Our exact 9-figure Shopify + TikTok Ads testing strategy (internal SOP)

Hey, welcome back. This is Peter Salvo, and in this video, I'm going to show you the optimal ad campaign structure, optimal testing budgets, and the testing approach. So let me dive in with some theory and then I'll show you a live example of exactly how to set it up. It takes like five, 10 minutes. Anybody can do it, and it's pretty simple. So whenever I create a campaign for E. Comm, I typically do one campaign. Within that campaign, I do three ad sets. And within each ad set, I put three ads. So nine ads in total. Now, budget at the bare minimum, I typically start with 20 a day per ad set, meaning the campaign overall is at 60 a day. Of course, the bigger you get, more advanced. You can start with bigger budgets. And it is recommended you can gather data faster, make better decisions, find winners faster. Now, the basic things you need to understand and remember is the three different, different layers. So the campaign is all about the objective. What's the goal that the pixel is going to optimize for? Is it engagement? Is it link clicks? Is it website conversions? And the answer For e comm 99% of time is website conversion, specifically website purchases. Now, the ad set, all you need to remember is that's where you define the targeting, meaning the audience of people that you're going to go after. The ad is the ad itself. So meaning creative and copy. Creative is the image or the video, and copy is the text within the ad. Now, whenever I set this up again, I do website conversions to purchases. Website. Within each ad set, I test one variable at a time. For Ecom, I like to keep one blank, meaning no targeting. Let the AI, the pixel do the heavy lifting. And then in the other two, I put one interest each. So something related, what I think could be relevant to the audience, that'll help me kind of zoom in on certain type of people. And I put one interest per ad set and then the ad. So typically, the biggest differentiating factor is going to be the creative, meaning the video or the image that you use. So I figure out a good copy that works good, call to action, and then I just switch out the creative. Now, at each time, you want to test one variable per section and at a time. So instead of doing a completely different ad, different video, different text, different cta, I like to keep the text cta, the same call to action, that's just the button at the bottom. And I just test a different video because that way if we keep certain things constant and we just change one thing, that can really let us understand which video is Performing better. Otherwise it's going to be a mess. Right. Otherwise you're not really going to know. Is it the different copy causing the change? Is it the different video? It'll be hard to know, but pretty much impossible to know exactly. So we want to keep it scientific and we want to keep it simple. So now that we have this basic stuff covered, let me give you a real example and then I'll go more into the budget. So inside your dashboard, click on campaigns. In here, click on Create. Now you want to click on website conversions and I typically turn this off. Now the campaign name, I typically just put the product name and the country that I target. Click, continue. Rest is fine. Now ad group. So here I put the name of the targeting option. So we'll start with no targeting. That's what I'm going to name it. Here you're going to put the pixel that you already created, you already connected it to Shopify and click the event. Complete payment. Now I typically like to do automatic placements targeting. So US age range is fine, typically 18 plus. I like to do both genders. Now if you're selling, you know, product very specific to women, then you can just target females, vice versa with males. Languages is all spending power. Typically I like to start with all women. With certain higher ticket items you can test it out as well. One more pro tip, there's no right or wrong here, right? Each and every thing we got to test. Like scientists, nobody has a crystal ball that'll be able to predict what campaign settings and what creative is going to work. Right. Your job is to go out there to the best of your ability, put together the campaign with best practices and then you have to test and some campaigns are going to fail. Completely normal. It's a part of the game and, and some are going to be successful. Now pro tip number one here. If you're going to target different countries other than the English speaking ones, Canada, UK, US etc. Like say you target Germany or Austria, you want to set the language to English, right? Otherwise some people might not understand the ad and you're just going to waste money. Now this is really where you put the targeting, what's relevant. So right now I'm going to leave this blank because this is going to be the no targeting campaign where we let AI try to figure out who are the right people. Here's another pro tip. The more you spend on this, the more you train and season the pixel, the more it's going to understand who the right buyers are. So this is also one of the reasons you need to spend a little bit of money to build momentum, get results because you need to give it data. Once people start buying it, the AI is going to understand like oh, these people with these demographics and character traits and psychographics are the ones most likely to buy. So it's going to streamline it and try to run it more towards those people Device you don't have to worry about Just leave it as is daily budget output to 20. Since we're setting the budget on the ad set level. Right. So you're going to see this is the campaign level, ad set level and underneath