Blog - Latest News

Split Testing Landing Pages – A Beginners Guide

If a landing page (a page designed specifically for first-time visitors) is crafted well enough, it will almost certainly result in a better conversion rate than simply dumping people into the homepage of your website. In short, if a landing page is of a superlative design, has undergone numerous changes and rigorous testing in order to boost its performance, then your chances of successful conversions will be far higher than sending new visitors to a general page that is not optimized to the needs of that visitor.

Generally speaking, there are two broad techniques to test the performance of your landing pages; Multivariate Testing and A/B Split Testing, and before you go pitting one landing page against another, you should first decide which method to use.

Understand that both A/B testing and multivariate testing are used to compare a set of page “blueprints” or creative designs, but the depth and intensity of testing differ for each method, and the need to use one over the other will depend on the extent you want to optimize your landing page. There are times for each.

A/B Split Testing is essentially as it sounds.

One page (A) is tested against a completely different version (B), with traffic split between the two so that visitors experience entirely different web page content on both versions, and visitor actions are then monitored to determine which version produces the highest conversion rate.

Where multivariate testing will test a variant of a specific element right down to the wire for the sake of refinement, split testing is simply used to test the overall mechanics of a web page, comparing one idea against a radically different one in order to establish a framework that performs best. It’s like testing the foundations of a house before you commit to amending the subtler, finer details later on – and the reason why A/B split testing is typically the better method of testing your web page whilst in the early stages of your campaign. Even if your landing page only has a few conversions per day, you cannot yet use more advanced tuning methods, but with the proper selection of test variables you can still achieve significant results in split tests, despite having low traffic, and ultimately improve your conversion rate.

Often A/B split tests are carried out on an ongoing basis. This format is commonly referred to as ‘champion/challenger’, where your preliminary design is tested against a re-designed variant comprised of radical changes in order to see which performs better, by way of measuring conversions. This isn’t a case of subtle optimization, but establishing an overall framework that works best one element at a time.

Image: Split testing – 2 landing pages split 50/50 in an A/B split test

Depending on traffic velocity (the volume of traffic), you can test any number of alternatives (A/B/C/D etc.) with traffic split right across the board to see which version works best. Testing yields the most valuable results when you test repeatedly, ensuring that the testing conditions are consistent. Split testing is simple to implement with Clickthroo, and you can dramatically challenge any assumptions about the ‘best’ way to design a page at any time during your campaign.

There are any number of elements to test in order to find what works best, but initial test considerations should include:

• The main headline – Typically a break down of your product, service or any other offer.

• Some words might work better than others, and get the message across more clearly, whilst some words can be optimized for the keywords used from the search engine results, ensuring the visitor is immediately in sync with your message before they’ve even landed and providing them with a stronger sense of affirmation in what you are offeringThe call to action – This is the text within the button or text link that represents your conversion goal. You want your visitor to hit this button to complete conversion, so ideally the text needs to contain a clear, concise message that not only properly informs the visitor where to click but also encourages them to do so.

• Button color – Very self-explanatory, but a science in itself. Different colors have different connotations. You want your visitor to associate that color, and by extension that potential transaction, with something positive, all the while pertaining to your brand identity or the product that is on offer.

• Graphics and photography – Using an image to depict your product or offer is a clear indication to the visitor that they’ve arrived at the right place. However, a picture is worth far more than just a thousand words and depending on the context of the image, a visitor may or may not complete conversion. You can build an entire narrative around a single image and provoke an emotive response from the visitor – one that either prompts or dissuades them from completing a conversion.

• Video – Using video as a convincer can be a great way of quickly introducing new visitors to your product or service, but only if presented well. Split-testing your explainer video can produce huge upturns in conversion rates very quickly.

• Form length – In the interest of keeping your visitor fully engaged, it’s important to minimize the amount of fields that visitors are required to complete. However, if your aim is to capture as much data as possible (with lead generation, for example) then cutting down on the number of fields may not be an option. Instead, you can test different variations to the order these fields are completed, varying the amounts of information each time, and stick with the iteration that performs best.

• Positioning – The on-page position of your landing page elements (images, call to action buttons, text, etc) can have a dramatic effect upon your visitor behaviour. Although tools like CrazyEgg provide detailed analysis on typical user behaviour patterns, the need to test should not be ignored.

Image: Some of the elements to test on a landing page

Testing these elements will result in conversion data that will provide you with a clear idea of which variants perform better to different keywords, traffic sources, and other, more granular performance metrics based on factors such as the visitors location, operating system, browser, time-of-day, device, etc.

Clearly superior versions of your site will become apparent with the Clickthroo internal tracking platform, and the determination of which changes are responsible for such any improvements will be abundantly obvious.

Try the Clickthroo Landing Page Marketing Solution: Free For 14 Days 

(Landing page builder, integrated tracking platform, A/B split testing, traffic segmentation, template and image libraries, integrated trafic sources, optional affiliate marketing module, and much more…)