5 Split Testing Tactics You Can`t Ignore
Do you know what separates the pro split testers from the wannabes?
It’s knowing what needs to be split tested and when to do it.
You see, the guys who get results often know where the big wins lie. They know what needs to be changed and how to change it.
Those new to the game, however, don’t have this luxury.
More often than not, newbies will split test the wrong things in the wrong order. They will then complain that this split testing thing is a pipe dream that doesn’t work for “their business.” They may even say something like they don’t have enough traffic.
It does not have to be this way.
We can learn from the pros and emulate their tactics.
In this article, we’re going to look at the top five things you need to focus on if you want to boost conversions quickly. We’ll specifically focus on homepage elements, as that tends to be where a lot of big wins lie.
We’ll go through each homepage element in order of importance. We’ll also break down examples from other websites to see why they’re so effective. As you finish each section, you’ll have a handful of takeaways. You can then use that information to improve your own conversion levels.
As you start implementing these tactics it won’t be long before you too, can be called a pro.
5 Split Testing Tactics that Can Make You Look Like a Pro
Before we get started…
Before we get started, we need to touch on a few things.
In an ideal world, you would go into a split testing experiment with some actionable data. This article is of value, because you don’t always need data to get results — you can just emulate others and follow some best practices.
However, it can help to know what website visitors are already doing on your site and what they want to learn more about. When you have such data, you can then make informed decisions that will improve the user experience.
Even better, those decisions are more likely to produce results because they’re based on hard data rather than best practices or data from someone else’s site.
If you want to collate such data, the CrazyEgg tool is helpful. Here is a case studythat documents how Crazy Egg heatmaps boosted leads by 24%.
Ready to review the top homepage elements you need to be testing? Roll up your sleeves, and let’s start.
1. The headline
The headline is often what people see first. Therefore, split testing a headline can contribute massively to conversions.
On a homepage, headlines generally explain what your business does and how it can help the person who is visiting your site.
Homepage headlines and subheadings
When it comes to homepage headlines, you can’t just copy another company.
Even though you might be able to use another website as inspiration, you’ll often have to create a homepage headline using your own skills. That is because your company is often unique in its offering.
More often than not, the rule for homepage headlines is to explain what your business does and why people should care. This should be done as simply and succinctly as possible.
Let’s view a few examples to help you better understand what a great headline looks like. As you look at them, you’ll notice some commonalities.
This image below is from the Airbnb homepage.
The headline isn’t aiming to prompt people into taking action (that’s the job of the button below). Notice how the subhead clearly explains what the business has to offer.
The next image is from Groovehq.
Observe how the headline and the subhead explain simply what the product is and what it can do for you. Again, the button below acts as a call to action.
Here is an example from the KISSmetrics homepage.
Once more, the headline points out what the product does. The text below explains it. The button then tells you what to do.
The same pattern is to be found on the Unbounce homepage.
What you need to focus on when creating headlines
As you can see, most homepages need headlines and subheads that simply explain what your business does. The button acts as a call to action. This gives you the space to talk about other things in your headline.
When you split test a new headline, always focus on simplicity. In as few words as possible, explain what your business does and how it can help people. Make an attempt to avoid jargon too.
Collecting impressive headlines
As you browse the internet, make an effort to take a screenshot of any headlines that appeal to you. Note the ones that prompted you to take an action.
Once you’ve found a few headlines you like, use them as inspiration for your own website. The more headlines you see, the more you’ll understand what makes a great headline.
A quick note on color
Before you split test any headlines, always ensure that your headlines are easy to read. The color should stand starkly contrast the background.
If your words can’t be read, it won’t matter how good they are.
If your current headline color does not contrast well with the background, consider split testing the color before you test the headline itself.
2. The image
When it comes to what people see first, the image and the headline are closely tied.
The images we’re talking about here are the images seen above the fold. These are often the first images that you see on a webpage.
A bad image can sometimes be saved by a good headline. However, a good image loses its effectiveness if the headline is bad —which is why we suggest it is second in importance.
If you want to split test your images until you get results, here are a few things that you can do.
Include real people and ensure that they’re looking in a certain direction
Images of people are already known to boost conversions.
But you can amplify this effect by using a person’s line of sight of in an image. When someone in an image is looking somewhere, the website visitor automatically follows their gaze.
Below is an example from a Salesforce homepage. It shows what appears to be a happy customer. So far so good!
However, on the right-hand side of the page there are several buttons. What’s interesting is that the person in the image is also looking in the direction of these buttons. Well done, Salesforce!
Evernote understands this as well:
Evernote’s home page changes to display different screens. However, a few things remain consistent on each screen. There are always people present and they’re always looking in the same direction.
A lot of them are looking in the top right-hand corner. That’s where the download button is located for the Evernote app.
The fact that they are looking in that direction makes you focus on the right-hand side of the page. Observe how there are no obvious buttons on the left-hand side of the page — could it be because no one’s looking there?
