Maximize Conversion Rates With A/B Testing

A/B Testing to Identify the Best Design and UX Solutions

A/B testing is a technique of a multivariate comparison that helps to identify the best design and layout of a website. The test provides valuable insights of users’ behavior that serves as a ground for improving on the usability that eventually leads to increased conversions and revenues.
 

It is recommended to run a test at least once a month with specific attention to headlines, calls-to-action, landing pages and so forth.
 

To perform an A/B test you will need to measure a conversion rate with the objective of increasing it. The most obvious form of conversion rate is sales and can be worked out as the number of sales per one hundred visits.
 

For example, if you average 3 sales per hundred visits your conversion rate is 3%. Raising this conversion rate from 3% to just 3.5% would mean a 35% increase in sales.
 

This illustrates how important the conversion rate is and outlines the process as a vice-versa process from setting the result before moving on to finding the best solution to achieving it.
 

What is tested?

As the e-commerce field rapidly evolves, so does web design and site architecture, thus A/B testing is widely expanding far and beyond traditional steps. However there are some general recommendations of what is usually tested.
 

Product descriptions

Very important part of selling a product since it outlines valuable information for the customer with a reassurance of making a smart purchase decision.
 
zappos-product-description
 

Sign up page

As a rule people don’t like to spend time on registering and want this process to be as quick and painless as possible so make sure that you find the best way in signing them up as conveniently as it can be.
 
mail-chimp-sign-up-free
 

Transactional buttons

Those are your bottom line actions that actually brings you money. Everything from a size to the colors are important and should be tested thoroughly.
 
elegant-themes-transactional-buttons
 

Product displays as a search results

How many items should be displayed on the page in response to a search query? What pattern should they be compiled in? How responsive the search was for a customer and did it help him/her to make a purchase decision?
 

These are the kind of questions that get answered during A/B testing and, as a result, bring insight not only in regards to design solutions, but marketing efforts and their offerings as well.
 
amazon-product-displays
 

Pricing layout and strategy

This could be an area of psychology but none the less enormously impacts conversion rates and should be treated with the utmost respect.
 
web-seo-analytics-pricing-tables
 

What tools are used in A/B testing

A/B testing has never been a cheap affair because it requires an audience, resources and tools to gather and process generated data. Nowadays, marketers and developers alike are equipped with quite a few different kinds of software and programming tools to define the best UX solutions for their clients.
 

Here we outline two of the most affordable tools that will enable smaller businesses to take advantage of testing.
 

1. Google Content Experiments

Google Analytics Content Experiments (Formerly known as a Google Website Optimizer) is a free Google Analytics tool allowing its owners to set simple A/B tests as well as more complex experiments involving multiple variables.
 
google-content-experiments
 

2. Optimizely

Optimizely features a browser-based WYSIWYG page editor for creating alternate versions of a given page. The only modification of your own site that’s required is to drop a single line of code before the closing tag. Once an experiment starts running, the data is available in real-time, which can be sort of fascinating to watch. This is one perk that Google Website Optimizer does not offer.
 
optimizely
 

What are the results?

The test is complete when it’s accurately telling you which variation (A or B) is the best. A/B testing is often aimed to deliver a 95% confidence level meaning that there is only a 5% chance the results are random. Obviously you are choosing a variation that would bring you the targeted conversion rate. However, as complicated as it sounds, A/B testing is a simple comparison of elements on a page. Sometimes, far more extensive testing is required in order to identify the best solution for your business.
 

Have you used A/B testing for your business website? Please tell us in comments. Your input is highly appreciated.
 
 

About the author: This article is prepared by Astound Commerce – a Demandware partner and leading provider of ecommerce solutions and services.
 

Featured image by: unbounce.
 

Leave A Reply