[0:00]What's up guys? In today's video, I'm going to show you exactly how you can increase your conversion rate. This is going to be the 2025 CRO Guide. Now, before I jump into this, let's just talk about why CRO Conversion Rate Optimization is important.
[0:12]CRO is maximizing the efficiency of users that you're paying through your paid ads, whether it's Facebook, whether it's Google, it doesn't matter, the amount that you're paying to get people to your landing page for the overall efficiency of what you're going to have with that page.
[0:27]So, how many people can you get to convert and how can you maximize the amount that each person pays when they land on your page. So, what I'm going to break down in this video is the exact strategy that I'm currently using across my multiple eight-figure e-commerce brands, that over the last three years alone, I've been able to do $60 million in sales with.
[0:44]This year, I'm planning to do over $50 million by year's end. We'll see how Q4 goes on that. And I have absolutely nothing to sell you. I'm doing this all for free. All because I'm trying to build a personal brand right now. I'm trying to make cool connections, network with different people.
[0:58]So, I really would appreciate a like and a comment if this video shows you some value. That's the only currency I'm asking for. Now, to jump into what we're actually going to cover in this video is number one, we're going to go into my overall conversion rate optimization philosophy at a very high level.
[1:13]So, I'm just going to show you guys how I look at different types of tests from a bird's eye view and the different ways we can approach it. Number two is going to be the exact strategy I'm currently using right now to do conversion rate optimization, and I'm going to actually walk through some live examples looking at different pages where we can then determine what to test in a real life example so that you guys can really start to see how you can apply it for your own store.
[1:37]And number three is going to be a tactical step-by-step breakdown of how to run these tests step-by-step. So, after this video, as I go through everything on this list, you guys should have the full blueprint to be able to go and start running proper split test on your existing store right now.
[1:54]And whether you're just getting started or whether you're already scaling up, this should be able to make you a lot more money if you do this correctly. So, with that being said, let's jump right into it.
[2:04]Okay, so what's important to note is that the way I look at it is that there's macro and micro split test. So, the differences between each of these and let's use an example with a little bit of a beautiful illustration that I put together over here, where let's say we wanted to do a split test on an advertorial page, right?
[2:20]So, the difference with a macro split test would be, you would be completely changing this advertorial altogether. So, you take it would be a completely different version of that page, where you rewrite the entire copy on the page, whether you reformat and redesign the entire design and the way that the information is presented, okay? So, that's on this side.
[2:40]Now, a micro split test, the way that I look at it is I look at it as different building blocks. So, as you can see over here, this would be, let's just say the above the fold section. This could be the section where they're really describing the problem on the advertorial. This could be where they introduced the product as a new solution. This could be the call to action.
[3:00]So, the way that I look at it is I like to group things into different building blocks and decide on which of these building blocks I would want to test. So, that would be more on a micro test.
[3:08]Now, there's two big differences of how I would look at using each of these. So, macro tests are typically done when you're first gaining momentum and gaining traction as a necessity, because at that pace, if you're below 10,000 monthly visitors, then a macro test is, so first let me define, I guess, what the big differences are.
[3:31]Let's grab this little pen feature. So, for a macro test, what will happen with a macro test is the, if you have a successful macro test, the chances are you're going to have a very high return on your conversion rate.
[3:45]So, the, let's say you have a 1% conversion rate, you scrape the entire advertorial, you rewrite it and it actually hits, that might get you to a 3% conversion rate. You can literally three X your conversion rate for doing that, but here's the thing.
[3:57]The probability of this working is much lower. So, getting a macro test to actually hit and convert, the chances are very low. So, in essence, the impact is very high and the chances of it working are super low.
[4:10]Whereas a micro test, if it does hit, the impact will still be good, but it'll be much lower. Let's say we use that same 1% conversion rate example. If you can get a 10% boost, which means you go to a 1.1% conversion rate, that would be pretty good for a micro test.
[4:26]If you can get a 20% boost, that's even better. So, like within the range of 10 to 30% is typically what you can see with micro split test, which means that you're going to get solid returns in your conversion rate or efficiency, but it's going to be lower.
[4:39]But the chances of it working are much higher. So, if we look at this compared to this one, is that you're going to, the goal with micro split test is that you get consistent gains with high chances that it's going to work if you do high conviction based tests.
[4:55]For example, if we look at this advertorial example, the chances of me going and then scraping an advertorial, coming up with a completely different concept and then getting it to work is relatively low. But if it does work, the impact of it can be monumental.
