5 Mistakes Small Businesses Make When "Testing" Their Meta Ads

Hello, and welcome to Metta ads under the hood, the podcast where we explore what drives high performing Facebook and Instagram ads. I'm your host, Megan Winter, and with over a decade of digital marketing experience, mentoring hundreds of business owners at the Facebook ads academy and running my own ads agency, I am here to lift the hood on Facebook and Instagram ads. Each week, I break down the mechanics of Metta ads, unpack real campaigns and share how I use ads to drive results and how you can do the same. Ready to take a look under the hood? Let's dive in! Hello, and welcome to the show. Now, if you have been here for any amount of time, you have probably heard me say until you test it's just a guess. But what does that actually mean? And how can you use that in your business to make sure that your meta ads and your marketing are driving a result? In today's episode, that is what I'm going to dive into, so let's get started. So until you test, it's just a guess. What does that actually mean? So until you actually make something real, it is just theory. Until it becomes a physical, practical thing out there that you can look at the data, you can see, you can touch, you can feel it, it is literally just theory. Until it's put into practice, it is just a concept. And so until you test the thing, until you make it real, it is just theory. It is just a guess. So until you test, it's just a guess means that until you actually have something to measure, you are literally, you have a concept, you are just guessing. And it might be a very, very educated guess like if you have come to a marketing strategist or, you know, you've done a really amazing plan and you've got some advice and you've worked with people and you've got this amazing plan. That is still just a very educated guess. It is a strategic plan. Now, the plan is a roadmap. The plan will give you all of the steps, but until you actually roll that plan out, you don't know how it's going to go. You can have projections, you can have forecasts, you can have our possibilities. And really, all that is is a hypothesis. So if we look at the scientific method, we've got a hypothesis, which is like a, I think this is going to happen, then we test it, then we get our results, then we analyze our results, and then we work out the next step, which is usually another hypothesis. So what we want to do is we want to make sure that what we're testing is actually giving us the answers or giving us answers that mean something to us. We want to, yes, make a very educated guess. So we don't want to just go, and I've seen this a lot, especially with AI coming into place. People are just making images with AI, they're running them through. They add, so using the AI add image generators that meta have brought out, and they're like, oh, well, until you test, it's just a guess. And so they're like, you know, running them and seeing what happens. Now you've got to also bring into the mix, brand reputation, and all of these other things, it's not just like a siloed approach. So we don't want to just test for the sake of testing, and we don't want it to be testing rubbish, because rubbish in rubbish out. So we want to test things that are in line and aligned with our business, with our brand, with our strategy. We want to make sure that that idea is actually aligned with all of those things. We don't want to just test slop, because even if it gets a couple of quick wins, it's going to do damage to our brand, and that's not good either. So people ask me all the time. They ask me like, okay, so tell me what videos I should create for this masterclass, or tell me what my landing page should look like. And what I can do is I can say to them, well, here's what's worked in the past. Here's some videos that are working really well for other creators, or here's ones that I've used for myself, or if somebody's got an e-commerce brand, I can say, here's the ads that are working really well for some of our other clients. And so you can kind of look at that, you can analyze that, you can then create your own, but until you test it for yourself, again, it is just a guess. So we want to test everything, but we want to test one thing at a time. And we also want to not get to hung up on the results, like we want to go through and test as quickly as we can. So we want to create, like I said, a good educated guess, but we don't want to dwell on it. So this is where the balance kind of comes in. We don't want to dwell on it, we want to create it, we want to get it out there, we want to get data back, and then we want to analyze it. So when I say test everything, but test one thing at a time, when it comes down to meta ads, we can test the creative concept. So for example, we can test like UGC content, we can test Foundal ed content, we can test different problems that your product is solving or different reasons why people should use your service. So there's a lot of different angles that we can test, we can also test the individual creative elements. So within each one of those creative concepts, we can test different videos, we can test different images, we can test different types of creative. So we can test videos versus images versus carousels, and sometimes we actually need a mix of all of those things. So I'm not saying that we want to just find one thing and stick with it, but we do want to test everything to see if that works, and if it works with the other things. We can also test things that are not necessarily on meta. So we can test the destination. So where are the