Boosting the impact of your images
Line of sight can have a big impact on conversions. If you’re using an image of a real person, incorporate line of sight. That means you need to make sure the person in the image is looking at some form of call to action.
Show your product off
If split testing images of people produced no results for you, consider using product images instead. You might want to use images of people using your product, or just images of the product itself.
This example from QuickBooks shows how someone can use the product:
The picture is great because it answers a few questions that potential customers might have in their head already. For instance, it details synchronicity between devices. It also shows how simple the interface is. These tend to be things that consumers care about.
Here’s another image from a company called Workday. Though their homepage image changes to show a number of things, the main image displays their offering.
Once again, it shows that the platform works on different devices. It also shows that the interface is easy and simple to understand.
If the product looks appealing, people will be more inclined to learn more. They will be even more inclined to find out more if your images implicitly answer the questions potential customers are asking.
What if you don’t have an image — should you readjust everything to include one?
Sometimes the inclusion of an image might require you to change the entire layout of a page.
Here is an example from the website Highrisehq.com. In order to make room for an image of a person, they decided to change the entire layout of their page.
As you can see, the layout change was quite dramatic. Yet it looks like it was worth it. Their conversions went up by 102.5% as a result of their actions.
If you don’t want to readjust your whole page, that is fine. Just move on to the next section of this article. However, when you do have the time, try implementing an image somewhere into your site. In doing so, you might be able to boost conversions massively.
Even if you do have an image — always test your assumptions
If you’ve already logged a lot of time split testing images to see no results — try using no image at all. It might be that the use of an image is actually causing you more problems than it is solving. Remove it and see what happens to your conversions.
You might even try replacing your images with a simple graphic that explains how your product works.
Here’s an example from Unbounce.
Remember, when it comes to improving conversions, you might have assumptions. They don’t matter. What does matter is how well you test your assumptions. This article is based on case studies and best practices. Yet those mean nothing if they don’t work for your site.
Your assumptions don’t matter. It only matters how well you test them.
3. Copy below the fold
The third element you need to focus on is the copy below the fold.
You need to ensure that you’re displaying the right kinds of information, and that it is displayed attractively. Split testing can help you figure out how to do this the right way.
Homepage copy is often designed to sell your product. To ensure that you do a good job at this, you need to make sure that this copy is benefit rich.
Homepage copy breakdowns
Here is an example from Grammarly.com. It’s the first chunk of homepage copy that you see, and it’s designed to quickly impress you and sell you on the product. Notice it explains everything in simple terms — enough to pique your interest.
If you scroll down the page a little bit more you see a second chunk of copy. This part talks about product features in a bit more detail — there’s even some jargon.
It explains clearly why you need these features. Pictures are used to back up the claims made.
This section of copy is great for people who want to learn more about the platform. It makes it easy for them to learn more, avoiding the need visit and search another page for information.
Here’s another example from Evernote.
They combine both the features and benefits into one paragraph.
As you can see, the headline tells you a feature and the paragraph below details why it’s a useful feature to have.
Here’s another section of their homepage. Can you see how they’ve set things up?
Evernote uses the image formula used by Grammarly: to describe a benefit and then show that benefit in action with the help of an accompanying image.
Here is a snippet of copy from the Freshbooks.com homepage. This is the first chunk of copy near the top of the page.
As you can see, the copy is designed to explain benefits — a simple snapshot view of why you want to use the Freshbooks platform, similar to Grammarly’s approach.
When you scroll down you see this screen.
It gets a little more technical and explains everything with a bit more detail. Great for people who want to learn more about the benefits mentioned earlier. The use of a call to action also gives people an easy way sign up.
Creating winning copy
There’s no point in split testing copy if you can’t come up with improved copy in the first place.
If you want to create winning copy, you need to know the problems your target market wants to solve. You then need to figure out how you can position your product as the solution to these problems.
That means you have to get used to explaining your product/service in terms of benefits. These benefits have to appeal to those who might buy your product.
Website Optimization Tip: Explain your product in terms of benefits.
Freshbooks appeals to small business owners. They could tell small business owners that they can make admin work easy. Yet instead, they tell them how Freshbooks can save them time — something a lot of business owners don’t have enough of. They position Freshbooks as a solution to this problem.
You can always detail the features of the product further down the page — as was shown in these examples. Just make sure your benefits speak directly to the people you’re trying to target.
Great images tend to amplify the impact of great copy.
Using the copy breakdowns from earlier, you can see that images emphasize the copy’s messaging. In addition, images make it look as though the product is super easy to use.
When creating images to accompany your copy, ensure that they’re closely tied to what the copy is describing.
It might help to think about how the image would perform if there was no copy there. Could the image alone sell the feature mentioned in the copy? That’s often a good way to judge if an image is effective.
4. Button copy
Button copy is also a crucial element that you will want to split test on a homepage. After all, the button is probably one of the few calls to action you have on a page. You need to make it count. The error most people make is to focus on button copy first.
