Welcome, welcome, welcome everybody to today’s Top Line Live at Five. My name is Brian Devine and thank you for joining us to talk about Split Testing.
I know it’s not the most exciting topic that we can possibly cover but it is so important I thought it is worth taking a couple of minutes to discuss a little bit. Now the reason it is so important is because the terms between a winning campaign and a loosing campaign, meaning a campaign that makes money and a campaign that looses money is usually split at testing and it is usually a lack of split testing or lack of understanding on how they do.
So what is split testing? What makes sure we are using the same language when we talk about split testing? The basics of it, is that when you run anytime an ad, an offer campaign, that at some point, at some level of the process, you have your control. I think this is going to work great and then your test, your “I think this maybe, these changes will have impact.
If you don’t know which way it’s going to impact, you test and the great thing is that there’s lots of software up there that makes it really easy to run split test. Meaning, when somebody clicks on your offer, or if it’s like you put something in the mail and day go to a web page. That, there’s a software that can show Option A and Option B and then do fifty-fifty or you could say “Yeah I want this 20 percent, this 40 or you know,” You can kind of, you can go crazy.
So I want to make sure that you really understand especially what we’re talking about with split testing. It is having a control on it and test against it and whatever the winner is, that becomes the new controller and we’re testing against that. Then you kind of keep on doing that. Now if you have a campaign that just consistently withers and you don’t think there’s a way to improve on it, then you’re good, right? But it’s always that one little thing that can surprise you. It also does run the risk that you made that one little change and it tanks. That’s why we always leave the control so you don’t go casual about it. What was the one that won and what do I change? That’s really important.
Now, what do you split test? What are the things that you want to actually split test? You can split test your titles, your headlines, your offers, the landing pages, the thank you pages, the bump upsell. If you don’t know what a bump upsell is or a bump offer, there’s another video where we talk about landing pages and how do we bump offer on a thank you page. So, lots of things that you can split test. Your imagination is limitless.
So, with this new software there are new possibilities. If you think of it, you can probably split test. Okay, how many changes should you make when you’re running a split test? Because I’ve seen really good landing pages and the split test was so completely different from it. There’s no way to know what was working. Now when you’re first starting at it, you might say “Okay, I’ve got landing page A and I used to be forwarding this work but I’m going to test it against landing page B which is completely different. The same offer but everything else is different and what we’ve actually done is we’ve tested a complicated landing page that’s otherwise, I man there’s a lot of text on it. I mean, there’s exclamations, there’s sort of step one, step two, dodge this, do this; and in some areas of the country, it works great! In another areas, it has flopped and it’s good to be able to sort out by saying “Okay, we’re going to split test big and complicated versus super simple and we’re not going to explain very much because when you see a big difference between those two and then we’re talking about, that’s just about as dramatic as you can get. Once you see that there’s a difference, there’s a winner like “Boy, in this area, we didn’t have to go with the super complicated.” Then, we move the similar over to become your standard.
That is now your control and then when we test our small small, what I am actually trying to say is one thing. You don’t want to take a winning ad and then split test it against five changes because you’re not going to know what change had made impact whether it’s good or bad. You might find out that there is no difference but you change five different things and the other just can’t pull each other out. What you want to do is you want to have your control like “Okay, this is the one we can keep,” and then you test a change in your headline and that’s it! Nothing else, or you just start with images.
Now, we tend to start with images. The reason for that is in different areas, different people respond to different images and it tends to have a meaning in that. The second thing that we are going to be testing and that we recommend that you test are your headlines. What is a headline? This would get some of the attention in your market, in your area because this… I would love to say that I have a magic formula everywhere in the country and it just would work that way so there’s always little testing and tweaking that has to go off. Now when you got the winning headline , then you would nod them over to that is your control and you would change no more thing and maybe it’s something as simple as a bump color and this one is orange, this one blue and you may find the blue outlive the orange in that so there’s all the things that you can do.
Don’t run into the risk of split testing two uneven things. I’ve seen that happen over and over again. Once you have that winner, you have the control. Set your next test up and you can keep on getting those little detrimental improvements in your campaign and it can be a lot of them. I can go on and on but here’s the thing.
There are software that can do it on your websites. There are software that can help do it for display ad and a display ad is where you go to see and then see wherever it might be and they have apps. Those are software now that can actually split test those those kind as well so whether you came with you know, to it in the afternoon or in the morning, you’re going to see a different ad and then you can start adjusting your ad spent to whichever ad is doing best.
So obviously I can go on and on but I want to make sure that you do split test so you can have people talking about it and this thing, it really is straightforward and the explanations are almost in the name however you want to know what to split test and how to set your control, and then you start split testing more and more against that control in finding the little things that will have a big lift for your campaign.
I think I’ve covered everything. If you need some help, have some questions, get in touch with us. You can view us up on Facebook. Give me a call at 720-989-1932. I will be happy to answer your questions. Help you out any way that I can. That’s it for me! Thanks so much for your time. Bye for now!