Usability Testing – A Beginner’s Guide

If you’re thinking about becoming a UX designer it can be hard to know where to start. In this post we’re going to help you get to grips with usability testing, a crucial component to the UX designer’s day job. We’ll outline the basics and give you actionable steps so you can perform your first usability test at home or at work. If you’re the sort of person who likes to know the ‘why’ as much as the ‘how’, then you’ll love studying human behaviour, testing products and validating your hypotheses through usability testing.  Let’s get started!

What is usability testing?

The UK government defines usability testing as: ‘Evaluating a product or service by testing it with representative users’. The main goals of usability testing are to double-check overall user experience of a product and to collect empirical evidence that the product does actually do what it is intended to. It is very easy to rely on your own opinions or that of your team when you are working closely on a product. This is because when you eat, sleep and breathe that product, becoming an ‘expert’ on it in the process you can become blinded to its faults. Even if your product is perfect usability testing enables you to validate that.

Why conduct a usability test?

The number of advantages to conducting a usability test during the UX design process are almost infinite, but perhaps the most important reason for conducting a usability test during the design process is that it’s considerably cheaper and quicker than fixing a site that has already been built. Even if the majority of the usability feedback is positive, users will still be able to help you identify some small tweaks to improve usability.

When should I perform a usability test?

Usability testing can start as soon as you have agreed on the Minimum Viable Product and made some initial rough sketches, and, as recommend, testing ‘early’ and ‘often’ is best practice.  It could be as basic as showing some hand-drawn sketches to a group of colleagues or friends. This is known as guerrilla testing as it is ‘quick and dirty’. Usability testing should happen multiple times throughout the testing process. As well as the guerrilla testing, it is best to add stages of usability testing to your project plan so it isn’t forgotten and your team have something to work towards. Testing could be performed at each of the following stages:
  • Initial wireframes
  • More detailed wireframes (I like to link mine together using the free prototyping tool, Invision)
  • A working prototype built by a front-end developer
  • The actual product before launch

How do I start?

Come up with a plan to ensure you get maximum benefits from the testing. Write a brief test script with around five key questions you would like to focus on. Examples of questions could be: what would make you want to use the product more? And, can you imagine yourself using the product regularly? Here are some examples of different testing resources and methods: Usability lab A usability lab, although an expensive option for a startup, is becoming popular amongst bigger companies. Craig Spencer introduced the idea to companies such as the Guardian and Tescos and Kate Towsey from the Government Digital Service are aiming to set up labs across the UK so government organisations can knowledge share. If hiring a lab isn’t an option for you simply set aside a quiet area of your office. Use online tools Tools such as Validately or allow you to watch and listen to every move your participants make. This allows users to perform tests at home, rather than coming into a foreign environment. Because of this you can often achieve more accurate test results as users feel more at ease in their natural surroundings and therefore behave in a way that they usually would. Starbucks testing “Starbucks testing” is a term used by Justin Barr Young that describes the ultimate in guerrilla testing whereby you simply grab five people from your office, offer them coffee and ask for some quick feedback! The negative side of this kind of testing is that these users are probably not your target users, and they likely already have previous knowledge of your product. This can lead to biased test results. Embeddable prototypes Embeddable prototypes is a new trend for sharing prototypes on social media which allows you to bring your work to life, communicate ideas and get user feedback, e.g. Invision allow their users to embed gifs into 100s of places online

So who should I ask to participate?

Jakob Nielson states only five users are needed to conduct a usability test, but as the saying goes, the more the merrier. Publicise your research intentions on social media, your blog, to friends and family and try and get as many people as possible to participate. You’ll probably find that people are really flattered to be asked their opinion and will quickly agree to take part. It is really important that the participants match your personas as much as possible in terms of characteristics, their reasons for using the product and which devices they use. For example, it is much more efficient to ask people to test a new mobile site on their mobile than on an old desktop computer. If you work for a company with employees who are similar to your target market it is worth asking them for feedback. And even if they aren’t completely similar, speaking to frontline employees like sales staff or customer service staff is particularly useful as they have contact with users on a regular basis. Beware of biases and try to be as holistic as possible. e.g. taking feedback such as ‘Kelly in accounts always says that x customer thinks y’ lightly as it can be taken be taken out of context and exaggerated. Don’t forget to get feedback from the participants about your testing methods so you can improve them for next time.

What should I do with the results of my test?

Firstly make sure the data is stored somewhere safe. It is essential to keep well-labelled digital copies of all usability tests so they can be used to compare and contrast future results. It is also useful to tag the data so you can use across multiple products. Next hold a debrief meeting with as many different stakeholders as you can to discuss the feedback. It can be particularly helpful to allow them to watch or listen to any recordings so they can make their own minds up rather than having an opinion inferred by you. It can be particularly helpful to invite developers to this meeting so they can get excited about the product they have built and also understand feedback from a user perspective. Finally set actions based on the outcomes of the meeting, task someone with making sure they are achieved and then iterate. The product is likely to go through several iterations until it has reached desired usability so be prepared to hold multiple meetings over the product lifecycle.


If you only remember three things from this post, make them these three:
  1.  Test early and often
  1. Ensure you test with people who are in your target market. You need to know who they are, why they are using the product and what device they will be using
  1. Be prepared to iterate, iterate, iterate!
Caroline White is a graduate of the CareerFoundry UX Design Course. She now works as a UX designer for New Zealand-based digital agency, NV Interactive.