What is usability testing?
Usability testing is a method used to test a system or design in a controlled environment. In web design, this primarily offers us the chance to monitor the strengths and weaknesses of our applications and websites in real time, with real users.
Often when a project length spans weeks or months, the entire team may all too quickly become accustomed to the system, its structure, and its processes. This can sometimes cause tunnel vision when it comes to usability testing the product. Whilst internal testing can be carried out on the basic physical principles of a website (cross-browser compatibility, design elements, coding standards etc), it is often difficult for a project team to recreate how users behave around certain processes or towards certain designs.
At Creode, we therefore user test to find areas of the site/system that can be improved and streamlined for the user, with an aim to increasing usability and in the area of e-commerce; goal conversion.
What do we use it for?
Our primary requirement for usability testing comes from our e-commerce builds. When it comes to online stores, companies can often spend many £1000’s on marketing campaigns, which can all be put to waste if the website in question is not good enough to convert those marketing leads.
Conversion is key for e-commerce sites and usability testing is essential in understanding how customers are using a store, and the most efficient methods of reaching high-percentage conversion rates.
Goals for usability testing
There are several criteria we look out for when testing:
Nowadays, users will make almost instant decisions on where to click on a page, or even whether to leave a particular site. Users expect to understand and navigate a site with ease and speed; make them work too hard or hang around too long and they’ll disappear. It’s therefore imperative that you can offer a simple and fast route through key processes, should the user wish to do so. This becomes far more important when you are dealing with personalised products, like with our recent work on http://www.yourimage2canvas.co.uk.
Everyone makes mistakes; whether it is clicking the wrong link or a simple typo in their personal details. However, it’s important to establish those possible problematic areas and identify how you’re going to deal with them. Simple usability tests can highlight all sorts of scenarios your internal testing didn’t get to, based on ways different users manipulate a site.
When you meet a client there will no doubt be a plethora of new acronyms and terminology you need to take on board. As a digital agency, our first few meetings are often about trying to understand the actual business at hand rather than the intrinsic details of the project. When you’re wrapped up in your own business, it can be easy to forget that your end user may not always have the same level of understand about your product as you do. Making sure your test-subject understands terminology is vital, so quiz them at the end.
It’s always important that a user remembers your brand and the unique selling points your website offers any potential competitors. There are several methods to identify whether a user has retained this information, however in most cases a simple post-test quiz asking questions such as “What was the website called?” and “What were the main offers?” can help clarify whether the site brand and main calls-to-action are instilled correctly.
We use a great bit of software called Silverback as the first port of call for our usability testing. It allows us to sit test subjects in front of a computer and Silverback will film the users’ screen, upper body and face. This allows us to retrospectively assert if they were clicking in the right place, and also gage any frustrations through facial expression.
You only get one chance!
First impressions apply to websites too! Offer a user a poor experience and you’ll lose them as a customer forever. If you get it right before you deploy, then you can look at techniques such as A/B testing to further examine user behavior.