ads pointing to, which is the landing page? And then once you've rolled out the test, so we want to test everything, but test one thing at a time, we want to make sure that we've got enough data to analyze. So usually as a rule of thumb, if we've got about 2000 impressions or 2000 reached depending on what we're doing, we can then analyze that. So you don't need to spend hundreds and hundreds of dollars on a test to see that something doesn't work. You can see quite quickly if it's working or if it's not working. Now if it's not working, or even if it is working, the data will always tell you. The data will always give you a story or a whole picture of what's happening. Now there's a couple of things here. One, I want a deep dive into, I mean, I don't give you two case studies actually. I was thinking maybe I would do these as separate episodes, but I'll touch on them here. And then if there's enough, I will dive into them in separate episodes. But I'm going to give you a case study where the data or the results are great. And then I'm going to give you a case study where the results are horrible and what the data is telling us. Now a little caveat here, something to remember when we are looking at how to analyze the results. We need to remember that Meta's dashboard is gaslighting us. All the platforms are. It's really important to understand how attribution works. So if somebody discovers your branch through a meta ad and then they go directly to your site and then they may be like retargeted with a Google ad, all of those platforms. And then and let's just say they also get an abandoned cart email. All of those platforms are going to say, yes, that was me. So at some point or another, because you've got first click attribution last click attribution, you've got all of the different platforms are trying to say I did that. That was me because they all want to claim the glory. So it's really important to know that whatever your meta dashboard is telling you, it might not actually be what's really happening. So we need to understand that it's actually more important to look at the source of truth. So look at your website, look at your sales, look at your calls, booked, whatever the outcome is, look at that. The other thing that I think is worth mentioning here is how the pixel works. So the pixel is just a little piece of code that goes on your website. It's like literally a couple of lines. But what it does is it pings back activity from your website to your meta account, like two meta essentially. And it has to be your pixel. So you have to own it. It has to be linked up into your art account, into your business manager. And what happens is if you have a purchase pixel installed on your home page, for example, or a pixel purchase event every time someone goes to your home page, meta will say someone made a purchase. And that happens. Like I just had a client on board and they have had a triple fire on their purchases since 2024. So their art account is super confused. Like it is thinking that people are purchasing way more than they actually are, thinking that it's getting like six dollars cost per purchase when actually it's getting at $18 cost per purchase. So the way that the pixel works is that no matter what is happening on your website, unless the pixel is coded correctly and unless the pixel is actually sending back data correctly. Meta doesn't know. Meta cannot understand. stand what your customers are doing unless your pixel tells it. So it's important to make sure that you're actually tracking things correctly as well before we go into until you test it's just a guess well how do we analyze the data we need to make sure that we're actually looking at the right data. So I always look at in the meta add dashboard I look at click through rate and the click through rate the CPM and the cost per click will all work together. So one will go up and the others will go down or one will go down and the others will go up and so all of these these kind of play together but if you've generally got a good click through rate it means that your ads are doing their job and so you can kind of have a look at if you're testing your creative you can go okay well what ads are people stopping on and you can look at the thumb stop rate there and then what ads are people clicking on and you can look at the click through rate there you can also look at the CPM which is cost per thousand impressions and you can see like what ads or creative is meta-charging us more for i.e. what does it think the good creative is because it'll charge us less for that. So there's some of the things that you can use to analyze your test for your creative and then you want to look at the actual destination so have a look at your website so have a look at your conversion rate on your website look at your repeat customer rate look at your average order value one thing that we don't look at is the rowers the tracked rowers within your meta dashboard and i think it's fairly obvious as to why by now because like i said the meta dashboard is gaslighting us the attribution is all over the place what matters is money in money out is your accountant happy what next so are people coming through from the ads to your website are people purchasing on your website and does your website have a good conversion rate so when i say until you test it's just a guess we have to understand what we're testing then we have to understand how we're going to measure that so that we're not guessing anymore and then we need to understand how to read that data oh hi it's me interrupting me just to let you know about my