Despatch Bay Hub Hub Visit main site
Ecommerce A/B Testing: Increase Your Conversions

Ecommerce A/B Testing: Increase Your Conversions

By James Barnes, July 30th 2019

Familiar with the term A/B testing but not the benefits it can bring your business or the method behind it? Read on! When you understand the benefits, we'll give you some examples of the A/B experiments you can run to optimise your ecommerce site.

What Is A/B Testing? 

A/B testing gives you the opportunity to continually optimise your website and marketing activities, using data collected from your prospects and customers. There are many features you can optimise including, functionality, design and content. 

How? 

Two slight variations of a feature or text on your website are automatically shown to a number of people when they visit; with one group getting version A, and the second getting version B. Their behaviour is analysed to see if the changes have had a positive impact on the customer journey and for the business. The winning variation can then implemented to optimise the performance. 

Example: Which call to action (CTA) button will receive more clicks on this product page?
Example: Which call to action (CTA) button will receive more clicks on this product page?

A/B testing isn't a one-off process; there will be many variations that will continue to increase from previous tests. Ideally, you should test the various improvements until the best possible version is found for your site or marketing activity. 

Why Is A/B Testing Important? 

Because, you will always be improving your prospects and customers experiences. The implementations you make will be based on data – you have proof they work. 

 Depending on the experiments you undertake, you will positively impact on the success of your business too. Keep an eye on those all-important metrics like, cart abandonment rate and customer acquisition cost! 

But don't just take our word for it, here's what Jeff Bezos, CEO of Amazon told Fortune magazine.

Experiments are key to innovation because they rarely turn out as you expect and you learn so much […] If you can increase the number of experiments you try from a hundred to a thousand, you dramatically increase the number of innovations you produce.

Where Do I Start? 

Choose the Right Tool 

Choosing the tool to complete your A/B tests is affected by various things including: 

  • Your employees and their skills 
  • The A/B test and its objectives 
  • Time availability 
  • Cost of the tool 

If you have an established team, between them, they may have the skill-set to perform A/B tests. As a brief overview, they will need analytical and marketing knowledge, and you'll also require developers and designers. 

However, you need to decide if they have the time to perform the A/B tests efficiently. A/B experiments need designing, coding and testing before they even go live and, don't forget, it's an ongoing task! Can you afford to give A/B testing your team resource?  

If you can't perform A/B tests in-house, there's no need to worry, there are many A/B testing tools on the market. If you're testing email subject lines, most email service providers already provide that service, and if you're testing everything but, this article will help.

If you choose to use an A/B testing tool, the steps may vary slightly, however, the process will remain similar. 

Step 1 – Prioritise Pages for Improvements 

Decide which pages you're going to optimise first; these should be the pages people visit quite a lot, like a landing page or a major part in your customer's journey, and also, contain metrics you can measure as a business. 

You can review your webpages' analytics and also, your behaviour flows in analytic tools like Google Analytics. By doing so, you will begin to understand your websites' and individual page's performances as well as, common behaviours on your site.

Once completed, you will be armed with knowledge for you to prioritise which page, or perhaps pages, you're going to optimise first. 

Tip

Heatmap tools will also help you understand how visitors interact with your website. 

Step 2 – Get Your Thinking Cap On

Now it's time to think of a hypothesis to try and improve your objective, improving peoples' experience on your website.

Step 3 – Design and Test 

This is the stage when your experiment starts to come together! Your hypothesis becomes a design. Remember to test it, no matter how small the chance, to make sure that it's fully functional and working. 

Tip

Set yourself a SMART goal every time you A/B test, and if you're new to the A/B testing game, focus on a single metric; you can always test another metric later on down the line.

Step 4 – Go Live!

Launch your experiment! Keep your A/B test live long enough for you to have the data for you to review and analyse. Unfortunately, we can't advise a time period on this – it's different for everybody as everybody has a different amount of website hits. 

Don't end your experiment if you've had few visitors. You won't be optimising your website if only a few people have seen your variations.

Step 5 – Number Crunchin' 

Once your experiment has concluded, it's time to number crunch. You will need to review and analyse the performance of each variable using the same data you gathered in step 1, referring back to your SMART objective. If there was a stand out winning variation, implement the improvement! 

Step 6 – Begin Again 

As we've said before, you should continually optimise your website for the sake of your visitors and your business. You will have many ideas, and some that come as a result of previous experiments too. 

A/B Testing Ideas 

The sky's the limit when it comes to what you can A/B test. If you're new to this, it can be a little overwhelming, so we've included a few examples below. You'll be expanding on them in no time.  

Call to Action (CTA) Buttons 

As you familiarise yourself with A/B testing, we bet you'll find yourself performing more and more tests on CTA buttons, after all, it's the button that creates your conversions!   

Which CTA button colour will generate your more orders?
Which CTA button colour will generate your more orders?

HubSpot concluded in their A/B testing that red was better; but will this be the case with you? Perhaps red doesn't even grab your visitors attention, and then there's the actual content of the button itself. That's when it can get interesting! 

Pricing Strategy 

Pricing strategy is a biggie for all businesses; if you look around, you'll notice there are many variations. 

Are you going to test various pricing strategies on your site?
Are you going to test various pricing strategies on your site?

Some businesses include pound signs and some don't, do you think there's a reason for that? People seem to have an opinion on .00's too, which would you rather read £10.00 or £10? Now the important question, which one gives you more conversions? 

Navigation 

A/B testing your navigation could be great, it's always at the top of your website so will always get lots of attention! 

Can you see how much attention 'sale' grabs now?
Can you see how much attention 'sale' grabs now?

You could highlight a key point of interest in your navigation bar; this could range from seasonal events like summer to sales, or move items around your navigation – will it achieve more sales? 

There you have it, you're now clued up on A/B testing – you should now have a grasp of what it is, how to do it and you've got some ideas to start it off. What will be your first test? 

Sources

  1. https://fortune.com/2015/09/17/amazon-founder-ceo-jeff-bezos-skills/
  2. https://blog.hubspot.com/blog/tabid/6307/bid/20566/the-button-color-a-b-test-red-beats-green.aspx