[5:08]Whereas on a micro test, again, the chances of this working, if I say, okay, let me change the above the fold section and change the headline to be more optimized, make it that you can see the full GIF, the headline, all that stuff or maybe the intro image, whatever it may be at the above the fold section, optimize all that.
[5:27]Chances of something like that working with high conviction is there's a pretty good chance that it works. But the impact of it won't be that high, but the chances of it working will be much higher. So, the way that I look at using each one of these tests from the toolbox of like how I would pick which one to do, micro split tests are when you're getting usually above 10,000 monthly visitors.
[5:45]Like ideally over 100,000 monthly visitors, you can really run these much more efficiently, but bare minimum, you should be getting at least 10,000 visitors to run one of these tests properly, because you have to remember that in order to actually get the data on this, you need to split this traffic in half.
[6:00]So, if I want to run a micro split test on an advertorial that's getting 10,000 visitors, I have to send 5,000 to variant A and then 5,000 to variant B. So, this would probably take me a month maybe or at least a few weeks to be able to get the proper amount of data, but you can still start to run this once you're above 10,000 visitors.
[6:18]Where this is where I'd run this one on below 10,000 visitors, because I don't want to have as much traffic. I just want to make some monumental shifts, especially if I can't scale because my conversion rate isn't high enough, which this could likely be a variable of that.
[6:31]So, that's like the main way I'm looking at it. But now, I don't want to say that macro split test are out of the question when you're already scaling, when you're already above 100,000 monthly visitors. It definitely does play a factor, but usually at that point, I won't run a macro split test on the main funnel that I'm using.
[6:48]If I do run a macro split test, I'll run that as a separate campaign, not as an AB test against my control variant. So, usually I'll just send separate traffic through like a Facebook campaign or something like that, just because there's too much risk that you're going to lower your efficiency that at that level is just not worth the risk.
[7:05]So, now that you understand hopefully the philosophy between macro and micro split test, what we're going to be focusing on this video is going to be mainly micro split test. And the reason why I want to focus on micro split test is because this is where you're going to get those consistent gains and higher returns and over time, if you keep on having winning test, then that's going to compound very nicely.
[7:25]So, let's go into what the strategy is now for micro split testing at a very high level. First thing is the goal is to identify what is the biggest bottleneck in your funnel. So, before going into the split test, you have to be able to determine what are we actually trying to solve for?
[7:40]Where do we have the biggest drop off and what metric is the biggest drop off? Is it, which is point number two, what metric are we lacking on? Is it conversion rate? Is it another variable there where maybe it's the LTV and the amount of customers we're getting into a subscription program?
[7:54]Those are different things that you can look at across the board, but that's what you have to first understand these two things, otherwise you won't have a direction of what you actually are trying to accomplish from making these split tests happen. So, once you have that, though, then you have to figure out how do you get the right idea of the test that can actually improve that metric?
[8:06]So, once you identify this is the metric that needs to be improved, now you have to figure out how you're going to consistently get the right idea, which I'm going to show you guys right below the exact process I use to get there. The next thing is then test based on the blocks. So, pretty much what I said over here.
[8:21]We're going to talk about how can you group these blocks and how can you determine which block to test? And then how to actually set up a proper split test will be the next component that we go into. And from there, like what will either happen is one, you will get an improvement like we said.
[8:35]So, either it goes up or maybe it goes down for some reason and it doesn't work, which is definitely possible, and it will happen from a certain percentage of the test that you do. So, in that case, you are learning something. If you set this up correctly, then you're learning on what you could apply on the next test and what you should refrain from doing if this doesn't work as well.
[8:57]So, either way, it's a win because one, you improve it, which is obviously ideal. But two, if you don't, then you're actually ideally getting learnings like it's almost like a scientific experiment. And then from there, your job is just to pretty much repeat this every single week.
[9:09]Once you get enough volume and enough traffic, if you're at 50,000, 100,000 monthly visitors, you're going to have enough volume to be doing this on a weekly basis, where now you can really start to see the compounding of this.
[9:22]And once you get above 100,000 visitors, then you could even start running multiple micro split tests at the same time, as long as you're separating out the traffic, which we'll touch into in a second. But that's the overall high level strategy of this. So, now let's start to break down the actual tactical step-by-step guide and through doing this, we'll go into some examples that I've already linked into here so that we can look at this in real time so that you guys can apply it.
[9:45]Tactical step by step guide. Let's start with the first part of what we spoke about, which is goal is to identify what is the biggest bottleneck in your funnel. So, this is where you have to do a data review process. So, the goal with a data review process is to understand where is the biggest hole in your funnel.
