Have you ever been told to “trust your gut” or “follow your intuition”? It can be great advice for your personal life, but as a marketer, not so much.
Image source: happyfoxblog
During my first year as an affiliate marketer, I was tasked with creating a new landing page for an astrology service. I followed my instinct and created a killer page, used great images, and wrote my copy from scratch. The result? It absolutely failed.
As someone who does not keep up with astrology, I was very different from my client’s ideal visitor. I didn’t truly understand my target audience, and as a result I failed to relate to them.
If you’re a marketer, especially a young one, you will often have to promote offers and products you are not always familiar with to target audiences which have nothing in common with you. In that case, it’s better to rely on data and facts.
Of course, there are a few general best practices you can follow to create high-converting landing pages. We have an entire article on how to create a high converting landing page. However, each industry and offer is different, so you will need to go through some trial-and-error to find out what works best for you. This is where A/B testing can help.
What doing A/B testing for landing pages?
In its essence, A/B testing is the process of comparing two different versions of something to decide which one is better. It’s a common research method that can be used for many different purposes, but for the sake of this article, we will focus on how to use it for landing pages.
The idea is to place two different versions of a landing page against each other and find out which one works better for your specific offer.
A/B testing is a great way to constantly improve and reach higher conversion rates without having to do much. This method will allow you to find what works best for your audience in the most efficient way possible.
How to actually perform an A/B test
Now that you know what A/B testing is and why you need to do it, let’s take a look at how to perform it in four simple steps.
First, you need to choose a landing page tool. Almost all landing page tools have a built-in function that allows you to perform an A/B test with no technical knowledge. LanderLab even offers A/B testing as part of its starter package.
All you need to do is upload the two different versions of the landing page you wish to compare. Tools like LanderLab will then typically send an even amount of traffic to both versions. All traffic will be distributed randomly to avoid any interference with the data. In short, you will get accurate results.
Second, you need to decide on which aspect you’re going to test. The most common elements to test are the headline, call-to-action, the hero image, and the copy. Of course, you can test additional things such as a long versus a short landing page, adding testimonials, key features, etc.
We recommend changing only one aspect per test. For example, in your first test, keep everything the same apart from the headline. That way, you can know for sure that any change in conversion rates is due to the headline. If you wish to test a combination of elements at the same time, that is called a multivariate test (you can read more about it here.) However, a multivariate test is more complicated. If you’re just getting started, a simple split test is highly effective.
Now that you have your landing page tool and you know what you want to test, it’s time for the third step – creating the two landing pages. You’re probably already using a landing page to promote your offer. This will be your “control” version. For the second landing page, you only need to change one element that you wish to test. If that element is the headline, keep everything else the same apart from your title. Here is an example featured on Adoric.
The fourth and final step is to choose a winner. There are two things you need to consider here. When should you end the test? And how do you know which landing page is the winner?
For the first question, most marketers recommend ending the test when you’ve received enough traffic. How much traffic is ‘enough’ will depend on your offer and industry. Others specify a timeframe, for example they let the test run for 10 days before deciding a winner. It can be useful to decide on a sample size depending on how much traffic you’re used to.
Let’s say you will wait until each landing page has received 100 clicks. If you are still not sure what is the right approach for you, there are a lot of free A/B testing significance calculators online. All you need to do is place your number of visitors and conversions for both pages in a calculator, such as the one provided by VWO, and it will show you if your results are significant. If they are not, you will need to wait for more traffic before confidently deciding on a winner.
The second question is, how do you choose your winner? In some cases, it’s very obvious. For example, one landing page has far more clicks and conversions than the other. Other times, the situation may not be so clear-cut. Maybe one landing page has a lot of clicks, while the second version has less clicks, but more visitors converted. In such cases, you need to evaluate these results against your goals. What did you want to achieve with this landing page? As a rule of thumb, conversion rates are a better measure of success. After all, your final goal in each case is to make money.
Once you’ve ended the A/B test and chosen your winner, the landing page tool will send all future traffic to your chosen landing page.
Repeat
If you’re wondering how many times you’re supposed to do this, the simple answer is that A/B testing is a never-ending process. There is always room for improvement, and you can continue testing different aspects or different versions.
Even assuming that you have found the perfect landing page that works for you, times and trends are constantly changing. What works for you today may not work a year from now. To ensure that you stay at the top of your game, you need to consistently seek growth.
Conclusion
Although there is a general formula behind top-performing landing pages, you need to undergo some trial-and-error to find out what best works for your offers. The most effective way to do this is through A/B testing. All you need to do is put two different pages against each other on a landing page tool, decide on a winner, and repeat this process in order to continually improve and get higher conversion rates.
