If you’ve been subscribed to the blog, then you’ll know you’ve probably seen it before. In fact, every Thursday I put out a poll asking you guys, then what kind of content you want me to cover that weekend come in so this Thursday, then I did exactly that I’ll put screenshot up here and, as you can see, then the clear winner was Facebook Ads testing.
So that’s exactly what today’s article is going to be about tomorrow, then Sunday will be on the second place winner, which was Facebook as scaling. So if you want to see that article then make sure you hit that subscribe button, and that means something guys, that’s a topic. Thank you for tuning in. I hope you enjoyed the article and let’s get straight into it. What is going on, then guys welcome to my ad manager account.
So the way I’m going to structure this article, then, is I’m going to start off firstly going through why this strategy actually works. I’ve got a Google document that we’re going to go and actually work through kind of. Why the reason in why we have to do things this particular way and why this particular strategy works and then afterwards, muchly going to go through on my ad manager, account and actually create this testing strategy.
So you can see exactly how to set up yourself now. I’m fully aware that some of you won’t care about why this thing works, even though I thoroughly or strongly recommend them that you do read this first bit. I will pause time stamp up here somewhere in the middle of the screen, for those of you that just want to skip ahead straight to the actual creation of this strategy. So that means said then, let’s get straight into it.
So Facebook Ads testing strategy new for 2019, so number one then before we get into this strategy itself, two things you need to consider. First number one: Facebook Ads optimize on the answer level. We’ve got proof of both these points as well, and then number two facebook recommends a minimum of 50 conversions per week for ad optimization and that is on the ad set level. So if you’ve got 10 ad sets within your campaign and each ad set is get in, say, 5 conversions, even though you’re hitting those 50 conversions, not one single one of those ad sets is getting 50 conversions on its own.
Therefore, it’s not going to be delivering optimally and just to prove it to you guys, then I’ve got Facebook’s own business website up here and the first thing I want to show you, then, is this bit here, so this is also crucial to making sure you get In 50 of the conversion events you optimized, for so, if you’re choosing purchases, then these must happen within your conversion window. So this is going to be a crucial point to the testing strategy as well, and then the next point is well probably the main point as well is that keep in mind that we recommend getting about 50 optimization events within your conversion window per week.
So what that’s basically telling us is that if we don’t get those 50 conversions within a week there now ads aren’t going to be delivering optimally and they’re not going to perform consistently. And if I do show you this next page, then you can see the Facebook state setting themselves as well. So each ad set accumulates data, so an ad set, not a campaign, so ad sets optimized on an ad set level needs each.
So each ad set there needs about 50 of the conversions it’s optimized for swear, that’s view, content, Add to Cart or purchase for Facebook to be able to deliver it with stability and efficiency. Now, as it says there and me, it says, needs about 50, the more the better the more data Facebook has to go on the better it’s going to be able to perform. So just give you an example, then if you have 10 and sets that get 5 conversions each in a week, I said this earlier, then you have 50 conversions, but because they’re not all on that one ad set.
Then all of those ad sets essentially won’t be delivering optimally. So what you’ll be better off doing is combining those 10 ad sets or combining them that budget into one ad set to make sure that one particular ad sess gets 50 conversions and that’s exactly what this testing strategy encompasses then. So that being said, then just two things you need to consider while going through this, because it will make sense, then why I’ve chosen these particular criteria.
So to begin with what locations should you target? I always recommend the big for english-speaking one to the UK. Us, Canada and Australia. Now you want to separate this per ad set. The reason you don’t want to put them all into one ad set is purely a couple of reasons really number one. The overall audience size will just be way too big. Are we absolutely massive and we don’t want audiences that are too big to start with I’ll get more into that later and number two as well? If they’re in separate ad sets, then you can make sure that you dedicate an equal amount of budget to each country.
Therefore you can, you can give each country its own fair chance, and then you can make a better judgment. The results won’t be skewed. Basically, if each country will have had say ten pound each spent on it. So if one brings in say five purchases and the other brings in two, then you know straight away which one is performed better, whereas if you combine them all into one and set them, Facebook is only going to give the majority of the budget to the best Performing one, so it won’t necessarily give each country its own fighting chance, so you won’t get an accurate representation in a which countries are the most profitable number.