[10:00]So, what I did over here, but what you can see over here is this is, let's say you're using an advertorial, maybe you're not using an advertorial, so this could start at the PDP, but the way that the traffic works is that whatever your first page is, and whatever sequence you have.
[10:11]So, if it's advertorial, this would be the first page, if not, it's your PDP product page. So, whatever that first page is, is going to get the most amount of eyeballs on it, the most amount of visitors, just because if you're sending your ads to this advertorial page, then you're going to get, let's just say 100 clicks to your advertorial.
[10:30]But likely, you're only going to have like 30 people reach the PDP. And then from the PDP, you're probably like going to only have maybe 10% of those people actually go to checkout. And then from the people that check out, you might have one or two of them actually convert and then go to your upsells.
[10:45]So, starting from top down is usually where you're going to be able to get the most benefits of improving whatever is the biggest bottleneck there. So, the different metrics to really start to understand where those holes come from is from the advertorial, what you can look at is your pre-lander CTR.
[11:00]So, out of people that land on your advertorial, what percentage of those people are actually landing on your product page? And I would say a good rule of thumb here is anywhere between 20 to 30%, 20 to 30% is usually a good range.
[11:22]Obviously, the higher, the better, but this would be like from what I've seen with different advertorials that I've ran, this is usually a range where if I'm in there, it's pretty healthy. Now, if it's below that, and obviously, take this still with a grain of salt, because what actually matters here is your overall conversion rate on the funnel.
[11:30]So, overall conversion rate is just tracking from people that drop off and land on your advertorial, they come off of Facebook ad, they land on your advertorial and then go to checkout. How many people are actually buying from those initial visitors that you saw on your advertorial page?
[11:44]That's how you track your actual overall conversion rate, which is the metric that matters. And if this metric is good and if you can scale your ads, then this doesn't actually matter. Even if this is 5% on your pre-lander CTR, but you're still able to scale it and you still have a good enough conversion rate, doesn't actually matter as much.
[12:01]This is the Northstar metric I would look at from a conversion standpoint is what is our overall conversion rate and then the other metric is what is your overall average order value? I'll get into the EPV in just a second. So, I'm not forgetting about that.
[12:27]If the conversion rate is low, that's when I'll start to look into things like the pre-lander CTR. And if that is within the range that I deem to be good and especially if I have previous data of times that I've scaled and it's within that range, then I'm going to start to look at what's my PDP, which is your product page CVR.
[12:42]So, what's my conversion rate of people that land on the product page? Or if that isn't good, because my add to cart percentage below 10%, it's really about identifying where is the biggest gap that you have in your overall sequence. And then based on that, that will tell you exactly what you need to optimize.
[12:59]If it's your pre-lander CTR, then that means you're going to have to do a micro change somewhere on this page, which we'll get into in just a second. Now, if it's somewhere between your add to cart or your PDP conversion rate, if it's your add to cart, it means maybe your offer isn't compelling enough, maybe your price point is too high.
[13:17]Maybe it's just the overall look and feel of this page, right? Those can all apply to the add to cart/initiate checkout PDP conversion rate. That's a combination of all those things that I just said. And also, you can't look at everything so individual either.
[13:30]Like you have to remember that this is the metric that matters, and sometimes you can get a little bit too zoomed in, where you say it's all about fixing your pre-lander CTR. But maybe put it like a sticky add to cart and yeah, you'll get more people to click through, but your conversion rate is going to dip much more in this page.
[13:46]So, you have to remember it's a blend between all these different numbers. But ultimately, this, the point of this is that you have to look at this data to see what then needs to be optimized. Now, from an AOV perspective, and by the way, guys, I have a full funnel breakdown video on this, where I go like super deep in detail on this.
[14:02]But this, I want to be more focused on conversion rate optimization. But just so you guys understand exactly where I'm coming from and what I'm trying to talk about, AOV, I usually look at it from two different sections. So, number one, I look at pre-purchase AOV. So, that's all the stuff that happens before people actually check out. So, everything over here.
[14:19]And then I look at post-purchase AOV. So, after checkout, this is like the upsell flow, downsell flow, what's going on over here. Now, there's ways to track all of these metrics. So, for example, if you have ways on the pre-purchase could be bundle framing, it can be your price point overall. It could be things.
[14:35]Pre-purchase AOV is a little bit more risky, because those things if you want to optimize it, it likely will also affect your conversion rate, because remember, it's all happening before someone actually goes and checks out and purchases. So, anything that you try to optimize on that section could affect your conversion rate.
