Usability Testing for Websites — A How To Guide

Usability Tests vs Qualitative Research

Usability testing your website is a must for improving User Experience (UX) and User Interface (UI). Most businesses understand that their position in Google depends on Search Engine Optimisation (SEO), and that great SEO requires a great UX. There are a lot of helpful analytical tools like Inspectlet, Hotjar, Google Analytics and Google Optimize which gather a lot of data about how people use your website (e.g. scrolling behaviour), however, they fall short of providing the deep insights that qualitative usability research can provide. In other words, these analytical tools are great at highlighting problems with your website, but they can't tell you why parts of your website are problematic or how to improve them. If you truly want a better website, then you should reap the synergistic benefits of using both analytical tools and qualitative usability research. It's absolutely worth the investment. So, without further ado, here is Adhesion's comprehensive guide for conducting website usability research:

How To Conduct UX Research (on a budget)

Conducting UX research can be affordable, even if you are a small businesses (you can't afford not to). We're going to try provide all the resources you need throughout this article, so the time and money you spend will largely depend on how much research you invest in. This research will mostly consist of setting a scenario for a group of people, and instructing them to complete a small list of pre-defined tasks on your website while you track, observe and question them. A Neilson Norman Group user testing article from 2000 suggests that, for example, instead of conducting 1 round of research on 15 people, it is most cost-effective to conduct 3 rounds of research on different groups of 5 people. After each round of research, you can make improvements to your website before conducting the next round of research. However, Neilson's article is out-dated and makes a number of significant assumptions. We have conducted research of our own and found that, instead of conducting 3 rounds of research on groups of 5 people, it is even more effective to conduct 5 rounds of research on unique groups of 3 people. Iterative design is key.

You are going to need to reward people for their time and concentration, so you should try not exceed 30 to 45 minutes per test. Each test must include an introductory brief (5 to 10 mins), 3 to 4 predefined tasks (15 to 20 mins) and a debrief session or focus group (10 to 15 mins).

UX Test Scenario, Tasks and Questions (examples)

When choosing the tasks to use in your tests, it's important to remember that Google prioritises mobile usability over desktop usability. Before giving people a list of tasks to complete, it is important to provide them with a scenario so they are in the right contextual mindset for your tasks. Here's an example of a scenario Adhesion would use for website usability research:

"You have recently invested in a new website for your sunglasses business and now you want to focus on online sales. You have decided you want to invest in a professional agency to handle your online marketing. Adhesion is an Auckland-based agency that provides small businesses with digital marketing services. Adhesion comprehensive range of online marketing services includes online advertising, websites, email and more."

As you'll see, it is also important to provide context in your tasks. Here's some tasks (in order of increasing difficulty) that you can apply to your business:

1. "Using your preferred internet browser, go to Adhesion's homepage and take 15 seconds to gain a basic understanding of Adhesion's services. After 15 seconds of looking around the homepage, explain what you have learned about Adhesion's business and services." (5 mins)
2. "You are now looking to advertise your business online. Use Adhesion's website to work out which type of online advertising is best for your sunglasses business. Choose an online advertising service and explain why you came to that decision." (5 mins)
3. "You now want more information about the online advertising service you chose. Download the relevant brochure from Adhesion's website, open it and have a read through it." (5 mins)
4. "You have decided to invest in one of Adhesion's online advertising services. By any means necessary, contact Adhesion." (5 mins)

Usability Testing Best Practices

• The introductory brief is necessary to reduce bias. Set a relaxed environment so each user can act normal and speak openly. Explain to everyone that you are testing your website, not their ability to use websites. If somebody apologises during a task, this is a big red flag indicating a problem with your website. During and after each task, ask open-ended questions to help each person express their thoughts (e.g. ask the 'Five Whys').
• Use screen sharing and/or screen recording software like QuickTime Player (make sure audio recording is enabled), and have someone else analyse each video file for more insights. Each subject is costing you time and money, so it's also worth using cameras to record the facial expressions, eye movement and hands of each user. Ensure to make any recording equipment subtle to prevent people feeling uncomfortable.
• Once you have made a round of improvements to your website, feel free to contact your previous subjects to gather a quick opinion. However, never test the same person more than once.
• You need to pre-empt extrinsic factors that will affect people's behaviour (e.g. time of day, your mood, your choice of words, lighting, internet speed etc). Try to keep everything constant during all your tests. The only thing that should change is the people you are testing.
• Inviting all 3 people in each group to be tested at the same time means you can host valuable focus groups after they have completed the tasks. However, it is very important to keep each person in seperate yet similar rooms to prevent them giving socially-desirable, non-candid responses (out of sight, out of mind).

Metrics to Measure for your Website Research

Again, each person you test is costing time and money, so get the most out of it by measuring everything practical. Here's a list of things you should record and analyse:

• Pages visited and time spend on each page
• Use of website menu and time spent/before navigating
• Time spent reading and unread sections
• Buttons and links clicked
• Eye movement tracking
• Conversion funnel drop-offs
• Cursor tracking
• Scroll tracking
• Verbal and non-verbal feedback (e.g. facial expressions)
• Post-task focus group insights

How To Choose People for UX Research

Here is Adhesion's Website Usability Screening Survey that can be used to 'filter out' people who are not appropriate to use for website testing (e.g. web developers). Our survey also gathers a lot of relevant information that's useful when analysing each person's results and responses.

Get Professional Help with Usability Testing

There is so much more detail involved with conducting website usability research than what is covered in this blog article. If you want more information and help, simply contact us!