Two then agent ginger to lead so leave this as default, especially if you’re going into an each-way. Not quite sure who your ideal customers are going to be so target anywhere from eighteen to sixty five-plus and then include male and female, no interest expansion. So that’s the little check box I’ll highlight it once we go through the creation process. Next point test: one interest at a time. So what we’re going to do is we’re going to use we’re going to flex target, but we’re only going to go kind of one level deep.
So we’ll have our base interest and then we’re going to flex it with another interest and that’s how we’re going to structure this testing strategy so we’re only going to flex with one interest at a time if that makes sense. Hopefully it does any questions at all. Obviously, do leave a comment down below, but all will come clear then, when I actually go through the process of Korean is strategy within the ad manager.
The next so emplacement again one per and set of four equal budget testing. So the two we’re going to focus on then is the Facebook newsfeed and an Instagram moving on into a daily budget. This is where things get interesting. So what makes this testing strategy unique, then, is we’re going to be duplicating the actual ad not duplicating that ad set not duplicating the campaign duplicating the actual ad and we’re still going to base it on the whole five dollars per day? So it’s going to be five dollars or just four dollars on there to make it a bit more clear times the amount of ads that you want to run.
So I recommend at least three, which will total a daily budget of $ 15. Obviously, the more the better and I’m going to explain why, later on in the article, because well to basically the more the higher percentage of an audience we can test, then the better results and more accurate representation. We’re going to get all over. That audience is actually a successful one or not, and the pros then, of having a higher daily budget and actually do Paquette do cocaine on the ad level is the number one we have a higher budget than everyone else.
Everybody talks about the whole five dollars per day, so if everybody’s doing five dollars per day and we’re spending at least fifteen dollars per day, then at the end of the day Facebook is a bidding market. So if we’ve got a higher daily budget, then we should get in theory, better results and everyone else, because we’re showing to Facebook that we’re willing to spend a bit more money and we’ll also be out bidding everyone else.
And if the majority of people test and a five dollar daily budget and we’re using fifteen dollars, then we’re going to be able to outbid people and therefore we should get the pick of the bunch when it comes to the customers. Also, we get better ad optimization going back to that original point of Facebook ads optimizing on an ad set level. If we’re spending more money per day, there’s going to be more data coming in for that one particular ad set, and therefore it’s going to optimize better and deliver more efficiently.
Moving on to the next point, there’s a seven-day click or one day view the fact that we choose seven-day click means that anybody who clicks on our ad and then purchases within seven days, they’re going to contribute and count towards those 50 conversions, whereas if it was Only a one day click then they click on your ad, but then they buy two days later. They don’t get counted towards those 50 conversion.
So therefore it takes longer to actually optimize totally optimize. The ad set, because they’re not included in that 50 of that makes sense and then the final point is our bid strategy which we’re going to leave a lowest cost and the reason we do this, then, as it says in brackets, use target cost when scaling is Because we’re testing we want to know, we want to see what the actual lowest cost is that we can achieve purchases for this particular product, because then that’s going to give us a gauge of whether it’s actually worth scaling and then once we do move into scaling.
We use target cost because then that way we can take to Facebook. This is how much we want to pay roughly per purchase. So, as we scale up, then try and stick as close to that as possible. So that being said, then guys that’s a strategy. Hopefully, it all makes sense. So what I’m going to do now we’re just going to go back to my ad manager account and I was going to build one out in front of you, so you can see practically how this strategy actually works.
So, to start with, then I’m just going to go ahead and click create new ad set looks like I was halfway through. So let’s click continue. In fact, let’s go right back to the beginning, so objective wise. Then I always start with conversions. Now, to be honest with you as long as you’ve got a strong product and you put it in front of the right audience, then you can always start with conversions, because with conversions comes everything else and what I mean by that, then, is that for somebody to Make a purchase on your store: they have to become part of your traffic if they have to view content, they have to add the car they have to initiate checkout.