[14:53]Whereas on the post-purchase AOV, but on the post-purchase AOV, this is all your upsells and your downsells, right? So, these are things that you've already made the purchase, so it can affect your conversion rate. So, usually this will be easier to optimize. But again, using this scale of chart, this is going to be less traffic, right?
[15:07]These are just going to be your buyers, which is going to be the bottom of the amount of traffic that people see. The post-purchase stuff is going to fall within this section, which means more people are going to see it, which means if you do crack it, it can have a bigger impact. But also you are risking your conversion rate. So, for tests like that, where it can affect both of those, that's where the EPV comes into place.
[15:27]So, this is earnings per visit, is what that means. And all this metric is you're taking your AOV and you're multiplying it by your conversion rate. So, if my AOV is 50 bucks and my conversion rate is 1% to say, which would be pretty shitty, then my EPV in this case would be freaking horrible. But this would be 0.5, right? So, this would be like 50 cents EPV.
[15:55]In this case, both of these probably need to be improved, but I would have to determine what would need to be improved more. And I would probably say the conversion rate at 1% would be way too low, and that would be what I would try to improve first and foremost. Now, in the cases where this really matters is if you're going to do a price test, which I'm not going to go into, because I have a video that's fully dedicated to price testing split testing on my channel.
[16:14]I posted it a few videos ago, so you guys can watch that after this. But that's where I would use that metric. Now, point is, this is only point number one, guys, so I'm going to try to move through it. But hopefully you guys understand the gist of it.
[16:25]The point is, you have to do a data review to understand what metrics are actually broken, what is your biggest bottleneck in your process and what metric is it, is your conversion rate, is it your AOV, is it your earnings per visit? And then understand from what sequence, what are the things that are broken? All contextual to where the traffic is being sent.
[16:40]So, this is a different point on this, but it's a similar like philosophy here. So, you have TOF, MOF and BOF. So, what that means is TOF is top of funnel, MOF is middle of funnel, and BOF is bottom of funnel. So, the way that I'm looking at this is the same way that this sequencing works, it's the same thing with your Facebook traffic.
[17:03]So, your top of funnel audience that might be watching an ad and going to a page, where this one might be your warmer audience that's going to a page, and sometimes when you're running different pages. So, for example, you might send traffic to an advertorial on Facebook, and that's where you're sending all your cold traffic to.
[17:20]So, a lot of times your conversion rate will be lower because it's cold traffic. Whereas you might do a retargeting or a middle funnel campaign where it goes directly to like your product page or your sales page, and that conversion rate might be much higher, but it's a smaller percentage of people that go and it's warmer audience that have more context that are going there.
[17:35]So, this is also important, just side note to mention when you're looking at these different metrics, especially if you're running multiple funnels, to make sure that, you know, everything is within context, right?
[17:46]You can say what's a good conversion rate, what's a bad conversion rate, but that's all relative to what's the warmth of the traffic that's coming on, where is that traffic coming from? So, that's just a reminder to make sure when you're looking at all these metrics, you don't take any one of these metrics too seriously.
[18:00]At the end of the day, it's about performance. If your performance is good, then this is all less relative. If your performance isn't as good or if it's struggling and you're backtracking to one of these, this is just a really good indicator of what might be broken and where you can start to try to optimize and fix it.
[18:13]Decide what needs to be fixed based on all that data. So, now going into point number two is let's talk about the actual building blocks of a page. So, we can start by using an advertorial, for example, which I pulled a few of these over here, which we'll look into in just a second. But going into these examples.
[18:30]The point is, this is how you can first get more clarity. So, this is all your data. Now, let's say we've identified after looking at everything that the main thing that we want to fix is the advertorial page. So, a way that you can get further clarity, especially if you're in the earlier phases of testing, I highly recommend this is setting up a heat map software.
[18:47]So, you can use Hotjar, you can use heatmaps.com. There's different softwares that you can use for this. So, what you do with this is, I'm sure a lot of you guys are familiar with it, but for those of you that are not, you pretty much can paste a line of code over here.
[19:00]Where on your advertorial page, you can track two very valuable things. One is you get a heat map, which shows you exactly where people are clicking, where is the most frequent clicking happening, and also you can see the scroll percentage. So, how far down is the average user scrolling?
[19:15]Where is the biggest drop off points happening on this page? So, this is going to help you further identify, going back to what we were talking about with different building blocks, where is the biggest drop off again on this individual page? This is how you can get even closer data when we're trying to zoom in to the page.
[19:30]And then another thing is just surveys overall. So, you can have surveys directly on a page. This is something I don't do as much as because it can affect conversion rate, but we do run it in different spurts of time, where we can ask questions like, what is stopping you from buying? What is stopping you from clicking to this next page?
