9 Split Testing Tips that Help Dramatically Improve Conversion Rates
Split testing has become one of the most important aspects of a successful marketing campaign.
Always be testing, or ABT, has become as synonymous with marketers as always be closing is to sales reps.
And there is a good reason for this. That reason is actually nothing more than common sense.
Ask yourself this question…how are you going to know what works if you do not put what you think works against something else?
That is a simple question that has an even simpler answer. You won’t.
As the marketing firm Rand Group tells us, “Marketing without a baseline is taking a shot in the dark.”
But how do you create that baseline? How do you perform split testing that can actually help you improve your conversion rates?
What should you test? How do you make sure that you are not losing sales in the process of split testing?
Let’s dive into it. Here are nine split testing tips that can help you dramatically improve your conversion rates:
#1 – Create a System
One of the biggest difficulties that marketers run into with split testing is that they do not have a system for testing different types of copy, headlines, emails, etc.
This is dangerous because it leads to confusion which then leads to wasted time.
You can remedy this potentially confusing scenario by creating a strong system for exactly what will be testing for each of your marketing methods.
If you are sending an email, decide beforehand what will be tested. Maybe you will test the subject line, time or day that you send, or which images to place within the email.
If you are writing content, maybe you will decide to test different headlines or where to put your call-to-action.
No matter what you are testing, having a system in place beforehand for performing those tests will ensure a smooth process and will save you a ton of time down the road.
#2 – Focus on Areas You Are Already Performing Well
While it may seem like a good idea to go out and perform tests on the marketing methods that are your lowest performing traffic generators, this can be a serious waste of time.
Using marketing campaigns or testing platforms that have too few data will probably mean spending several months completing the test.
Plus, even a 1% conversion increase for a platform that is only bringing in 50 visitors per day will mean you are gaining one person every two days. That is definitely not efficient.
On the other hand, your campaigns that provide plenty of data may only take a few weeks to gather reliable test results.
A 1% conversion increase for a method bringing in 1000 visitors per day means you are gaining 10 per day. While that still may not seem like much, it is 2000% increase over your other method.
#3 – Try Different Types of Social Proof
There is no debating whether or not social proof is important for conversions. What is debatable, however, is what is the best way to go about providing that social proof.
Three of the most often used methods include case studies, testimonials, or just cold hard numbers.
Your best bet is to test all three and see what works best for you. Just be sure that you are including an image as KissMetrics points out that social proof works significantly better with pictures.
Marketer Anumpam Bonanthaya talks about how to use case studies and testimonials effectively in this article.
#4 – Test Emails Vigorously
Emails are by far the easiest thing to test because they offer the opportunity for you to truly randomize two different sides.
If you are testing conversions from visitors, you are going to run into a lot of repeat visitors that influence the overall data that you will receive.
One of the biggest things you will want to test with your emails is the subject line. You should be committed to creating captivating subject lines in the same way that you would a headline for an article.
While many marketing experts will also tell you to perform tests about the time and day of the week to send your emails, common sense tells us that this does not play as big of a factor as we think.
Think of your own email response activity. Do you open an email simply because it came to you on Monday at 8AM?
More than likely not.
Instead, focus on testing subject lines, short and long copy, and different call-to-actions within your email. This is where the most significant increases can come from.
Becky Myers of Sales Cycle wrote a simple guide to email testing that you would be wise to take a look at.
#5 – Try Different Font Sizes and Types
Philosophy expert Eric Michael Weisenmiller did an extensive study about the readability of online text that gives a thorough explanation of how people react to different types of text.
In it, he found that Arial 12-point font is best for reading speed and comprehension.
You will only know what works with your audience by using different font sizes and types within your copy.
#6 – Test Your Call-to-Action
Testing your call-to-action is an easy, yet great way to dramatically improve conversion rates.
There are a variety of different things that you can test here which include:
- CTA button vs. simple link
- Color of CTA button
- Navigational cues to CTA or no navigational cues
- Position of CTA
Your call-to-action is the driving force behind getting your visitors to convert. Do not take it for granted.
#7 – Test Content on Different Pages
The content featured on the different pages throughout your website will heavily influence how long someone stays on your site.
If you are having trouble keeping visitors at your site or converting, testing different pages like your home page, about page, or contact page can help you figure out whether or not this is a large issue or not.
A lot of times visitors will come to your site with the attitude that they are not yet ready to convert. Writing solid content that appeals to them can keep them there long enough to convert.
#8 – Try Different Graphics and Videos
Graphics and videos are becoming an essential aspect of conversion. Today, not featuring an image, graphic, or video can be deadly for conversion rates.
Oli Gardner from Unbounce has a great infographic that gives marketers an idea of the different things that they can try with graphics and images.
The important thing here is that you do not get so involved in testing your graphics that you stop testing other things.
While certain small changes can definitely be effective, be sure that you are not wasting too much time on tests that yield insignificant increases.
#9 – Don’t Always Listen to the Experts
While many so-called marketing experts will tell you that you have to do this or do that, you need to understand that you have a unique set of visitors that no one else in the world has tested.
You will only know how that audience reacts to different things if you test them.
Be unique with your content and test those things that you feel will resonate with your specific visitors.
The Final Word
Split testing can get extremely confusing and is a time consuming process. It is clear, however, that is a worthwhile practice that can dramatically improve your conversion rates.
Focus on creating efficient systems that can save you time down the road with your tests and do not be afraid to try something that has never been tried before.
Hopefully these tips will serve as a guide to help you get started towards becoming an expert split tester. Good luck!