PerformanceININside Performance Marketing
JOIN THE PERFORMANCEIN Join the PerformanceIN Parnter Network
International Performance Marketing Awards
Change in Deadline – Entries Close 02 August
Download Your Entry Pack
Can A/B Testing Maximise the Value of Website Traffic?

Can A/B Testing Maximise the Value of Website Traffic?

PerformanceIN

There are several ways that businesses can use A/B testing to help maximise the value of website traffic and in turn their sales. Here are a few tips on how to be successful in increasing your website traffic.

Seamless online platforms are key in delivering benefits to both businesses and their consumers. With the increasingly competitive nature of the e-commerce landscape, brands and businesses are continually trying to out-smart one another by increasing website traffic to win the war for each customer.

This is where A/B testing platforms can become priceless assets to businesses of all industries and sizes. As most of us are aware, A/B provides retailers businesses with the ability to compare variations on their platforms, statistically analysing the results of experiments to gain an insight into the most valuable layout for their store. Along with this, the latest platforms also enable users to apply more complex statistical algorithms that take past outcomes into account, providing more reliable results in a shorter time-frame.

There are many ways that businesses can use A/B testing to help maximise the value of website traffic and in turn their sales. Here are a few tips on how to be successful in increasing your website traffic:

Increased customer personalisation

The first way A/B testing platforms can help you make the most of website traffic for your business is through boosting customer personalisation. It is well known that if a website is populated with products, services and content that the customer is interested in, they are more likely to continue on the path to complete a purchase or sign up.

However, if you have invested money in a series of paid adverts, it is likely that the messages will be targeted at a variety of different customers. If one customer clicks on an advert for a particular accessory - let’s say a necklace - it is best that they land on a page with content that is consistent to that message.

This is simple on platforms that provide dynamic text features. A Dynamic Text plugin allows you to drag and drop it on your page, and then simply add the message you want to appear into a parameter on the end of the URL. When a customer clicks on this advert, the message you put in the URL will display on the page they arrive at.

Another personalisation technique is a geolocation plugin. This displays the visitor’s location when they arrive on your page, further enhancing the level of personalisation and increasing the likelihood of a sale. Implementing this on an A/B testing platform allows retailers to test and discover which personalisation technique works best for their customers.

Meeting specific targets and goals

Setting precise goals for e-commerce platforms enables retailers to measure how effective the changes are that they are making during A/B testing. Some goals that can be set include:

Page view goal - this allows businesses to track the number of times each page has been viewed.

Click view goal - this allows retailers to track the number of clicks on each page that have been modified.

The revenue goal - enables businesses to track and measure the success of their  A/B test by tracking the value of each purchase made by customers.

Improving agility in your strategy

New algorithms implemented into A/B testing and notifications platforms can now enable businesses to automatically allocate traffic to the best performing variant in an A/B test. Previously, those testing their website would have to wait for the end of the test to discover which version of their website was producing the highest revenue.

This meant that for the duration of the test you would be losing out on sales. Furthermore, once the test was over you would have to switch off the experiment and manually implement the changes before reaping the rewards - all in all, it could take weeks.

Now, with various features results can be taken into account before the test has finished: necessary changes can be actioned instantly to maximise visitor value. This allows you to produce something that will continually recalculate the best performing variant and will therefore automatically push more traffic to it.

A hybrid approach to statistical significance

Behind A/B testing lies a whole world of statistical debate. An experiment must run for long enough for a statistically significant (reliable) result to be achieved, to make sure that the new colour scheme or positioning of reviews is actually responsible for improving conversions.

There are two main statistical approaches to A/B testing: the Frequentist approach which is generally faster, and the Bayesian approach which is arguably more reliable but slower. Both have their benefits, but fully understanding them takes more time than most business owners have. That’s why the best A/B testing platforms take the best elements of both in a hybrid approach to statistical significance, which will give you more accurate results, more quickly.

The ‘hybrid’ approach considers how results are changing over the course of the test, whilst also taking into account the differences of consumer behaviour on different days of the week. Using a platform with carefully designed algorithms will allow you to run tests quickly and reliably.

However, subscribing to an A/B solution doesn’t always guarantee results for some companies. It is important to note that selected A/B testing platforms will often sell with a 12-month subscription in mind which can lead to misinformation - be wary of guarantees. The best way to optimise conversions is research the wide range of solutions available and ensure that the chosen solution is able to meet the requirements for your business.

Before you research a solution, decide: do I want a simple editor or one that will allow me to code the changes in? Do I want intuitive, drag and drop style features or more complex elements? Do I want to AB test, split test, or run a multivariate test, and which platforms support these? How much support do I want when creating and running tests?

Don’t leave conversions to chance!

It is clear that A/B testing is worthwhile for businesses trying to maximise the value of visitors to their website. It forces you to pay closer attention to how your business appears to hard-earned website visitors, and the process of experimentation may also highlight other elements of your business that may need revamping. It can also help with team building within the company as it requires input from a variety of team members regarding new ways of engaging customers online.

Continue the conversation

Have something to say about this article? Comment above or directly on Facebook, Twitter or our LinkedIn Group.

Philippe Aimé

Philippe Aimé

Philippe Aimé is the CEO and founder of Convertize and has an extensive background in sales optimisation, with a passion for neuromarketing and consumer behaviour. 

He launched his first travel website in 1998, which is where he learnt to develop skills in web-design and optimise SEO. This was in addition to his consulting job, where he worked with companies such as Alstom and Renault.
 

Read more from Philippe

You may also like…