[19:48]So, this is usually more relevant on a PDP, but this is something where you can get actual feedback from customers with the easy one, two click to start to understand as the data compounds, what are the biggest things that are the friction points on this individual page?
[20:01]So, this is how you can take that data and now go deeper once you've narrowed down exactly where you want to optimize. And then based on all of this, you can start to just use some intuition and kind of if you're a good marketer, looking at the page, seeing what can be improved.
[20:16]So, again, using advertorial for example, this will be different for whatever page, but the different building blocks, for example, an advertorial, there's going to be the above the fold section. So, again, the above the fold section is everything that shows up if you click on a page, before you have to scroll at all. So, that's everything that showed and this is some of the most valuable real estate on any type of landing page.
[20:36]Because this is the one that if someone lands on that page, 100% of people are going to see this part of the page. So, getting this part dialed in is the part that can have the biggest effect to actually getting people to read, continue on to the page, and ultimately try to convert.
[20:51]So, that's why this is so important. Bridge paragraph. This is where you're now segueing people. So, this a lot of the times will go into what's important, what are they about to learn, what are they about to read? The problem section, so this is really where you're building up the pain, the emotion, all that kind of stuff.
[21:03]The product breakdown. This is arguably the second and sometimes first most important section, where now you're really selling them on your product, the benefits of it, all that kind of stuff. And then lastly, the offer CTA, which is always super important and especially on an advertorial of what are people going to be getting, incentivizing them to click and improving that CTR.
[21:24]So, with that being said, let's go into a few examples. So, here is one advertorial. Let's pull that up. And we're going to go into mobile view over here, just so that we can look at it.
[21:34]First piece over here is like I said, this is the above the fold section. So, to be honest, I think this above the fold section can be optimized, because I can't see this image. So, something I might have an idea on this above the fold section, because this is everything that shows above the screen, is maybe shrink the size of this headline.
[21:50]Move this image up closer. I like that they have a picture of the person that's actually writing this. That's helps with credibility. How my sleep apnea went from bad to terrible. A wake up call for people with sleep conditions. Improve it. Maybe lower this type of section overall. Maybe turn this from a regular image into a GIF. That might be something that I look at.
[22:11]So, these are all things above the fold that if I optimize this correctly, could have major impacts on the overall section. So, that's going from that page. Now, let's jump back into this guideline so that we're looking at everything in accordance, right? So, that's example of above the fold.
[22:23]Now, what about the bridge paragraph over here? So, let's jump right back into here. Let's look at the bridge paragraph over here. So, this could be, that's what happens when you haven't had a proper night sleep in months. This is first showing this. I woke up upside down with the worst headache of my life.
[22:36]Through my cracked windshield, I could see my car had rolled down in an embankment. The last thing I remember, fighting to keep my eyes open behind the wheel. Again. So, this is like the bridge paragraph. This is getting people interested based on this headline that they just wrote to now actually read.
[22:53]Now, this is the problem section. So, for example, like, I can go into details of what I would do to change it, but the point is, I could pick this as a section, narrow that down as my building block and say, I want to redefine how I'm showcasing the problem, or I want to redefine how I'm showcasing the bridge.
[23:07]Then eventually we get into the actual product section. The my simple solution. So, how are we describing the product? What are the testimonials that were showing? We can group these all into one building block instead of keeping that separate.
[23:22]So, hopefully this is starting to make sense in terms of how this all comes together. The point in all this is like what I'm trying to do and identify with, let's say that was our advertorial, what I'd be trying to do with these different building blocks is figure out what are parts that I can group together into its own block, because this is going to be what I'm going to say, all right.
[23:37]This is what needs to be hyper optimized and what needs to be the next steps of how I go about just optimizing that page. That's where I'm going to look at these different building blocks is a really important note.
[23:47]And the same thing applies. I know I'm showing you guys on an advertorial page. Let's go to a product page and just for now, let's keep it on the desktop version. So, I might say I want to optimize the image gallery is one thing, or I might want to optimize the product title, or what I'm saying over here.
[24:03]Maybe I want to optimize all of this section together. So, that could be something there. Now, I could look at these different building blocks and define this as the benefit section of just how I'm showcasing the benefits.
[24:14]Maybe I want to switch out one of these benefits or multiple ones of these benefits. So, a lot of different things that I can start to look at changing on there. This is a really common section you'll see this on a lot of different pages. How we compare. So, compare us versus them type of section.
