UPWORD Consulting - Week 5

A/B Testing


I want to help as many freelancers and agency owners generate high-value WordPress projects consistently and predictably and will keep the training and weekly support 100% free for as long as I can.

The resources in this course are crazy valuable ($15,000 worth of resources - but you won't pay anywhere near that), from all of my proposal templates, to legal agreements, to a fully built out WordPress agency site, these premium resources can help accelerate your growth.

You won't find another course like this that is so affordable and packed with so much valuable content. Unlock the resources today!

Learn More

Hello everyone, Welcome to Week Five and module five. So in this module, we're going to be covering a B testing. So this is after you've set up your original email automations that were included in the original scope of work. And now hopefully your client is excited and wants to work with you more. Also below, you'll see I've included a proposal, just a short explanation of what we do, and then kind of the cost involved with it or investment. So once you get someone that wants you to maintain and manage their email marketing monthly, you're going to want to set up a B test. And the reason why is you want to learn what subject lines work, what time of like time delays work, and I'm gonna show you how to do that on clay vo, it's super easy. So let's hop into it.

So let's start with just this abandoned cart sequence. So let's say we wanted to test the four hour time delay, and we want to test four hours versus two hours. So the first thing you do is what's called a conditional split. So you'll see right here, it's going to split it up, then over here, you're going to select the condition. So what I do is select a random sample and include 50% of people. So 50%, randomly, you're going to go down the normal path automation, and 50% are going to go down this other path. So we're gonna hit save, then we're going to add a time delay. So you see, I just drag this over, let's set this to two hours. Great, I've now have an AV test of this time delay. And from here, all I do is clone this email, you'll see it here, and I drag it over. And then I could either do another AB test on the split. Or I could just mimic it exactly. So if I wanted to test this time delay, then I would just copy everything over Exactly. If I wanted to AV test everything, then you would do more conditional splits. But I'd recommend just testing one conditional split at a time. And so that's how you test time delays. Now let's test let's set up an AV test for an actual email. So a lot of times where I like to start is AV testing subject lines, because subject lines are 80% of the battle, you want to figure out how to get those open rates up so that you can get more people reading your emails, so you get more people converting on your emails. So the first thing you do is click Add variation. And this is going to set up an A B test for you. So you see, this is the original one. Let's hit Edit variation. And it looks let's test. Let's remove the name and see what this subject line would do versus showing their name or even preview text. So let's test of it being personalized versus not personalized. So you leave the email the same don't change anything, you just want to test the subject line. And then from here, you see we've got two in here, right? So here's the original one. And then what you want to do is split this traffic evenly. So you see 50% 50%. Now we've got an AV test running on the time delay, and also on the subject line, I would recommend doing one or the other, not both. So that you can really get clear sample sizes and statistical significance. So you're going to want this to run long enough. So you get enough data depending on how many how big your email list means the longer you'll have to run it to get statistical significance. You can also Google statistical significance calculators, and you can put in this data exactly from experiment, and it'll tell you how confident it is that you have a winner. So that's really how I set up a B test. And then you can also test body content. So you just want to make sure that you're choosing either subject line or body content. So you're very clear on the results of your test. If you're doing too many different tests. It's hard to get significance and know honestly know what's working and why. So start very simple. That's how you set up a B test enclave.