Has anyone ever told you, “There’s always room for improvement”? This applies perfectly to websites and landing pages. Your page can never convert too many people into customers. Split testing, or A/B testing, is an easy way to learn more about what your customer’s are looking for so you can increase your conversion rates.
BUT FIRST, WHAT IS SPLIT TESTING?
This method of testing allows you to compare two versions of a webpage in order to see which one converts more. This type of testing lets you remove the guesswork when trying to determine the best way to optimize. From this, you can make the changes and adjustments needed.
There are many advantages to split testing, which is why it’s a popular tool among marketers. Because you are testing two different versions, there is very little effort that has to go into the design of the test. All you have to do is come up with different versions of the pages you want to test and put them up online. Once this is finished, you just have to split up the traffic. As you can see, the ease of design and the simplicity of the test is a huge advantage. As mentioned before, it takes out the guesswork. Split testing is a great validation tool. It gives you a clear image of what works the best, without incorporating too much statistics.
Another advantage is the ease of analysis. Due to the design of the split test, not a lot of analytics and statistics are required. You first have to determine what you are hoping to achieve. Once you establish this and have created the test, the last step is to decide if you are satisfied with one of the versions. Split testing doesn’t require a lot of time or effort when looking at the results. It’s simple: did you reach your goal? This simple question is all you really need when determining if the test is finished or not.
There are a couple of drawbacks that you should be aware of. One of these being there is a limited number of elements that can be tested at one time. You might have a number of problems but remember you can only test one element at a time. If you test more than one, you won’t know which one was the contributing factor to the increase of conversions. There is good news, you can always continue to split test your webpage. The best thing to do is figure out which element is your top priority and continue to work through your list. Even though split testing can be a lengthy process, it is still a favorite among marketers. Plus you can never stop improving your webpage.
It is also important to keep in mind, split testing, cannot help businesses come up with a solution for every single goal. Not every goal can be measured through customers’ actions online. These goals such as brand recognition or PR efforts can’t be measured. Think of it this way. Imagine your company had some coverage by the local press. A journalist called the CEO for a quote. You don’t know if the journalist visited your site before the interview or not.
Split testing is not a difficult test to set up. You can go one of two routes: Google Analytics, which is free and can be done yourself or you can do a paid solution. Sometimes creating and setting up your can sound like a daunting task, but Google Analytics made it simple. But if this still seems daunting or you don’t have to the time, no worries, there is still the paid solution. Optimizely is a great cost efficient option for a paid solution. Plans with Optimizely start at $17 with advanced features available at a higher price. Their software is excellent but if you are starting from scratch, you might want to look into Unbounce. This software is an excellent choice if you are new to analytics and marketing. All of the templates that are offered can easily be customized to match your business.
How long you run your test is ultimately up to you. They can run anywhere from a few days to a couple of weeks. You don’t want it to be too short of time, because you won’t get accurate results. At the same time, you don’t want to run the test for too long because you won’t be able to implement the changes. Running a split test can also give you skewed results, especially if you don’t have much traffic. Just make sure you are confident in your findings. It is also important to take into consideration holidays because most webpages will get a fluctuated amount of traffic. This won’t give you the most accurate results. So make sure you’re testing during a time in the year that gives you an accurate result.
THERE ARE FIVE MYTHS WE WANT TO ADDRESS BEFORE WE GO ANY FURTHER.
You should test everything. Yes split testing can be used for many different elements on your page. You should want to change and make your site better. But with that being said, there are two types of pages you shouldn’t even bother to test.
- The broken page, there’s a pretty good chance it just needs a complete new design.
- A page that has little to no impact on revenue.
Bottom line, if you’re testing just for the sake of testing, there’s no point in continuing with the test. All of your tests should relate back to the overall goal or idea.
Split testing will make my conversion rate go from awful to great. Don’t rely on split testing to solve the problem of conversion. Once again, it’s just a tool. It is there to help you validate the cause of the problem. It brings awareness to you. Maybe you have a traffic problem rather than a conversion problem and in that case, split testing won’t do too much for you. Once you find the cause, you will have to continue to research, test and analyze.
Split testing is expensive. Split testing doesn’t have to be expensive. In fact, many marketers like this type of testing because of the low costs. Google Analytics allows you to set up a test on your own for free. There are paid options too but these are not crazy expensive numbers. There are plans that start at $17.
Split testing and optimization are the same thing. This is wrong. Split testing is still reasonably new but is continuing to grow in popularity. Somewhere along the lines, it’s gotten a little bit confusing on the relation between split testing and optimization. It is important to remember split testing is just a tool to help you optimize your site later on. Conversion rate optimization utilizes data to help improve their customer’s experience so in turn the person converts to a customer. Split testing is just a tool that helps you validate your strategies and ideas. It tells you what is and isn’t working. From there, you make changes, which will ultimately optimize your conversion rate.