[24:28]So, you can start to define and categorize all of these different sections, no matter what type of page you're looking at, and be able to group that into those different building blocks to actually be able to make those tests. Now, something I want to mention with building blocks, which is an important note here, is that the trade-off often becomes how much you want to group in one building block.
[24:50]So, let's go back to the above the fold section over here. So, what I could say for this above the fold section is maybe I just want to change the headline. So, that's going to be super straightforward, and it's going to be much easier to understand what the impact is of that test I did in that section.
[25:03]So, if I just change the headline, then if the results improve, then I know I can track that back to me changing the headline and the new headline driving the results. Whereas if I were to change all of the stuff that I mentioned before, including making a GIF over here and changing another part here where we change the title of the person and reshaped the whole outline of the page and move all those things around.
[25:25]So, then it becomes much harder for me to get all the proper learnings from it, because I'm not going to know which section it is. But the trade-off on that is that if that does work, the odds of me getting better results, going back to the macro versus micro philosophy, the same thing applies on an individual building block.
[25:43]The more things that you change, the more impact you could have, but also probably the less chance that you're going to get better results, and also you're leaving out some potential learnings by taking that route. So, that's just something to note with these different building blocks.
[25:57]And oftentimes, I think it comes down to just like the this next part I'm going to go into in terms of how to determine what to actually test. That's really what I'm talking about in this section over here. You have to decide, do you want learnings or do you want result?
[26:10]And often times, I think it's better to try to get results in these types of sections, especially if you're on the lower end of traffic. Now, the bigger you become, the more you should start to prioritize learnings and safety, rather than going more tests into one.
[26:23]If you're below a million visitors a month, then I would do more of what I said with the changing multiple elements within a different building block, so that you can try to maximize the results of a test, especially if you have high conviction on different elements, right?
[26:38]So, that's what I'd recommend. Now, smaller tests with less variables will give you better learnings, like I just said, but slower progress. And bigger mashed up tests will improve results more, but harder to identify what makes those improvements.
[26:50]That leads us to point number four, which is once you start to identify all of these things, you should make a general list of what will make the most impact. This is a list that I've put together of what I've seen usually get me the best results.
[27:03]So, first and foremost, it's your actual offer and the offer presentation. So, this is always the first thing that I optimize, because everything stems from this. The offer that you put, for example, on the advertorial, that will affect your product benefit section, right?
[27:17]Because if you're talking about your product, that's all going to stem from also what are they getting at the end of it? What are ways that your full offer can help solve the overall problem that people have. It also will impact the amount that you can justify your price point to be and how much you can charge for the product.
[27:32]It also will impact conversion rate probably more than anything else, because if you give them so much value, this is something that can change everything. So, this is always the first place that I really try to hyper optimize before anything else. Second place is then the price point.
[27:44]This is the easiest test to run and the test that will probably have the most impact consistently. So, price test is, it affects your AOV and your conversion rate. So, doing this on top of an offer test, these two together are really the most important variables.
[27:59]And then from there, usually it's going to be above the fold sections because of what I mentioned, where above the fold sections are where you're going to maximize the amount of people that actually see it, which is super important. And if you get that, you'll get people to actually read it. So, this can have the biggest impact no matter what page.
[28:14]And then usually it will be somewhere along the lines of a product breakdown, especially for an advertorial page. The product breakdown is where you're really selling people on the product. So, there's a lot of benefits to doing tests there. And then it's just about getting creative, right?
[28:26]It's really using the next part of like how to get the ideas, what I'll talk about in just a second and always tying back to the data that we looked at before. So, based on all this, now, how to get ideas for testing blocks? This is what's super important and how you can start to actually get creative.
[28:41]First off, it's about finding competitors and building a list ideally. And it doesn't have to be in your niche. Ideally it's actually not in your niche. So, that means that customers that are searching for your specific product that you're selling, haven't seen that layout before. So, getting inspiration from bigger brands or brands in completely different sectors that have this optimized well.
[28:56]So, always be on the lookout for really good landing pages and build up a list so that you can get all these different ideas. Then, whenever you're looking at a landing page, make sure you validate it that they're actually working. Use Similarweb. Ideally, it has above 100,000 visitors a month before trying to utilize that as a reference point.
[29:15]Then the next piece here is look at how competitors are presenting that block you picked to optimize from a variety of them. Whatever block, let's say it's the above the fold block, look at a bunch of different pages and see how they're optimized above the fold. Get some ideas. For example, we just looked at that one, where maybe my above the fold section is similar to this, but I see that they have a picture of the person that actually wrote the article.