Either way, effective button copy ensures that people do what you want them to do. It’ll get people clicking onto your signup page so that they can eventually click the buy button.
When it comes to creating great button copy, you need to stick to these rules:
- Give it a contrasting color
- Convey value in the button copy
- Make it easy to find
- Don’t be afraid to model others
In other words — give people a reason to click on your button!
As we did before, let’s take a look at some effective button copy examples.
Here is an example from Mailchimp.com
You can see that there are two buttons to be found. Each are very similar. Both are also very easy to see, contrasting heavily with the background color.
However, the most important thing is the copy. Take a look at what it says: “Sign Up Free.”
It’s telling website visitors that they can get started for free. It’s also telling them what to do. Free tends to be the magic word with a lot of these buttons and rightly so.
You have to always remind the visitor what they’ve got to gain.
Here is another example from AWeber.com: “Start Your 30-day Free Trial Now.”
They’ve managed to stuff even more value into their button. Firstly they tell you it’s free. Yet they also address any questions related to, “What’s the catch?”
That’s because they clearly tell you it’s a “30-day free trial.”
This mention reassures potential clients that they’ll be able to use the platform properly, without having to deal with restrictions. It also lets them know why it’s free — because you can only use it for 30 days.
And of course, the button contrasts heavily with the background.
Let’s take a look at this example from 99designs.com: “Get Started Now.”
The button is easy to see. Though the copy on this button does not allude to anything free, it does prompt the visitor to engage with the process. The call to action is strong and encourages you to click.
Something we’ll touch on later is the reassurance copy found below the button. Notice how they mention there is a 100% money-back guarantee. This makes people feel more comfortable with clicking on the button.
Borrow from the best
Brilliant button copy is everywhere. If you visit enough sites you’ll begin to see some patterns.
If you see something you like, try to emulate it into your own website. Then let the results speak for themselves.
5. Advanced split tests
If you want some big wins when split testing a homepage, focus on the points mentioned above in their respective order.
However, if you’ve already taken those steps and want further improvements, or if your page was fine to begin with and you want to A/B test more stuff on your page…
This section is for you. We’ll look at some of the other actions you can take to dramatically boost conversions.
Often the best and simplest way to incorporate social proof into your homepage is to leverage a testimonial. All the homepages we’ve looked at have used a testimonial as social proof.
Because it works.
Test a page with a testimonial against a page without one. Make sure you’re testing different kinds of testimonials too. Ideally the testimonial should come from someone who fits the profile of your target market.
Here are two below examples from Grammarly:
Each one is from someone that a potential customer can relate to — a writer and a journalist.
The testimonial counters the objection that this product is not worth the money or that it fixes a problem they do not have. The individual testimonials describe how the product is better than the ‘Word’ software.
Here’s another example from the AWeber homepage. It details two in-depth testimonials from people who represent the ideal customer.
Testimonials are a great way to boost credibility. Experiment with them on your homepage.
Sometimes people have doubts — it’s your job to chase these doubts out of a customer’s head.
You can do that by using some reassurance copy.
This kind of copy might remind the potential customer that all the risk is on you. It might even remind them that you will not use their email address for reasons that they did not agree to.
Let’s look at an example of some reassurance copy.
The example below is from Leadpages.net, a company we looked at earlier.
Notice how they mention the guarantee and the secure checkout feature below the button.
The 30-day money-back guarantee is great because it lets people know that they’re not risking anything. Leadpages is accepting all the risk — so you don’t have to.
The secure-checkout mention reassures people too. It lets everyone know that dealing with this company is safe.
Here’s another example from 99designs. It too mentions that there is a money-back guarantee.
Be aware, not all websites use reassurance copy because it does not work every time. What matters is that you test it and see how it works for you.
Just ensure that the copy you use deals with some of the doubts the website visitor might be facing.
Layout/website element locations
Additionally, you can also split test the layout of a page. What would happen if you put the social proof elements of your homepage higher up — maybe even above the fold?
You should ask this question for each of the elements that can be found on the homepage. As with everything else, these changes are better carried out when you have data.
The CrazyEgg scroll maps can help you figure out what people are looking at the most on a page. If visitors are looking at something that’s near the bottom of the page — consider moving it up.
Ready to reach more people?
Split testing can be confusing.
Hopefully, you now know which website elements you need to focus on and in what order. As you make these methodical changes, you should begin to see an up-turn in results.
In an ideal world, you want data to inform your decisions. Understandably, that is not always possible. Yet you can still follow a few proven principles and emulate some of the better examples out there, so that you get results.
You’re free to experiment outside of the box and test a lot of unconventional ideas when split testing. However, this article aims to provide you with a split testing framework that’ll help kick you start ahead of the curve.
Once you begin to work in an organized fashion you’ll begin to see how this whole split testing thing works.
Before you know it, your business will reach more people every day — all thanks to split testing.