So as long as you’ve got a strong product put it in front of the right audience, then everything else will come along with it. If that makes sense, so moving on to the actual ad set level, then your pixel name will be in there. Obviously, now the reason mine is called that is because I’ve got a few different ones. I just name them to make sense to me relevant to the right stores, conversion event wise now.
The reason that’s like that as well is because I I have a store that sells multiple niches, so I has set up different custom conversions, hence why it’s called that. So as long as you’re only advertising one niche at a time then just use the standard purchase. Conversion event moving down then into the first section, which is locations now. What I’m going to show you is setting up just kind of one ad set one particular answer, but if you want to get a bit more advanced and create multiple ad sets at a time, then what you do is you scroll to the top? Let’s get rid of this annoying message as well, and then you can click create multiple new ad sets and what that allows you to do is pretty much self-explanatory.
Just create multiple new ad sets and you can split test everything and essentially set everything all up in one go, but I’m not going to go into too much detail in this article, because I have done a specific article on that topic. Actually so for this example of this article, then we’re going to leave it as UK aging gender stays as default, and then this is where we get into the actual tags. In section of our strategy – which is this section here so one interest at a time and no interest expansion, so what that means, then is make sure you unclick that.
Therefore that means that Facebook is only going to target people that are included in our detailed tags. In Section, so what we want to do, then, is make sure they were only tagged in one interest at a time, but we’re going to flex it by one level to make sure that we’re targeting the most focused section of our audiences as possible. So, just as an example for you guys, if we’re in the dog nation, I always use it as an example on this blog.
What we’re going to do is we’re just going to use dogs as like the base interest, but, as you can see, the potential Reach is still 15 million people, which is actually huge. We want to make this a lot smaller, purely because if we’ve got a new pixel in a new niche, then for us to go out and advertise in an audience – that’s 15 million people and spend in 20 pound per day. Then, as you can see, our daily reach is anywhere from 1 K to 8 K so to spend 20 pound.
I’m not sure what that works out percentage-wise. But it’s probably like point zero zero one percent of this overall audience, so the chances of actually finding people who are going to purchase our product within that potential reach size is very, very, very absolutely tiny, purely because our pixel is not matured. It’s not optimized. It doesn’t know who our purchases are, so essentially it’s going to go out and show our ads to random people until people start to take notice a bit and start to engage with it.
So, to begin with, we have to flex down to make this number even smaller. So what I recommend, then, by flex, is by flexing by at least one level. So if we click on our audience, we can put another dog interest in here then to essentially bring this potential reach down. So I’m just going to use dog Today magazine because it’s one of use quite a lot and select that and see what that does to our potential reach up brings our potential reaching down to 61 thousand people wish you’re more than happy with.
I recommend, if you’re testing one interest at a time, then try and get your audience below 300,000 people just because that’s small enough to adequately test or get an idea, then of whether your product is going to work or not. Now a lot of people will say: there’s not enough scale, not enough room there to scale it and you’d be 100 % right, but what you’re forgetting is that to begin with, our pixel is not going to be very well matured.
So we have to do this to build up the traffic, build up the purchases, that’s going to mature our pixel and then once our pixel was matured once it knows what and who our ideal customers are, then we can just get rid of dogs. Today. Go back to 15 million people and Facebook will know who our ideal customers are. So now we can go out and actually find those people within this audience of 15 million people where’s to begin with, it won’t be able to do that because it won’t know who our ideal customer is so essentially, then we have to do the tags in for Facebook ourselves to begin with, until we’ve matured our pixel any questions on any of this at all.
Then obviously, please do leave a comment down below, so I’m just going to fill this back in again. Dogs today magazine now. The recent magazines work so well is because, for anybody to be subscribed to a dog magazine, they’d have to be pretty crazy if they don’t own a dog. So it pretty much just tells you straight away that people within this audience owned dogs and therefore they’re, going to be interested in a dog product moving down then into placements, make sure you hit edit placements and then start off by just getting rid of all of Them and then just select the ones we want now.