[29:38]So, that could be super interesting. And based on that, what I might want to do is get inspired by that variable. Whereas if I look at another advertorial, like maybe let's pull up this one, which we didn't look at yet. So, going into this one, they have that GIF element, which I spoke about, which seems to be super powerful.
[29:54]So, the more types of advertorials or not even advertorials, just funnels or pages, whatever you're trying to test that you look at from competitors, that's going to give you a lot of these really good ideas and you can start figuring out different elements, especially when you're zooming into a building block that you already know that you want to optimize. And you just like hyper focus on that from your different competitors. It'll give you a lot of really good ideas of ways that you can frame it and present it.
[30:17]And a hack that I do is I usually will take a Miro board, I'll do a full page screenshot of different competitor pages that I like. I'll outline maybe like 10 of them together in one Miro board so that I can see it all in one place. And then based on the block that I want to optimize, I'll look at those different elements. Get inspired by different ideas, and then use this as my basis for the ideas that I'm going to actually go and test.
[30:37]So, once you have a bunch of these different ideas, then what you got to start doing is creating a list of all these ideas where you track it. So, based on these ideas, you should know what's the page that you're going to optimize first and foremost. So, is it going to be the advertorial page, the product page, a different page altogether? And what is the optimization category that I'm going into?
[30:57]So, is it going to be the above the fold? Is it going to be the problem section? Is it going to be the product section? This is where you're really defining pretty much the building block of where you're going to be optimizing. And then what is the main KPI that you're planning to lift? So, before running the test, like what are you expecting to get out of this test?
[31:13]Are you trying to improve your average order value? Are you trying to improve your conversion rate? So, really understand what this is before going into the test. And then after you run a test, this is what's really important to keep track of is impact after the test.
[31:26]So, did it win, did it lose? What percentage did it win by if it did win? And if it lost, you don't need another percentage, but at least keep track of the learnings after the test. So, you know, if it worked, you say this worked because XYZ and I think because of this test, because I did this and this.
[31:44]And if it's just a singular test, where it's just like the headline for example, then you'll know exactly what it is. And if it lost, say, I saw that doing this and this actually decreased the results. I think the reason might be because whatever you come up with as your own reasoning to that. But just make sure you have a log that you're keeping track of all this stuff as you're trying to run the test.
[32:00]Now, in terms of actually prioritizing different ideas, this is super important. I think overlooked, where I want to give a shout out to our head of CRO and our head of funnels that helped implement this really simple framework, which probably a lot of you guys have heard of, but it really helped change the way that I went about optimizing.
[32:18]Because beforehand, I would have a lot of ideas all the time. The ICE framework, what it stands for is ICE, which is I for impact, C for confidence, and then E for ease. So, all these together are scores that you're going to create. Each one of these, you're going to have a score based out of five. So, 5, 1 through 5 for impact, 1 through 5 for confidence, 1 through 5 for ease.
[32:39]And then you're going to get a score out of 15. So, the ones that are closest to 15, those are the tests that you should prioritize first, because it means that you can have the highest amount of impact with that test. The confidence level that you have that it will work can be very high, and how easy it will be for you to execute on it will be much easier.
[32:56]So, example, if I wanted to do a full macro test where I redesigned and scraped the page, that would take much more time for me to do. The ease of it would be less than if I just wanted to change a headline. Right? And the impact of a headline change could be just as much potentially as me doing the whole different page if I see a high degree of confidence on it.
[33:17]Versus my degree of confidence of changing everything and making that work. So, pretty much once you start listing out all these different ideas and you get the list automatically like pre-ranked for you, just by having the score filled out. Now, you're going to have a much clearer understanding of how to prioritize different tests that you run and what order to actually run it.
[33:33]So, going into the actual software. So, first piece is looking at the data, understanding exactly what section you need to optimize. Based on that section, you can use some more data, like heat maps and surveys to understand what section of that page that you actually want to go ahead and start to optimize it.
[33:48]And then categorize it into your own building blocks so that you can figure out what elements of that page you want to group together in in individual test. Then, you want to decide on what building blocks or what elements to include within a building block, as we mentioned, and from there, you want to look at what are the different ideas that you can have for that building block.
[34:07]Then it's about just having a list of all those different ideas that you can have to improve that individual test. And this can also be other ideas you can keep track of ideas that are outside of that building block that you might want to come back around to to implement. And then ranking those different ideas with this framework.
[34:23]And to actually now launch and implement a test assuming you're using Shopify, what I use is Shoplift, but this is pretty much the same thing as intelligence. I have no affiliation with Shoplift, it's just I like their software, it's been much smoother for me to use. So, I recommend checking them out. And also, depending on what type of page you're using, like there's Replo as well, but the problem with Replo is that it creates different links.
