AP 0050: Which Split Testing Tools Do You Find Most Useful?

AskPat 50 Episode Transcript

Pat Flynn: Hey what's up everybody? This is Pat and welcome to Episode 50, Episode 50, were halfway to 100 which is awesome. Over 500,000 downloads, this is just amazing, I can't thank you guys enough. Again, this show is powered by you and I'm here for you, I'm here to help answer your online business questions five days a week.

I also want to thank this episode's sponsor which is FreshBooks. If you go to GetFreshBooks.com and put “Ask Pat” in the “How did you hear about us?” section, you can get a sixty-day, that's two months, extended free trial to use FreshBooks for all your counting needs, bookkeeping and all that good stuff, invoicing, it's just such an important part of business that we don't realize we have to do when we get started and I only wish I'd gotten started with FreshBooks sooner. So, now let's get onto today's question from John Dennis which is about something that a lot of us can do more of and that is split testing and these are specifically about the tools. So here's John Dennis.

John Dennis: Hey Pat, what's up? It's John Dennis from The Smart Time Online Podcast at SmartTimeOnline.com and I have a question about split testing. I want to know what tools for split testing are you finding most useful for your website and why? Thank you so much for all that you do and for the opportunity to pick your brain. Take care man.

Pat Flynn: John Dennis, thank you so much for your question and like I said just a moment ago, split testing is something that is crucial to our businesses and to understanding what's happening and what is best suited for us and our audience but we don't do it for some reason and I think it's because it's complicated, I mean, how do we split test? And for those of you who don't understand what that is, it's essentially, split testing or AB-testing, as they call it, is when you drive half your traffic to one version of a website and the other half of your traffic to the other version of the website so Version A and Version B and you'll be able to see, based on tracking and conversions, which one works best and it's such a powerful thing because then you can take that one that works best, and you know that one's better or you're doing it better, and you're maximizing the conversions on your website. A lot of people talk about driving traffic and SEO and doing these things to bring people to our websites, but we don't focus enough on what happens when people get to our websites and when you can understand what converts better, you're going to make more money, you're going to build your email list faster and those sorts of things.

So what tools do we use? Well I'm experimenting with a lot of different tools that use split testing and a lot of specific tools that are built or meant to be used for split-tester sites. The first one I ever used isn't around anymore and that is Google Website Optimizer. I used it to understand that the sales page on my website at GreenExamAcademy.com is the best one. And it's funny because if you go to GreenExamAcademy.com and go to my sales page, it looks terrible. It is disgusting. There's side bars, it goes against all the rules of what makes a good sales page, but the numbers don't lie. I tested it against several other versions, one that looked like Amazon, one that was at—you know you see these crazy long, super lengthy sales pages with testimonials and all this stuff: No, they didn't work as well andx the original one I put up, which is what you see there still, has converted the best. There would be no way for me to know that unless I had tested so I'm glad I did. And that was using a tool called Google Website Optimizer, which like I said, isn't around anymore but it's sort of, it's folded into Google Analytics and it's free, you have free access to it, it's a little bit complicated but there are ways and tutorials on YouTube or on Google where you can understand how to quickly use it.

What you'll need is you'll need a page that you want to test, variables on that page, and when you test you really just want to test just one thing, one thing only so you know that that's the one thing that works better. Once you start testing five or six things against another version that have five or six different things, you're not going to know if it's one of the things or all of the things that made those conversions higher and or lower. So it's important to test one thing at a time. Split A versus B, one variable, and you'll also need some sort of thank you page, or a page at the end where people can arrive on, cause that's the way it works technically. People arrive to your site and once you put these tools on your site, half the traffic goes to Version A, half goes to Version B, but they end up in the same place which is typically a thank you page or perhaps it's, you know, a billing page or something like that and that's a page you'll have to set up in these different tools and that's how you know which one converts best. So that is what used to be known as Google Optimizer which is now known as, I believe, Content Experiments. So when you're in Google Analytics look at Content Experiments and you'll figure out where you can learn how to get that all set up.

