Split It Basic Helper
(Ranked #17 of 56 Helpers)
Split testing can be very powerful, but it can also get quite complex - we like to keep it simple for you. The Split It Basic Helper does just that - keep it simple. It does one thing and one thing only, but honestly, that's all you need for 90% of your split testing needs.
The Split It Basic Helper will randomly divide your contacts in a campaign into two separate paths - path A and path B. It's designed for 50/50 split tests, or, what is commonly called an AB Split Test. In an AB split test you simply divide the traffic in half, make two variations of whatever you're testing (email subject line, for example) and measure the results.
Many solutions will complicate this by tracking your conversions and even auto-switching to the winner, etc... The problem with that is that this will vary dramatically based upon the volume of traffic that you send. So our approach is simply to create the split and let you track it and monitor it as desired.
- Simple AB Split Testing
- 10/10/80 Split Testing
- Split by Tags or by Goals
Simple AB Split Testing
No muss, no fuss, simple 50/50 or AB split testing. You have the flexibility of testing whatever you want, from simple email subject alterations, sales page copy, completely different offers or prices, or even completely different funnel paths.
Split by Tags or by Goals
The Split It Basic Helper will divide your contacts into two lists and either apply a different tag to each list, or fire off a different API Goal for each list - the choice is yours. We know that excessive tags can clutter up your app, so we give you the ability to use API Goals instead.
10/10/80 Split Testing
10/10/80 split testing is a variation on basic AB split testing. What you do is take your total list and only select 20% of it and do an AB split on that 20% (thus the 10/10 part). Then, after you determine which 10% segment wins, you send the other 80% of your list to it.
If you have a large list of 20k or more contacts, 10/10/80 split testing can be very effective. It overcomes the big problem of 50/50 or AB split testing which ensures that 50% of your results will underperform. In a 10/10/80 split test, only 10% will underperform and the other 90% will be thrown to the winner! I talk about this more in the video above or in the transcript below in this post.
Above is the toolbar options you'll see for the Split It Basic Helper. Here are the Toolbar button options from left to right:
- Copy Link for Infusionsoft - This first blue copy link button copy the URL used for the http post snippet inside of the Infusionsoft Campaign Builder.
- Configure Button - The last button on the far right is the start configuration button which allows you to View, Edit, Copy, Delete, view the Usage Report and the Usage Log.
Tips, Tricks and Shortcuts
Hello and welcome to MyFusion Helper. My name is Troy Broussard and this short video I’m going to show you how to use the Split It Basic Helper. Now when we go through this I’m going to show you how to set it up, first of all, inside of our application and we'll go through that process it's very straightforward and simple.
Then I’m also going to show you how to deploy it inside of Infusionsoft and in this particular video I’m going to go into a little bit more detail than I normally do in that because I want to show you a few different samples of how you might use it in some different campaign examples so that you have a little bit more structure and guidance than normal because split testing, there's a lot of different ways that you could do it and so I just want to give you some ideas. And then lastly we're going to come back and I'll go through and you know show you some best practices and give you some other ideas of where you might want to use it in your business. Now before I do all that though I want to give you a quick warning because of the way Campaign Builder works, it's not recommended that you run this split test on large contact lists.
So, if you have more than 10,000 contacts, it's not recommended that you use this particular solution. And the reason for that is that Campaign Builder runs one contact at a time. It doesn't run in batch mode. So if you have a large email list of 50,000 people and you drop them into a campaign that's going to result in 50,000 API calls and it's going to be something that will probably delay quite a bit the execution of the campaign and cause a lot of network churn.
Now, don't worry we do have some solutions for you for that. First of all, we do have an advanced split testing add on which you can contact us at the help desk about and if you're interested in that reach out to us at firstname.lastname@example.org on our help desk. It is at a premium price point. It's a very advanced split testing solution that allows you to send in batch mode to very large list, so even if you have a quarter of a million people on your list it will generate only a couple of API calls and send that out immediately.
So that is a solution, it is an option. But I’m also going to show you another way and that is for larger list let's say you've got 30,000 or 50,000 people on your list. Then instead of running a 50/50 split test which is considered an A/B. split test. You can run what's called a 10/10/80 split test. Now it really is still a 50/50 split test. It's just that we do the following, we divide your list into the first 20% of your list and so let's say you have 50,000 people on your list, we're only going to look at 10,000 of those people first. Now, we're going to take that first 20% or that 10,000 and we're going to split that up and do an A/B split test, two 5,000 segments of that test. Now when we get the results from that we then send the remaining 80% to the winner.
That's called a 10/10/80 split and honestly, it's one of the best ways of doing A/B split testing. And so you know you can still use this feature even if you have a larger size list just by running a 10/10/80 eighty split and once again all you would do is just start with a 20% subset of your list and run a split test on that pick the winner and then send the 80% of your list to the winner. So that's another kind of approach that you can use.
So, let's jump in and take a look at how to set this all up. First and foremost let's go in here and create a new Helper. So we'll go in here and type in “split test” and we'll do split test demo and we’ll go ahead and configure it, now, this is a very simple thing to configure there's not a lot of options here. The biggest choice have to make is whether you want to do a split test with tags or whether you want to do a split test with goals and it really doesn't matter, you can do it either way.