The results are immediate. You might see changes right away but that doesn’t mean these results aren’t skewed. Like any test, the factor of time is important. You can’t stop the split test after a few days. It doesn’t give you broad enough information about the results. Once you give it time, you can then start to make changes.
IF YOU CAN CHANGE IT, YOU CAN TEST IT…BUT…
With this being said, remember you don’t want to test every little thing. Only test the things that will ultimately have a huge impact on a visitor’s decision to convert.
Websites and Landing Pages
Almost every one of your customers will go to either your website or a landing page. With this being said, it is extremely important that your site is optimized to its full potential. Websites and landing pages are a great place to start when split testing.
— Headline. This is the first thing your visitor sees. It has to be captivating so it can draw someone in. As we all know where, there are many ways to say one idea. Some words are better than others. Headlines are not just about the words you use, but also the size and color eve. This is why headlines are a great element to test. There are so many parts to a headline that can be effective.
— Call-to-Action. First thing first, test the wording not the color. The wording is ultimately what impacts someone’s decision to covert. Similar to the headline, the wording used can positively or negatively affect the decision on the visitor. For example, most people find the word ‘submit’ does not convert as much ‘download’ or ‘sign up’. Also test to see if the placement of your headline makes a difference.
— Images. Test the images that have a direct correlation to your sales. Does the image of your product or an image of a person convert more people? Images play a vital role when trying to convert. Humans are visual creatures hence why the images on your site or landing page are important. Use the split testing to figure out which one benefits your website or landing page the most.
— Sales copy or product description. Does your copy or description need to be improved? This is some of the most vital text on your website or landing page. Testing this is extremely important because of the impact it has on people. This text helps someone make the decision on whether or not they want to purchase your product or sign up. The way it’s worded will definitely impact the person who is visiting the page or site.
Even with the constant increase use of social media, emails are not dead. Don’t forget to split test your emails to make sure it’s reaching and converting as much as they can. Tools are built in to CampaignMonitor and MailChimp.
— Subject. Many of us will read the subject line and then make a decision of whether or not we open the email. Some of us don’t even make it this far; we just automatically delete the email. This is why getting the correct subject line is essential.
— Welcome emails. Test the time at which you send them to see if it affects how many are opened. Send half instantly and send the other half two or three hours after the person signs up.
— Testimonials. Do you include them or leave them out? People always like to hear a third party confirm what is being said.
— Timing. When is the best time to send out an email? Some people will say you should be sending them out anywhere in between 8:00 p.m. to 12:00 a.m.; while others say 8:00 p.m. isn’t the best time. This is the perfect thing to split test because truthfully, it depends on your business and your consumers.
Funnels have to be used when a product or service is being sold online. Funnels, as you can imagine, are extremely important and have to be set up correctly. Split test to see if the order of steps is affecting your conversion rate. How do you motivate people to buy your product?
— Order of steps. The flow of your pages can definitely affect conversion rates. This can also mean, once of your designs has fewer pages. Maybe your audience doesn’t want to click through five pages just to buy a product.
— Number of steps. Shorter is usually better. You want to keep your funnel simple and to the point. If the step doesn’t add value, you can probably leave it out.
REAL-LIFE EXAMPLES OF A SUCCESSFUL SPLIT TESTS
Electronic Arts, EA Games, is an American developer, marketer and producer of video games. You probably know or have heard of their games. Some of their post popular being FIFA, SimCity, and Need for Speed.
Their goal was to increase their revenue. Thanks to split testing, they were able to sell 1.1 million copies of SimCity 5 during the first two weeks it was released. 50% of those sales were digital downloads.
While EA Games was preparing for the release of the game, they used promotional material to help drive more pre-orders. On the top of the pre-order page, the offer for SimCity 5 was displayed. It was easily accessible for visitors, but even so, EA Games weren’t receiving the numbers they were hoping for.
EA Games decided to release variations of the promotion. One variation removed the promotional offer all together. The results were very surprising. The page that didn’t include the offer drove more sales. There was a 43.4% sales increase on this version. EA Games found that people just wanted to buy the game.
Kiva is a non-profit organization that lends money to low-income and students via the Internet. They are currently working in 82 countries to “connect people through lending to alleviate poverty.”
Kiva’s goal was to increase the numbers of donations from their first time visitors via their landing page. Their idea was to add more information in hopes of receiving a higher number of donations. An information box that provided statistics and FAQ was added to the bottom of their page. As a result, their donations increased by 11.5%.
Split testing, as you can see, is a simple and easy way tool that can be used to help you reach the conversion rates you want. There are many things that can be tested but it is important to remember that while this is true, you don’t want to test every little thing. Test the elements that will have an impact on conversion rates. The easiest thing to do is to figure out what your top priorities are and go from there.