[34:53]So, what's important to look out for whatever software you use, at least from my experience, is you want something that can take your existing link that you're already scaling and seamlessly be able to split test it directly from the same URL. That's what's really important. And you want to make sure that you have a clear way to track that data.
[35:10]Because it can get very confusing with a lot of these split testing softwares and also be very expensive, especially with softwares like VW or convert.com. If you're using Shopify, recommend using one of those apps, because it's going to be much cheaper. And as long as you have the data in Shopify where you can track individual landing pages.
[35:28]So, the way you can always break that down is go to your conversion rate. Once you're in your conversion rate, you can filter by from dimensions into page path. And then on page path, as long as you're properly naming your different URLs, you should be able to see the exact conversion rate on an individual page. And you can always cross reference to see if it's going up or down.
[35:49]And last point over here is just once you start running these tests, it's about understanding when to actually pause the test and when to implement and execute on a test. A few rules to go by is typically try to run a test for at least seven days to get a full conversion rate cycle.
[36:05]So, every day can be different. Saturday is going to be different than a random Tuesday. So, you want to make sure you can give it a full cycle ideally to see how it converts and how it holds against each other over that full time frame. And within that, you need to have statistical significance.
[36:18]So, this is just making sure you get a proper amount of visitors on each page before making any decision of what you implement. So, 95% plus is bare minimum, but ideally you can get this closer to 99% before making something actually live.
[36:31]So, these are the two most important variables, more than anything. It doesn't matter how much revenue a page does, it doesn't matter how many visitors, even how many on each page. This is really what ends up mattering. So, I actually attached a statistical significance calculator over here.
[36:47]So, if you guys haven't seen this, I'll just show you quickly how this would work. Let's just say, using this example, you put the visitors over here. So, let's say it's 10,000 visitors. On that 10,000 visitors, we get 10 conversions, and then let's say on variant B, it's the same amount of visitors, 10,000, but this one gets 20.
[37:07]This wouldn't be significant. Now, let's say that the conversions over here go to 400 conversions, this one goes to 600 conversions, and then we're talking about, I don't know, random number here, let's say 30,000, 30,000.
[37:20]So, now this would be 100% significant that this result would win at a 2% conversion rate. This would be enough for me to then implement the next queue. Now, you can play around with this. There's obviously a bunch of different scenarios. You can set it for 99%, 95%. This tool is called SurveyMonkey. And you can just do a basic search on statistical significant calculators, but that's how you can then determine what to actually set live.
[37:43]So, pretty much with all this being said, like the way that this all kind of goes in motion for these micro split tests, which we just went over, is you're going to determine, so we're assuming you're doing a micro split test again, you have to define what is the bottleneck, again, based on that bottleneck, your goal is to find what is the thing that you can improve.
[37:57]Which page is it, what section is it, what building block can it be? Based on that building block, getting ideas of what you can do around it, based on those different ideas, ranking them properly, then having a software where you can actually do the split test and then run that split test until you have statistical significance. Now, imagine you start to do this week over week, right?
[38:20]Where you're running a test, you're ideally getting some form of an improvement of at least like a 10% lift. And by the way, like in terms of what actually wins or not, what we've typically seen is anywhere between, if we're at a 50% win rate, that's pretty solid.
[38:35]So, one out of every two tests that you run actually ends up being significant, you're going to start to see compounding effect very quickly, because remember, if a test doesn't work, you'll lose money within that testing time frame, but you can always just revert back to what was working before. So, it's very hard to lose money when you're split testing.
[38:50]If you can get out of a 50% hit rate on these test, and ideally more, but at least 50%, you're going to be in a really good spot because you're going to start getting, even if it's a 5% improvement one test, 15% the next. All that starts to compound on each other until before you know it, you take a 1% conversion rate all the way up to a 2%, maybe a 3% just over a longer time frame than if you were to do this huge macro test over here, where you would have a lot less chance of this working.
[39:16]These are things that, you know, especially as you're at scale, and, you know, as you're doing this, you're going to start to see differences of $50,000, $100,000, $200,000 in a month just because you ran good split test that improve the overall CVR.
[39:31]So, this is super important. Hopefully you guys got a full understanding of everything here. Again, if you guys enjoyed this, I'd really appreciate if you guys gave us a like, subscribe to this channel. I don't sell anything. I have nothing to sell you guys. All I'm doing is trying to build the personal brand. So, thank you guys so much for watching and hopefully you guys got some good value out of today's video.