Now there are some specific tools out there, third-party tools that are built just for split testing traffic that comes through site and the different variables that you put onto different pages and there's two tools out there I've been experimenting with and both are great and I'm still trying to decide which one is best but, if you check out VWO.com or Optimizely.com, both are great third-party split testing tools. It's literally technology that could change your business; test things and see what works best. I highly recommend getting involved with one of those three previous tools I just mentioned. If you don't, you're leaving money, you're leaving subscribers, you're leaving results on the table. It's just plain and simple that that's what you're doing if you're not split testing. You should always be split testing something to improve your site.

Now there's other tools out there which do something specific but also have split testing involved and that's what I've been using more of lately. For example there was a tool called Hello Bar, which puts a nice little bar at the top of your site and you can split test different variables of the Hello Bar so if, actually I'm testing this right now, if you go to GreenExamAcademy.com you might see one at the top of the page and you might be seeing Version A or you might be seeing Version B, you don't know as the visitor but I know in my analytics in the backend of the Hello Bar, which will tell me. And Hello Bar is really cool and it's free if you go to HelloBar.com, it's by Neil Patel. It's a great little tool you can use to sort of increase either conversions into a product or conversions into your email list, so check that out. And again, that has built in split testing.

Another thing that has built-in split testing is a tool called LeadPages. I'm an advisor for Lead Pages which is co-founded by my good buddy, Clay Collins. Again, that's LeadPages.net. Check it out, it's amazing. What it does, it allows you to build these incredibly beautiful landing pages for collecting email addresses or a number of other different things, you know, webinar pages, all types of pages, any types of landing page you can think of, there's templates there that make it incredibly easy to customize and create and you can split test. You can create two landing pages, for example, if I were to say, go to EbooksTheSmartWay.com, that could go to the page where I promote that free ebook to get people into my email list and I can have two different versions of that landing page which is created by LeadPages.net. So again, you can go to LeadPages.net and check that out, it has split testing built in and yeah, so those are the tools that are really helpful and I have been using lately that involve split testing. [Full Disclosure: I'm a compensated advisor and an affiliate for LeadPages.]

So, in Google Analytics you have your content experiments, you have VWO, Optimizely, Hello Bar, which is that tool at the top of the site that I mentioned earlier and also LeadPages.net. So check any of those out. You should always be split testing something. John Dennis this was a great question, thank you so much because this is really important and I think people need to realize, and again, if they're not split testing something, you are not doing the best you can to make your website the best it can.

So John, thank you so much for your question. An AskPat teeshirt is headed your way, thank you. For those of you listening, if you'd like a question answered here on the show or potentially answered on the show, you can head on over to AskPat.com and you can potentially win a teeshirt as well.

So thank you so much and to end the episode as always, I'm going to reiterate our sponsor which is FreshBooks.com, again the link for that is GetFreshBooks.com and enter “Ask Pat” in the “How did you hear about us?” Section. I mean seriously, hunting through receipts and invoices and all that stuff, I mean this software makes it so easy to deal with all the accounting needs of your business. FreshBooks.com, that's GetFreshBooks.com and then put in “Ask Pat”, thank you so much for that. And the quote of the day I'm going to leave you with is from Neil Patel himself, and he says, “If you don't have enough conversions on each variable, you shouldn't stop the test. My rule of thumb is each variation should have at least 100 conversions and the tests should run for at least 2 weeks.” Now that's not like, an inspirational quote from Neil but that's a very very important rule of thumb from Neil as far as when you do split testing, because I know a question out there, “Okay well how long do I split test for? I have Variation A and Variation B, how long should I go until I find out which one is better?” And again, these tools will tell you which version is better. But you want to give it some time, so 2 weeks or 100 conversions so you can understand really what's working and what's not.

Thank you so much for listening to AskPat and I'll see you in the next episode. Peace.