it is going to be the ad level which I'll show you in a sec. All day conversion, Target CPA typically leave blank. Later down the road more advanced you can try to do Target cpa. You don't have to worry about it. Click continue. All right, ad name. So typically I put video ad number one or if you know like a, you know, keyword or some sort of label like you know, oh this is influencer one video ad. This is more like a presentation ad. You can put it there. Again these are just labels for you to remember and distinguish which is which. Right. So again not a nobody's going to see this. No end user is going to see whatever name or targeting naming convention that you use. All right, you select your page. I typically split this this as well. So sometimes I do Spark ads and then I duplicate the entire campaign. Don't do spark ads. I like to leave it on typically and start with Spark. Now single video I'm just going to add from library. This is where you upload your video ad. Again this is an example campaign just to show you the structure. So it's not going to be an actual video ad for the product and then say we run with this. I'll just copy the URL again. This should be the proper.com domain. You see this is a subdomain since this is just a test store. So the link is going to go here under the URL. Now CTA's I like to select just a couple or just one. Check it out is a good one. Interested? You want to keep a low friction. You know you don't want to put stuff like order now, buy now because some people will actually think, especially older generation, like oh my God, it's already going to order the product itself. So keep it low friction. Just check it out, interested, look at it, stuff like that. Now for copy, there's many things you could put from a discount to scarcity 50% off today. Whatever. I like to just take one of the testimonials and reviews that's posted here. You can either take the whole thing or shorten it and then just put it here. So I'll take this and boom, there we have the text and it says, check it out. That's the video ad, all right. That I typically leave unticked. Let me just double check everything. All right, looks good. So now we have the campaign almost done. Right? So if I go back over here, what you're going to want to do now is click duplicate and duplicate this twice. All right, so we have video ad number two, number three. And you're gonna notice I'm gonna leave everything the same. I'm just gonna put a different video ad here. Alright. And here to change the video, you're just gonna click on update. You either upload or you already have it up there. So pick the different video. And then video ad number two, same thing. So update and just put a. Perfect. All right, so notice I kept everything the same. The product links the same, the text is the same, the CTA button that says check it out is the same. At least it should have been the same. Let me fix it. Okay, this is good. Let me make sure this is good as well. Yep, that's good as well. So now we're almost done. Now you want to click on the ad set level, duplicate twice, make sure this is not ticked. Right? Because you do want to copy the ads over. And so now we have two other assets created. So now you scroll down here and now you're going to try to find something related to the product. So say this is a calf massager, compression people have pain, etc. So you want to find maybe beauty and personal care. So maybe this, we just keep that like that. And then you can copy, paste and just change it to beauty and personal care. Let's go over here and try to find something else. So we can do purchase intentions as well. Or maybe in this case baby, kids, maternity, it's probably going to be females, maybe 30, 40 plus that buy this stuff. So maybe you could target moms through this. Targeting again, there's no right or wrong here, right? So you want to pick certain audiences. Some of them might be also indirectly related, right? So you got to think about like these people that you're going to target. How old are they, what kind of magazines they read, what kind of movies they watch, you know, all these different things, who do they listen to, who are the thought leaders, what, what kind of books you know, are on their shelves, stuff like that. And then you can try to target these people. So there we go. So essentially in a couple minutes what we set up is we have a campaign with the website conversion goal set to purchases. We have three different ad sets targeting one different variable. So here we have beauty, personal care, baby kids, maternity and no targeting. And then within all of these assets we have the three identical VADs or video ads, or let me correct that. So we have three identical ads. So these are the same per ad set because of course we want to figure out which ad set is performing the best. And then you also see there's certain combinations. So certain ad, if you target it to two different sets of people with one audience it might not work, but another audience it might work. Right. So we have nine different variations here and this is pretty much how a campaign should look like. So I hope that's helpful. And you see that once you do this once or twice, it'll be very quick, very easy to do. And then here's the beautiful thing. When you click publish all and it's out there, I'm not going to do it because it's only a template. But essentially once you're in here and you have these campaign shelves set up one campaign, three ad sets, nine ads altogether, then all you have to do is click here and let me try to find the button of copy. Right, so now just going to copy the whole thing and then very quickly you can flip out products and test different ones. All right, now