More often than not, in order to find the best example of something, you need to contrast it and compare it with other things from the same category. Affiliate marketing is no different. Even with so many different variables, the comparison of data is what leaves you with the best of the best.

You can a/b test almost any variable including traffic sources and affiliate networks. However, once you do have that one perfect offer, it’s time to find the perfect landing page.

So, if you want to make sure that your marketing efforts are not in vain, read on to find out everything you need to know about a/b testing your landing pages.

Do you need a landing page for affiliate marketing?

Let’s get this popular question out of the way. Beginner affiliate marketers might feel overwhelmed with the number of campaign elements they need to take care of. Research and find an offer, research and pick a traffic source, spy on the competition and pick a creative. And then find and tweak a landing page to have the highest chance of scoring conversions.

Unfortunately for the lazy affiliates, in 90% of cases using a lander would bring them better results. Pre-landers have a lot of functions and sometimes their main purpose is to engage visitors while other times they will serve as a sales page that can discuss all the details of why you need the product, what it is and why you need to purchase it right now.

In short, there are always many marketing approaches to every offer. But if you want to make sure you have the best chance of reaching your goals, you need to test. And the element that makes the biggest difference is always the landing page.

Don’t believe us? Ask the Internet. Oh, wait. It’s already been done. And everyone on Quora will tell you the same thing. Even if you are placing your affiliate link in a blog post about the product, using a landing page is still likely to bring you better results.

Why is split testing important in affiliate marketing?

Let’s face it. Split testing is important everywhere, as long as achieving the highest performance helps you achieve your goals. The thing with affiliate marketing is that the performance of your campaigns depends on the performance of each of many smaller elements. 

You should test similar affiliate marketing offers to see which one performs better. The one that loses is most likely the one that was losing you money. Good riddance. Now onto the next component. You should test targeting options. That’s usually the logical approach to running campaigns that are not restricted to one GEO or device. You begin with a broad run-of-network campaign and finish with a well-optimized source or target campaign.

As a beginner, split testing is the best approach for quick and conclusive results. However, as you begin to grow your business, you’ll find that multivariate testing is one of the best practices for maximizing the number of conversions you get.

How do you do a split test for affiliate landing pages?

Split testing in affiliate marketing usually requires additional tools. While testing targeting-related variables (such as GEO, device, OS, OS version, carriers, and day-parting), it’s possible and even advisable to create separate campaigns on the traffic source level. But that’s not the case with a/b testing landing pages.

As a beginner affiliate marketer, you might not need anything other than a traffic source and an offer. But a/b testing is one step further. The third tool/service you’ll need is a tracking solution and that’s also what will allow you to split test your landing pages. There are of course other options so let’s discuss the tools needed to start a/b testing your landers.

A/B testing tools

Ad tracking solutions such as Voluum, Adsbridge, BeMob, and many others, give you the possibility to use many landing pages within one campaign. If you want to test a different version of a lander, all you have to do is add a new flow and start gathering data on that fresh option.

Landing page builders are your second best option. While this is a completely different tool than a tracker, some lander builders such as Convertri, Landingi, or Clickfunnels allow you to set a/b testing within the platform.

While using a tracker for a/b testing your landing pages allows for greater flexibility and optimization, some traffic sources such as Facebook might appreciate ‘seeing’ only one outgoing link from your ad. And that can only be achieved by testing a/b landing page options within a lander builder.

Additionally, the internet is full of helpful calculators. If you have any doubts about the prospects of your a/b testing samples, you can take a look at these links and compare the results yourself:

• A/B Testing Significance Calculator is a simple tool designed by a veteran affiliate marketer Neil Patel. It has a simple purpose – to calculate the conversion rate of your landing page based on the number of visitors and number of conversions.
• Bayesian A/B Testing Calculator is nearly identical to the first one, except for the additional benefit of providing more detailed data.

Of course, these are only the two most popular options. If you’re an affiliate marketer who sticks to paid ad exchange platforms, these are the tools you are most likely to need for a variety of purposes. However, that list is by no means exhaustive and there are more things worth mentioning.

Advanced testing tools for email marketing, social media marketing, SEO, company web pages, or product descriptions are different and sometimes more specialized. According to HubSpot, trusted and renowned brands can benefit from using tools like VWO, AB Tasty, Crazy Egg, or Convert.

These are not only capable of a/b testing landing pages but they also offer segmentation tools to split visitors based on their behavior, JavaScript events, cookies, and sometimes even weather. So, if you are running a medium to large-sized business, you should read up on HubSpot’s expert insights and pick a capable a/b testing solution for your company.

What can you a/b test?

Yes, you can a/b test your landing pages but what does that even mean? Taking two completely different pages and running a/b tests is a valid option but it’s not what you’d normally opt for.

Here is a list of a/b tests you should run on your landing page to ensure the highest possible conversion rate for your campaigns:

Using a landing page vs. direct linking – we’ve said it before and we’ll say that again. 90% of offers work better with a landing page. However, it still might be worth testing both versions of the funnel. Just like direct linking instead of using a landing page can decrease your conversion rate, going the other way around when it’s not the best option can significantly lower your prospects. Always run a quick test before deciding on a strategy.

Call to action (CTA) button – a call to action is one of the most important landing page elements so running a/b tests between different versions is a must. If you think changing ‘Click here’ to ‘Get your deal now’ can’t possibly affect your campaign conversions then you are in for a surprise. Change the copy, the shape, the frame or shading, and the positioning as well. After all, this is where your affiliate link is so it’s crucial you pick the most clickable version.

Header – this is one of the most important elements of your landing page as well. A/b testing your landing pages with different headers will ensure that the first thing the website visitor sees is exactly what they are looking for. The bounce rate of your landing page might depend on this element so if people quit your page after seeing the subject line, you might need to do some optimization. Or split-testing.

