generate enough leads to know, particularly with five different running. So, you can adjust that accordingly. But, this is how you do it cuz what Meta will do is part of this creative testing tool, they will separate out segments of your audience, only put the different variations in front of that segment, so there's no overlap, and give you really good ideas on what performs well. And this gets

20:24past the whole Meta's looking at these two different ads with different hooks and seeing them as the same ads and only giving one of them ad spend, only entering one of them into the ad auction, well, you're going to force that to not happen. If you click confirm on this, um then you can see, look, we've got the five different ads here underneath the new sales ad with the you can tell it's a test cuz you've got that little signal there. And then uh yeah, so you then just go ahead into each ad and change up the ad creative and create

20:48those ads with the separate elements. Then you run your test, and that's how you get round that. In fact, we will use this creative testing tool for a lot of different testing in order to really get to understand, okay, this performs better than this, which is the big part of testing. And if you use this, you have more control as an advertiser to get those answers and to prove results over time. Hyros is the best tracking and attribution software for Meta ads. It's a software all the big players in this industry use, including myself. I use it to track results for my own ads as well as all my

21:18organic results as well. So, if someone books a call for our done-for-you services, for example, I can see exactly where that person came from with a lot of accuracy. That allows me to make better decisions for my marketing and my business. Hyros has been able to track that from our ad spend in this campaign, we've generated 96,000 pounds. But 58,000 pounds of that was not reported by Meta. So, if I was just looking at my data within my Meta ad account, I might come to the conclusion that this campaign is doing okay, but in reality this campaign is doing a lot, lot better. Now, this is a

21:48relatively extreme example because of the nature of our business. So, most of our product services have a recurring billing element. Meta might be able to see what we make initially from our ad spend on that initial transaction, but not those subsequent payments. That's why it it shows the importance here of being able to accurately see what did we actually get from our ad spend in order to be able to make better decisions and optimize correctly. So, if you want to improve the performance of your ad campaigns without having to touch your

22:13ads, go ahead and sign up to Hyros. There is a link in the video description. Now, one thing I want to quickly bring back up cuz I mentioned it briefly, but I want to emphasize how important it is, is using influencers in your ad creative. That is one of the most common recommendations I will make to Meta advertisers. It is also probably the single recommendation that is ignored the most. And it's not ignored

22:36because people don't think it'll work. It's ignored because it's difficult to do. Going back to the beginning of the video, it's very much outside of people's comfort zone. What I'm going to try and hire influencers to create video ads for me to promote my stuff? The fact that it's difficult to do is part of the reason why your competitors aren't doing it, which A is great because if you can make yourself do it, then you gain massive competitive advantage. But also, there is nothing that I see repeatedly improve Meta ad results more than that. Than hiring influencers to produce video ads. Can be statics as well, but

23:08mostly it's going to be video ads that you can run as partnership ads, so it comes from both you and the creator is in the the from from the Facebook page, Instagram accounts, etc. Two main reasons for that. If that person has influence over your target market, which is exactly what you want, they have a following, the hook rate is likely to be way higher, as in people actually stop and watch the video as opposed to just scrolling past it at a much higher percentage, which is we've already talked about is really important to getting someone to actually find out about what it is you do and then

23:36convert. So, the hook rate is much much higher. But also, their recommendation then carries weight. If someone who has influence over an audience says, "Oh, by the way, I love this product. I would recommend you buy it." The conversion rate from views to clicks, then clicks to conversions, potentially even repeat purchases if you keep doing it, all of

23:57that is much higher. And when you're hiring an influencer, you also circumnavigate some of the other potential issues you might have with creating video ads because, right, the influencer is going to be good on camera. That's why they're an influencer. So, that box is ticked. Um you haven't got to go out and hire videographers, work out editing, production. They're going to handle all

24:16that for you. So, yes, it's going to cost a bit of money, not as much in many cases as most people think. You can absolutely start small with micro influencers and then look to to scale up. But it is it's as close to a Meta ad cheat code as you can get. And I spend a lot of time going to like Meta events and meeting with other agency people. And it's one of the two things that consistently comes up as like, "We are pushing more and more influencers and partnership ads for our

24:43clients and it's just working so well." Now, coming back to influencers, you might be like, "Okay, yeah, Ben, I get it. I should be working with influencers, but how do I get started?" Well, I'd recommend you go ahead and check out this video cuz in that video I walk you through the creator marketplace and how to use it. It's a free tool that

24:56Meta provides, and so no cost required. And in my experience, one of the best ways to find creators and influencers to promote your stuff. There's like 1.5 million plus creators growing all the time. You can search by category, etc., etc. Go ahead and give this a watch. That's the best way to get started.