let me dive a little bit more into the optimal testing budget. So technically there's a couple different ways to look at it. It should be 2 to 3x the product price or it could be 3 to 5 eggs the target cost per acquisition. Now to keep things simple, since we're starting out, let's just say it's 2 to 3 the product price. So if this one's 50 bucks, then you should spend 150 on that ad set to figure out if that ad set is going to work or not. Okay, so let me maybe clarify here that this is on the ad set level. Okay, so kind of like the example I gave here. If it's a $50 product, spend at least 100, 150 per that ad set to figure out if this is going to work. Well now if initial results look promising, meaning your click through rate is above a percentage or even higher, CPC is low enough and the math looks good, then we might want to spend more on it since variation A might be breaking even or losing Money, but variation B could be profitable. So this is one of the things I was saying earlier is say you have five campaigns. Now one target the usa. Campaign B has everything set up identically, but is going to target EU countries. Campaign C, same thing, but target the United Arab Emirates. Now you might find that USA typically the most expensive country to target. Maybe you're going to lose money, maybe break even, but in EU you might be making a profit or if not in eu, you might be making money in Dubai or the uae. Okay, now there's another variable besides country is say you targeted us identical link, but you test a set of different video ads and a certain set of video ads might be profitable, others might not. Or say you test the same video ads but you create a different product page that's more optimized. You create a pre sell page and it goes there and maybe the cost per click and CTR and all those things are the same. So you're paying the same ad cost, but you have higher conversions on a different product page and it might make the math work in your favor and become profitable. Okay, so you're looking for a winning combination. There's all these different variables that we combine and one of them might work while the other ones might not. Now if we see from the get go the math is bad, right? It's not, not optimal to begin with. Instead of spending more money to test these variations, we're just going to switch to a new product. There's still a small chance that a different combination might make it work, but the likelihood is kind of low. So what I mean by that is if you're paying like five bucks per click, CTR is really bad. Nobody's buying. You know, you're better off just testing a different product and trying to figure it out and find a way to make it work. Now last thing I wanted to show you is I'll put a sheet below that'll help you keep track of these different variables. Right? So this is kind of like a variable comparison sheet. So you'll be able to put the product here and then per each product, you'll be able to select different variables. So say you can try to sell it for 69.99 or 79 or 59 and then say the country goes to us, you do no targeting, you do a different angle. So an angle is how you approach it, right? So say if we're selling this calf massager, then angle A could be relieve your pain. Angle B could be give your leg a rest. Angle C could be you know, gift for mom, right? So it's the same product, but it's sold from a different angle. And certain angles might resonate better with certain audiences, right? Different ads, you can test different upsells, different funnels. So maybe it's product page A, maybe it's product page B, maybe you do a pre sell page, maybe you do a click funnels page, right? And you're gonna get different conversion rates. And then in the sheet you'll be able to track how much you spent, what's the CVR conversion rate, what's the average order value, what's your profit, what's your roas? And you'll be able to find winners. Now this is kind of like a pro tip. You don't need to use the sheet, but if you really want to take this to the next level and be scientific about it, this will be helpful because of course you'll be able to see this data inside of TikTok as well, especially if you do your naming conventions. Well right now in this case, if you want to keep all of this data within the naming convention, you're going to have a really long campaign name, right? Product name, price, country, targeting options. So it'll be a little bit too long. This is why we like to keep it in a sheet because it gives you just better overview. So hopefully this was helpful. I know I covered some more advanced stuff and I know some of the later things I discussed might be a little bit more confusing. But look, the main lesson in this video is very simple. This is how an optimal campaign should look like. And of course, if you're asking, can I have four or five ad sets? Yes, you can. There's no issue with it. Can you have more ads than three in an ad set? Yeah, I typically recommend three to five ad sets. Three to five ads within an. Excuse me, three to five ads within an ad set. So yes, you can have more, yes, you can have a higher budget. But just to keep things simple, this is an optimal beginner friendly setup. It's perfect to start with. And then the second main thing is just understanding these different layers. Again, you have the campaign underneath it. You have ad sets that set the targeting the audience. And then underneath each asset you have the ad itself, which is really consisting of the creative and the copy. So hopefully this was helpful. It gave you insight into how it works and you can see that even as a beginner, you can begin setting these things up within just a handful of minutes. So I'll see you in the next one.