I’m going to do it with goals. It's my preference to do goals because i don't like just creating tags when i don't necessarily need them. They tend to slow down the application and so whenever possible I use goals and you'll continue to see this thread of education in all of our videos so the likelihood is that we'll use goals most of the time. But we know that most people are comfortable with tags so we put the support in there. So you can either pick two split tags or two split goals. So I’m just going to create a goal and I’m going to call it “Split path A” and “Split path B”. So now we need to do is go ahead and create a Helper so click the button there to do that and this will now have our link ready to copy and paste in, so I’m to go ahead and copy that and then a switch into Infusionsoft to show you how this works. Now before I get into that, we're going to spend a little time just talking about some of the stuff with this and when you set up a split test you know this is kind of a common scenario so many opts in and then you send them to a 4 email series, promotion and then you have a purchase goal at the end.
Now, when you look at this and you think about what kind of things could you split test and I've kind of written him down here for you and this is not obviously an exhaustive list but it's the main things and honestly it's a lot of times in kind of the orders that you want to look out. Like a lot of people jump in and start split testing email content right away and one of the first things you really want to split test is price points before you do that because they have the biggest impact and yield on the outcome of your funnel. So, those are some of the things to consider.
Another thing that can have a big and it's easy to do is split testing email subjects. If you had all of these e-mails written and ready to go then you could really easy just click and drag the whole thing down here and come in and change the subject on these 4 e-mails and in 10 minutes you could have a complete split test ready to go testing out the same content of the e-mail but just different subjects just to adjust your open rates and see how that works so, take this and take the approach that's going to give you the most end results. So, let's take a look at a couple of different options here so here's an example of what we could do with a split test.
We could come in, we have an optin, we could create the split test and I'll show you how that's going to work using the goals that we set up and then we could have 4 e-mails here going to a purchase at 97 and then we could have the same emails going to a purchase at 127 and we could test to see how our conversions work at this price point. This would be one type of split test that you could execute now to show you how to do that, we already copied the URL so we would come in here and we would add in an API post method here and so we come in and we paste that link in here. So we paste the link in, set it ready and come back and set that ready and then that would be in here ready to go. Now, it's going to create two goals.
Now remember we did our goals here we had Split path A and Split path B, so we would put in MyFusion Helper. Split path A. for our A split test and here we would do Split path B. MyFusion Helper is always going to be the integration name and split path B. Now, what is going to happen is that when this executes it's going to execute that API call and then depending on which goal it sets, it's going to send them down one of two paths and now you have your contacts split up into two separate paths. That's how this works, so you're naming your split test goals and you can do it that way now you could also do it with tags. So it's no different, it's just that you know instead of an API call here you would have the, applies a tag and have and have a tag goal set. So either way you want to do it it's totally up to you we tend to prefer using API calls but you're absolutely welcome to use tags as well.
And then you would have a split test set up this way. Now let's take a look at kind of another example here so let’s kind of scroll down here and show you a different type of an approach and you know this is again it just depends on what you want to split test, in this particular split test were dividing up the list we’re splitting it A and B, were sending them through the same e-mails and we want to see which one fills out the application and so maybe we want to test our open rates on these e-mails and so, here we've got, four different emails here and the same or different emails with different subjects that we want to do an open write comparison on and that's a way that we could do it there. Now if we wanted to split test two different webinar registration forms, we could simply do this and make a copy of that.
And let's say now we had two different lead pages that we were sending traffic to register for a weapon our webinar. This could be lead page A and this could be lead page B and we could test that. So, it's really unlimited how you want to do it all we're really doing is calling an API call here to create a division and to automatically divide the traffic and or send them to two separate paths and how you want to use that to your benefit is really up to you. So again here are some of the things you could be testing price points, subject lines, different order forms on the checkout process checking, you know testing different upsells and you don't even have to test it for price points, you might be testing entirely different products you might have two different product versions that you're testing or two different offers that you're competing against each other to see which one works . So, there's lots of different options here that you can do but that's how you would set it all up.
All right so let's jump back into the training and wrap things up. So, hopefully that gave you a few examples of things that you could do and just to summarize we talked about testing product price points or product A versus product B. We talked about testing e-mail subjects' and e-mail content and sales pages, that's one thing we didn't mention is sales pages and that could be very, very useful so maybe you want to send them the exact same thing. But just a different link that goes to a different sales page and one of them you have a video sales letter and the other you have perhaps just a traditional copy written letter and see which converts better again that's another possibility for a split test.
You know testing your upsells can also be another way that you can test which path is going to work best for you now I really want to encourage you to start big and then go small and what I mean by that is a lot of times people split tests the minutia they split tests the little things like colors on the page and silly stuff like that and that is not going to have much of an impact. What you really want to do is start big - split test against big variations.
Split test a VSL versus a copy letter for sales pages that's a big difference to know whether your audience is going to resonate more with a video or a copy written letter. That's a big difference right split test things that are dramatic. Split test price point split test your offer. Split tests the bonuses but stay away from the little stuff because that will just waste a lot of time and probably not yield much as an end result.
Another thing to keep in mind is consider using the 10/10/80 approach that we talked at the beginning. You take 20% of your list and you run an A/B split test here then you find the winner and then you send the remaining 80% of your traffic to the winner. That is a system that works very well, it's super easy to do and you don't need anything more than the Split Basic that we have in the app for you.
All right so this is Troy Broussard signing off, I hope that gave you some good ideas and get you rolling with some split testing in your business and if you need more help reach out on the help desk at email@example.com.