The two we’re going to be focusing on then are the Facebook newsfeed and the Instagram newsfeed as well. But what you’ve got to remember is that we’re going to split test these in two different handsets now. The reason we don’t combine them, then, is because Facebook is going to optimize our budget into the one that it thinks is working the best. But, to begin with, what I recommend is devoting the same amount of budget for each one, so each one gets its own.
Fighting fair chance to produce the results basically and then once you know which one produces the results, then just focus primarily on that one, so this particular ad set. Then I’m setting up for the sake of this article, it’s just going to be in the Facebook newsfeed. All mobile devices – that’s fine, now daily budget. What I’m going to do is, as it said in the strategy, I recommend using at least three, so I’m going to put the daily budget, then 215 pounds we’re going to use conversions when you use SEM and a click or one day view and we’re going to Go for the lowest cost as well.
So that’s pretty much our ad setup moving on to the actual ad section of this campaign, so now this is finally loaded up. Then what the default screen you will see, then, is this one which is the create ad screen, but what I recommend doing then, is using an existing post. Then that way, every time you run ads to a particular post. All that engagement comments. Tags are all going to build up on one post and therefore, when that accumulates you’ll build up the social proof and the more social proof you have, then the more likely people are to engage with it as well and actually click on it too.
So choose your ad, then that’s not the topic of this article, so I’ve done plenty of articles on ad copies and everything so go out. Do your research on that, but once you’ve got your ad done all set up, so this is just an example. One then, for this article go ahead and just click confirm place your order and then once this confirms and takes you to the next screen once it. Finally loads, so now this is finally loaded, then, as you can see, we’re back in my ad manager account on the ads level.
This is the one that we just created and what we’re going to do now, then, is just you placate this we can court. We can leave it in the same campaign, we’ll leave it in the same actor and let’s create two copies: click duplicate and then once it’s finally loads just go ahead and click publish, because essentially we want to keep everything the same. All we’re doing is duplicating the ad, so we’re running three ads under one and set.
Essentially so, let’s close this down so as you can see, then what we’ve got now then is we’ve got three ads under one ad set level and what this is going to do then, is it’s just going to basically allow us to test an audience more adequately Than just using one ad for a number of reasons, number one we’re using a bigger budget. Therefore, we’ve got better bidding power against other people. Number two I’d sets are going to optimize better than just using five dollars per day and number.
Three, then, by using three different ads, especially this works really well as well for larger audiences. Purely because, if you’re going out with one ad, then obviously Facebook is going to choose a certain segment of your audience to show your ad to. And if that certain segment doesn’t react well to your ad, then you might leave that audience thinking the audience doesn’t work and it’s over the audience or your product, whereas if you’ve got three or five or even ten different ads, Facebook is going to go out and Show those ads to five two three five or ten different segments of that audience, and then you can see which ads are returning the results and then once you’ve done that once you’ve, let it run for a day or two.
You can close down the ones that aren’t performing very well, and then Facebook will start to put even more budget into the ones that are performing well. So that means that the guys, that is the whole testing strategy from start to finish, I would love it if you guys actually go out, implement this and come back and let me know what kind of results you’ve been getting. I’ve been using this a lot, especially when scaling ad sets, and it’s been working really well for for me, in fact, I haven’t done a with proof article for quite some time, so maybe, if you guys want to see it, of course, I tend to go on Whatever you guys recommend, if you guys want to see it, I can do another with proof article.
I actually show you one of those ad sets where implemented this strategy, so you guys can actually see the kind of results I’ve been getting, and that being said, then, hopefully that made sense. Hopefully you guys are still reading. If you are, then, hopefully you enjoyed the article, otherwise you wouldn’t be reading this far and if you did then please do leave a like. That’s all I ask, and that being said, any questions at all make sure you leave a comment down below.
I assume we’ve covered quite a lot, so there’s probably going to be tons and tons of questions, but I do get back to every single person so feel free to spam. As many as you want down below in the comment section and that being said, then I’m going to stop talking now. It’s nearly. It is just gone three o’clock in the morning, so that means something guys have a great weekend and I’ll see you all tomorrow.
For Facebook, Ads scaling,
Looking for a company that will have your back? Check out the video below: