Split testing is a marketer’s best friend when it comes to perfecting the sales funnel. Why? Because it actually allows them to make the right changes for their sales funnel to become a converting machine. It’s extremely important to recognize that split testing is a way to strive for the best numbers on each landing page you release.
The more you test, the more you’ll make.
But what is split testing really, and how do you then go on to do the actual testing?
Putting it simply, split testing, also known as A/B testing, works by having two different versions of your website or sales page shown to visitors who come to said page.
Visitor A sees version A and visitor B sees version B and so forth.
Both activities are measured on the page where the split testing takes place and the goal is to have one page outperform the other according to your marketing goals.
For example if you’re split testing a webinar funnel you might want to increase the registration rate on the webinar registration page.
Now, there’s different ways to measure the success or failure of a page you’re split testing but if you’re using ClickFunnels you’ll be able to see the results right on the landing page within your funnel.
You’ll see each test side by side your control. You’ll also be presented with a confidence score which describes how ‘confident’ you should be in declaring that test the winner.
By consistently testing at each stage in your funnel this results in an extremely well optimized sales funnel, which in turn results in more revenue.
And we all love a good revenue stream, don’t we?
However, after the two basic problems of what split testing is and how to split test are dealt with, most marketers are left with one further question, and that is what should they be split testing.
This article is here to help all of you who are wondering where you should focus your split testing efforts to begin with.
Split Testing Your Headline
First and foremost in the split testing arena is the headline.
The large attention grabbing piece of text that sits at the top of the page.
You know the ones.
An excellent and well crafted headline, means more and more visitors are clicking their way to their site and reading further than the first line on the page.
After all, what’s the point of building a fancy landing page without having any visitors read further than the headline?
Most importantly, not consuming enough information to purchase your products!
That’s a surefire way to a shortage in sales.
There’s an interesting statistic related to the importance of headlines, as Brian Clark from Copyblogger explains, and it states that 80% of your people read the headline, but only 20% will read the rest of it.
Yeah, that’s how important a winning headline is, since it’s the only piece of copy the majority of people will ever read on your landing page.
So there’s really no room for error here.
So what do you need to know to write a winning headline?
Wel there’s a lot of different strategies that you can use to grab people’s attention but here are 4 simple steps you can take that Neil Patel reccomends you remember:
- Your headline should be unique
- Your headline should be ultra-specific
- Your headline should convey a sense of urgency
- Your headline should be useful
Write different variations of the headline, pick out the two best headlines (or even better, ask different people to pick out the two best headlines and then see which two really are the best ones) and off you go to split testing the two.
After having used the two different headlines in your split testing efforts, you should see a difference between the results these two headlines had.
I won’t say how long it takes for you to get clear and trustworthy results because that depends entirely on how many visitors you get and how you acquire them.
Visitors from paid advertising behave entirely different than visitors who found a blog post of yours entertaining and decided to check out your product.
But the rule of thumb here is, the more visitors you split test with, the better the results as you’re testing at scale. That part’s quite obvious really, but I just wanted to make sure you were aware of that.
Testing Your Offer
After the visitor scrolls down from line one, meaning the headline that you’ve successfully split tested and optimized to perfection, what they should see is an offer.
An offer is quite self-explanatory, because it’s exactly what it means in every situation, you give your visitor something in exchange for their email address, and that is the offer.
They act as the incentive for your visitor to voluntarily enter their email address into the field on your landing page.
The offer can be anything, from a free eBook to a webinar that you recorded last week. Just make sure it’s something your target audience would be interested in, because getting a free eBook about nutrition is great, but might not be what you’re looking for after clicking to a landing page that has “Amazing Photoshop Skills Fast And Easy” as the headline.
A good offer, however, is more than just an eBook relevant to the niche, it needs to deliver immeasurable value to the visitor
Everybody has free eBooks that they offer to their visitors, which is also a reason to think outside the box on this one. Maybe go for something different and more interesting.
The lead magnet is a piece of content that you want your visitors to get instant value from as soon as they give you their name and email address.
By delivering value in an easy to access, fast to go through and simple enough for a toddler to understand package, you’re golden.
Some examples of nice and creative opt-ins can be found here if you’re looking for some inspiration.
When you’ve come up with two relevant offers to use as incentive for your visitors, you’ll get to the actual testing phase of the operation.
This one is really the easiest one to test, because all you need to do is see which incentive leads to more signups over a certain period of time.
If you’re using ClickFunnels I’d reccomend using a ClickPop in the sidebar of your blog to drive optins for your lead magnet.
Split Testing for Your Ideal Price Point
While it’s safe to assume that there’s one good price point ($5 cheaper than your closest competitor’s offering) you shouldn’t always go for it.
Even though it’s a common pricing strategy, called the market based pricing strategy, you can grow your revenue rather easily simply by split testing two different price points.
You can for example take the market based pricing strategy as one option, using a price common to the market or one slightly below that, and then compare its results to a completely different pricing strategy. The alternative price could be based on the profit margin or alternatively change the pricing type itself.
What do I mean?
Well think about this, a membership site can be priced any of these ways:
* $37 per month (charged monthly)
* $30 per month (charged yearly)
* $1 trial for 7 days (then $37/month charged monthly)
* Free trial for 7 days (then $37/month charged monthly)
* A one time $397 charge
All of these are different pricing strategies which can be used for a membership site but finding the right one for your business model requires testing.
You need to take into consideration some terms which can be found in the internet marketer’s dictionary.
Testing for the Perfect Call to Action
The good old call to action made the list as well, because let’s face it, there’s really no website or landing page without a call to action (and if there is they obviously know nothing about marketing).
A proper call to action grabs the attention of your visitor in a way that arises intrigue and desire in them.
You’ll need to make them want to sign up to your mailing list or purchase your product.
By having an effective call to action in the right spot on your landing page, you can easily double the sign ups in comparison to those gotten with poor call to action efforts.
An amazing call to action is always clear and simple, and never asks for the visitor to take multiple actions at once.
Only request them to sign up, subscribe, watch the video or something else that you’d like them to do, but don’t ask them to do everything for you.
That just confuses them.
For an excellent CTA and all around conversion article, check out this post Kissmetrics published on what converts and what doesn’t.
When analysing the results of split testing two calls to action you’ll only need to compare the amount of signups each individual call to action managed to gather from the traffic you sent to them, which makes this one an easy measure as well.
This is turning out to be a lot easier than you thought, isn’t it?
Color Layout and Background
Last and least important this time, split testing different colours, layouts and backgrounds on your landing page.
How your visitors perceive the look and feel of your landing page is important, but a well thought out landing page should not see a drastic improvement from a different font colour or background image (unless the original image was horrid, seen that happen as well).
You can however gain a few new sign ups from the slight improvement made by these split tests. When testing elements such as font colours remember to consider the psychology behind colours and how they subconsciously impact the decisions made by your visitors.
I’ve covered colour psychology and landing page backgrounds quite comprehensively before and I definitely recommend you check it out!
While a tad harder to measure, the process is fairly straight forward. If you have the necessary traffic for it, you could try multivariate testing for the colours and layout, which allows you to test multiple different combinations in one testing instance.
No more split testing blue against green, then green against red and red against blue, just testing them all on different visitors in a single go.
Remember, you want to test BIG changes.
That’s where you’ll see the highest lifts in your split testing.
So What Now…
There you have it, a list and explanation into each of the five elements you should start your split testing efforts with.
I’ve put the five elements into an order based on importance too, but really the four first ones are nearly equal in importance, and that is vital.
You shouldn’t leave out split testing any of those four under any circumstances.
If you want more in depth, specific and complex metrics and results, you should try multivariate testing, which is basically split testing, but as the name suggest there are multiple variables and versions being tested at once.
It’s especially convenient when measuring features such as colours and their combinations as well as different font types.
As always, we love hearing your thoughts, comments and experiences, so don’t hesitate to share them in the comments section below!
With multivariate testing, you’re saying we can do a split test using multiple variables and options at once. So, for example, we could split-test a different headline and offer at the same time. Is that right? If so, how would we know which variant is resulting in the better conversion?