How can A/B Testing help you with Conversion Rate Optimisation?

There are several methods that internet marketing experts or search engine optimisation programmers can use to conduct site testing.  The method I'd like to examine more closely in this blog is A/B Testing.  This involves creating multiple versions of a webpage across unique URLs on the website, before splitting the traffic equally among them. By monitoring behavioural patterns across the various page versions, the webmaster will be able to determine which webpage achieves the highest conversion rate and implement it on their site permanently. A practical example might be to think of a website that sells DIY Accessories. One of their best selling products is a drill, so obviously they want to ensure the page where customers can purchase the drill is optimised to achieve the most conversions. As a result, the website owner creates three pages – hypothetically they are, www.example.com/drill and www.example.com/drill1 and www.example.com/drill2 – although page '1' and '2' are variations of the original 'drill' page. The testing software used by the webmaster would then split the traffic evenly between the URLs, and the analytics report that each page generates would be used to establish which page achieves the most conversions. To establish continuity in user experience, cookies are used to track which user visited which page variation to ensure they return there when they revisit the site. A/B Testing is normally conducted when substantial changes are being tested across several pages – although Multivariate Testing (MVT) techniques can be used, it is technically very difficult when large scale changes are required. The elements of the website that are often tested during an A/B Testing procedure can include the call to action (word sizing, buttons, placement etc.), headlines, form length, layout, product descriptions, images, and quantity of text amongst others. Because of the quantity of elements that are often changed in this form of testing, a new URL is demanded and this is a defining feature of A/B Testing. This is a perfectly acceptable method of testing the quality of a landing page in order to benefit conversion rate optimisation efforts. However, it often leaves website owners concerned that their prior SEO efforts will be in vain as the possibility of content duplication penalties prescribed by the search engines looms large. However, as is often the case with search marketing and search engine optimisation techniques, the concept of 'intent' is crucial. If a web developer intends to manipulate search rankings by maliciously reproducing content from another domain on their webpage, then they do face penalisation. However, as is often the case, if content is duplicated on a single domain in an effort to improve the quality of user experience, then there is no cause for concern. Hopefully this blog will have clarified some of the basic principles behind A/B Testing, but keep an eye out for the next blog, listing our Top 5 Tips for A/B Testing. If you have any questions about this process please don't hesitate to leave a comment or get in touch.