training program if you would like to learn how to run your own ads you might like to come and check it out in this training program i set your ads up for you and then over eight weeks we run them together every single question is answered by me and at the end you are confident to take the wheel and run your ads yourself back to the episode now like i said i'm going to give you a couple of case studies so the first one is of an ad account that actually isn't working so this is an actual email that i had to send to a client and i sounded exactly like this the overall results took a dive last week and the ad account spent more than it made which is bad but as always the data tells us a story and here's what the data is telling us the overall site conversion rate dropped to 1.84% which is still healthy and as expected i turned on the view content campaign to drive traffic the purchase ads are however converting at a 3.4% so the ads themselves are attracting the right audience the conversion rate dropped to 2% when we started pushing traffic to a different product and this clearly indicates that we need to optimize that product the AOV dropped to 34 dollars which indicates people are trying the product for the first time as in they're just you know buying a little amount rather than buying like a bundle or something because 50 out of 53 customers are new to the brand the AOV did however increase to 52 dollars when we started pushing the new product the click through rate on the ads is around 1% which is benchmark so the creative is engaging people and enticing them to click so we don't need to test any new creative at the moment cpm is healthy so meta isn't charging us too much in a nutshell the ads are doing their job the AOV has increased but the conversion rate has dropped since adding in the new product and this is fixable see the next steps below so we then went into what we were going to do so basically we're going to bring down the budget until we fix that conversion rate because there's no point in spending if you know it's not actually converting and we've you know paused a few of our more advanced campaigns which is like what we start to add as the clients go on we have then said to the client that the priorities are to optimize the landing page and we've given a full breakdown of how to do that and we've also said let's look at a free shipping threshold to increase the AOV to a hundred dollars or above and so you can see there right that we've started testing a few things there so we've got creative that's running and testing we're testing that we're testing landing pages so we're testing some different products we're looking at the overall result and we're going okay the ads like I forget what it was so let's just say for this week the ads account spent two thousand dollars and the Shopify account only made about two thousand dollars so it actually spent essentially that that is losing money because you've got your cost of goods and everything like that so what most people's instinct to do is to go okay well what creative do we need we need more creative we we need to like change the art account we need to like have some more campaigns in there what what do we need to do but actually what we need to do is we need to slow down and we need to look at the data because until you test it's just a guess we have tested and we're not guessing anymore the data is now telling us exactly what we need to do the creative is working it is getting people to stop their scroll it is getting a lot of people to click on the website those people who click on the ads are actually converting fairly highly compared to other traffic but still lower than what it was before so it's very very clear that the landing page is not converting now I could say and I have said here are my recommendations to fix the landing page but again until we test it's just a guess so what we need to do is we need to make the changes on that landing page we need to then run some more traffic to it which is so I've now reduced the budget we're going to run some more traffic to it and then we are going to test that new landing page to see if it improves and that is kind of breaking it down that is as simple as it is we don't freak out we look at the data we see what it's telling us we make one hypothesis which is let's change the landing page and at the same time I've reduced the budget because why would we waste money we're going to reduce the budget we're going to fix the landing page and then we're going to test it to see if those changes actually worked so until you test it's just a guess we are going to have a guess which is my you know educated guess based on my experience and what I'm looking at and what our other clients are doing and all that kind of stuff I'm then saying let's make these changes and then test it the other case study that I wanted to talk to you about is a client that actually spent 20% less than what they budgeted but they made 4% more in sales now 4% more in sales doesn't really seem like much but when you've spent 20% less in budget that actually is a lot so what this what we were able to do with this campaign as we worked out a very specific target this client wanted to generate 100 memberships we were able to then reverse engineer and again guess in inverted colors very educated guess but still guess we were able to reverse engineer the budget that she needed so we were able to say okay we think that your leads are going to be this much we think that your conversion rate on your landing page is going to be this much we think that this percentage of people are going to