I recently attended MeasureFest – an event dedicated to web analytics and CRO organised by Rough Agenda in Brighton, UK.
During his introduction of the Conversion Optimisation workshop, Joe Doveton cleverly illustrated how you can never predict which version of an A/B test will win by visiting some www.whichtestwon.com examples and asking us ‘experts’ which variation would convert the most users.
Of course, it worked beautifully, illustrating that no matter who you are or what you’ve seen, there’s no substitute for A/B testing when you want to know what works best for your website audience here and now.
While A/B testing is the most common test method for improving conversion rates, amazingly it is still only used in 38% of companies.
I was interested to find the most common A/B tests employed by the ecommerce retailers that I met at the event. These were the most common testing activities:
1. Layout of forms (horizontal v vertical)
While users are probably more familiar with vertical forms, laying them out in a horizontal bar can give them a prominent position while using less screen space.
2. Static v automatic ‘carousel’
Do users find a carousel helpful or tempting towards products and offers? Should the different themes automatically rotate or should there be navigation to switch between views?
3. Categorisation of products (taxonomy)
Testing which category product items are listed in to make them easier to find.
4. Navigation options
Varying the language, order, position and presentation style of navigational links – both for desktop and mobile views.
5. Form length (number of fields)
Adding and removing form fields and the status of whether data input is optional or required. Also testing the order of form fields and validation of the values inputted.
6. CTA position (above the fold or not)
Altering the location of ‘buy’ or ‘enquire now’ type call to action buttons on the page to measure the effect on conversion rates.
7. Image types/photography
Whether to use images or not and to decide whether a drawn style graphic or photograph performs best.
8. Button language
Testing various different words on call to action buttons to compare click through rates.
9. Product/user reviews (hide or show)
Whether the availability of a user review assists conversion or not and where best to position this information.
10. Image sizes
What is the optimum product image size? – balancing load times with order rates.
11. Hero banner with one image or a collage of images
Does a single, strong hero image convert to more sales rather than a collage of multiple products/offers?
12. Visibility of awards/accreditation
Should you display awards and membership logos? At which part of the customer journey do they have most impact on customer actions?
13. Phone number visible/hidden
To display a contact phone number or not – do phone enquiries have a positive impact on your KPIs? Where to position this information on the page?
14. Indication of payment methods accepted
It’s normally a good idea to display the kinds of payment types that you accept online, but where to display them, how big and at what stage?
15. Collapsed or visible search filters
Options for refining search results can be invaluable to shoppers but do they get spotted if the categories are collapsed down? Worth experimenting with making them all visible or specific sections open and closed by default.
16. Hide/show site navigation during the checkout stages
Does hiding your main website navigation options on landing pages focus your visitors on your intended journey to conversion or not?
17. Greyscale or colour share/payment icons
Does it make a difference to their usage when social sharing and payment icons are displayed in colour or black and white?
18. Visibility of product stock levels
Showing customers exactly how many items are currently available might spur them on to place their order there & then but does this benefit your sales figures down the line?
There’s really no such thing as a failed A/B test because a test always teaches you something. Testing is the route to making improvements, so if you are keen to make incremental improvements towards your business goals then testing should be an ongoing process that never ends.
What are you currently testing on your website? Let us know below.
A/B test Capture+ in your address forms and measure the difference it makes to your conversion rate.