Colors and design – your audience might be optimized perfectly for the kind of offer you’re advertising and the content of your website might be professional and informative… but if your design is horrible you can forget about conversions. The user experience should be among your highest priorities and presenting your visitors with a bold color scheme might require a/b testing to see if it sticks.

Product photo – the goal of your campaign is to make the product website visitors believe they need to make a purchase/sign-up/register. While abstract graphic elements can significantly improve the way people perceive your site, choosing a product photo to complement your content is another one of the a/b tests that can really make a difference.

Be careful though, using logos is not allowed unless you have the permission of the brand. So, as much as you’d want to present site visitors with the perfect product photo, you need to remember to check the campaign guidelines of your ad network before you run your a/b test.

Form – Last but not least, split testing the form of your landing page might be the most fun part of affiliate marketing. Apart from seeing the green numbers in your tracker, of course. If you want to test the potential of your campaign, you might want to test different approaches. And that would require creativity and sometimes creating content from scratch.

Most of the time, there isn’t one good way of promoting, for example, a diet pill. Affiliate marketing allows you for a lot of flexibility in a/b testing landing pages, ads, funnels, and anything in between. The form of promotion is up to you and reaching your goals often depends on how well you research and plan your a/b tests.

On a larger scale, you can split test promotion methods such as email campaigns vs. social media campaigns, push ads vs. pops, and blog content vs. a YouTube video. On a smaller scale, you can test version A of the page which would have one unique selling point (USP) against a version B which would have a completely different USP.

Your conversion rate might also depend on whether you choose a spin wheel or a quiz. But let’s discuss some examples now.

A/B testing examples

In a/b testing, it’s important to change only one parameter at a time. Sure, you will most likely have to run an a/b test on many different versions of the funnel but doing everything at once is not the ideal approach. You need clear data and you’ll have to get it one step at a time. Your steps don’t need to be taken in a particular order but here are some example a/b tests you could run.

★ HubSpot’s red CTA button vs. green CTA button split test was conducted to see if a simple landing page would gain more conversions depending on the color of the CTA button. The content of the page, including the copy on the CTA, the subject line, and the graphic elements all stayed the same.

The results: The red button outperformed the green button by 21%.

★ Convert’s benefit-driven headline multivariate test aimed to a/b test two new variants of a headline copy. Additionally, the original headline was used in the test as well. In total, there were three different site versions; one focused on the social proof, the second on the benefits, and the third was the original with no headline, just the website title.

The results: 27% increase in conversion rates and the “Add to cart” button click-through rate increased by 13%. Also, the social proof version (picture below) won.

★ VWO’s two magic words split test was very simple. The company called Viadeo took two CTA buttons and compared them against each other. Version a was green with a ‘Sign up now!’ button while version b was green with a ‘Sign up now!’ button… and ‘It’s free’ added right next to it. Can you guess which landing page was the winner of this split test?

The results: Version B with the words ‘It’s free!’ had a 28% higher conversion rate.

★ CrazyEgg’s Wall Monkey landing page split tests focused on improving the landing page elements that didn’t get enough attention. They analyzed user behavior using heatmaps and conducted a split test between the original (version A) and version B with a new design. Subsequently, another a/b test was conducted to see if replacing a slider with a search bar would drive more engagement.

The results: Both landing page tests caused a 550% increase in their conversion rates.

★ If you’re curious to read more about the surprising results of a/b testing, you can take a look at WordStream’s 24 of the Most Surprising A/B Tests of All Time and get inspired to produce your own a/b test examples.

Almost anything can be tested, so if you have a website or a landing page you start testing different versions of it to see if you’re not missing out on some conversions because of a wrong approach.

You never know what your page visitors might like more. Until you a/b test it.

A/B testing vs. multivariate testing

One of the a/b testing examples above mentioned multivariate testing. You might be wondering what that is. In simplest terms, it’s testing more than two versions of whatever you are testing (content, CTA, form, colors, etc.).

Of course, you can conduct split testing one pair at a time. An a/b test of something is usually not a standalone event. Unless you’ve gone through a whole lot of split-testing and you’ve gathered sufficient data to arrive at the grand finale of a/b testing. Then two options are definitely enough.

Multivariate testing allows you to test more versions of your landing page simultaneously. So, you can test three variants of a CTA, five content layouts, or 10 color variations. The possibilities are endless. You can even test multiple landing page elements at once.

The benefit:

✔︎ It’s quicker than a/b testing in the case of landing pages with many questionable elements.

The downside:

✘ You need huge volumes of daily traffic to obtain statistically significant data.

Multivariate testing is ideal for intermediate and pro affiliates as well as businesses that receive substantial amounts of daily traffic. However, you shouldn’t pick one or the other. Chances are that only combining testing a/b options for new campaigns and multivariate testing for landing pages that you’re using with a tried and tested offer can help you reach your goals.

Conclusions

If you want to make sure that visitors coming to your sites or landing pages have the best experience you should do a lot of a/b testing. But let’s face it, as much as user experience is an important factor, it’s the conversions that we care about. And a/b testing is how you optimize it.

We’ve talked about the importance of a/b testing, tools you need, calculators that might come in handy, a/b testing examples, and more.

You want your landing pages to bring conversions, sell things and be lead magnets. The only way to do that is to split testing your two ideas. Pick the best one and introduce a new variable. Conduct more a/b testing until you’re sure you’ve got the right thing. Expand split testing to offers, ads, networks, traffic sources, and anything you can think of.

In the end, you’ll find a funnel that will take your earnings to the next level.

Drive Traffic to Your Landing Pages
Join Zeropark Now!

Magdalena Bober