show up on your webinar this client was running a webinar we think that this many people are going to convert from that show up rate and so therefore if you want a 100 people we can really easily work backwards and we can say okay so for that we think you need to spend I think it was 5 grand or whatever it was 5 grand total budget and the client was like yep sounds great so we started running the campaign so we were now testing it so until you test it's just a guess so we started testing we started rolling it out now what a lot of people do is they'll just run the campaign and then kind of do a post-mortem at the end but what we can do and what we do and what we recommend what I recommend that everybody does is run a small amount a budget through pause and then have a look at what the results are and see if our estimates if our if our guesses are actually correct so what we found was that the lead cost was actually coming in at about half half of what we had expected so two options we can keep spending the same amount and get double the amount of leads or we can say we're going to spend less and hit the target and we asked our client what she wanted to do this was the first time that she was rolling out this product so she said let's just spend less and hit our target of leads and then we'll see what the conversion rate is and then you know well that's what she wanted to do so we actually spent 20% less in the end we spent 20% less we we did get some more leads then expected but we spent less than expected we then then she was able to run through the whole thing so she was able to run her webinar and see how many people showed up was that the same as what we had expected. And it was fairly similar. I think we'd expected 30% of people to show up and she had about 25. So a little bit less, but our lead cost was a lot less. We expected that 10% of the people who showed up would convert and we expected the average order value to be $400 or whatever it was. And that was pretty bang on. So we guessed it and we tested it and then it kind of proved to be correct. Now so many different things could have happened there. It could have been that the lead cost were actually double. So then we go, you either need to adjust our budget or adjust our projection. And so you can really see that it's quite an agile and adaptive process. So at the end of this, the client was able to convert at the rate that we had expected. And she sold 104 memberships. So she wanted 100. She sold 104. But we spent 20% less of her budget. So she made the same revenue goal, which she was very happy with. And she spent 20% less of her budget, which she was going to be happy with if we spent the whole budget. But she spent less with so she was even more happy. So that just goes to show, you know, we can roll out a plan for two different clients. And sometimes it works just straight away. Like for that client, it actually just worked on the first goal. Like that's the first time she sold this thing. And she sold 4% more, spent 20% less. For this other client that we've just started working with their back end is like, it's a bit of a mess. So it's going to take a while. So we're rolling out our strategy. And it's actually not working on the first goal. There's a lot of things that we need to fix. Like we need to fix their landing pages. We need to, you know, there's a lot there that we still have to do. But the data is telling us exactly what to do. So if anybody comes to you and says, I know how to make you 20 grand a month. I know how to get you this, you know, just follow my formula and your guaranteed results. What I want you to think is until I test, it's just a guess. Until you test, it's just a guess. You can follow the exact same formula that I do for something. And it will work differently for you. Your audience is different. Your business is different. Everything is unique. I hope otherwise you're just copying someone else. You need to work out what works for you. So I just really want to get this mentality of until you test, it's just a guess into everybody's minds, everybody's business. But that is what business is all about. It is literally just having a plan, having a hypothesis, testing it, analyzing it, seeing what works and doing more of what works and less of what doesn't work. So I really hope that this episode has been helpful. I had a loose plan of what I was going to talk about. And I hope it's made sense. But so what what I really want to just leave you with is like I said, you can have the best educated guess that there is. But at the end of the day, there's no way that you can guarantee what is going to happen until you actually put it into action until it becomes a reality. It is just a concept, which is another way of saying until you just it's just a guess. So I hope this episode has been interesting and valuable. Please come and see me over on Instagram @meganwinter, M-E-G-A-N, W-I-N-T-E-R. Let me know if you have liked this episode, let me know if you have any questions. And if you've liked this episode, I would love for you to leave a five star review on Apple or wherever you listen to your podcasts. That would absolutely mean the world to me. Thanks guys. I'll see you in the next episode. Bye. This episode was brought to you by my business, loom marketing. Whether you want someone to take the wheel and run your ads for you or you want to learn how to run your ads yourself, we'll make sure your meta ads are driving results. You deserve meta ads that don't let you down. Come and check out loom marketing. That's l-u-m-e-market team.com or pop a link in the show notes.