Usability testing has been at the center of UX/UI design conversations lately, and it’s a concept that is rightfully gaining more importance. Companies are finally realizing that UX design mistakes come with a price tag. A usability issue that goes unnoticed can easily cause a large number of users to abandon a digital product or service.
That’s why usability testing is so important – it gives you a chance to test your design, validate your assumptions, and iterate based on user feedback to ensure the acceptance of your future product. It also helps you avoid spending huge amounts of money building a solution that is inherently flawed.
However, usability testing is still an emerging field and there are no universally recognized standards. UX teams in different companies will have different tools and processes in place and they’ll rely on different usability testing methods.
To help you get usability testing right, we reached out to nine UX experts from around the globe and asked them to share their number #1 usability testing tip. What they came back with is pure gold and I hope you’ll implement at least one of the following tips to improve your usability testing process.
In our book Tragic Design, my co-author and I discuss how to avoid unintentional harm to users through bad design. One tip would be to consider how the product you are designing would work with different types of people and not just your "ideal user."
Often times, when we dream up our experience, we think of the ideal scenarios. For example, when designing a social network you think about friends connecting for a party, setting up a first date, or wishing your mom a happy birthday. This is great when you're thinking up new concepts. However, once the idea is ready to start making its transition into a real product, you have to consider more experiences than those.
You have to think about the negative ones. People have a wide variety of experiences we aren't aware of. What if someone gets divorced? What if they are depressed? What if the date ends up being abusive or stalker? What if the person is out of data this month? What does your product look like in an emergency? What if someone doesn't feel comfortable having M or F on their profile? The list is a big one and that's the challenge.
You won't be able to think about all the different cases but you can cover important ones and it will improve the effectiveness of the experience you create because it will support more people and more of life's experience, more of what your users need from your product.
In short - don't forget to think about people's backgrounds as well as the negative situations that might arise.
“Do a heuristic analysis of your product ideas internally even before starting usability tests. You will catch a lot more of the low hanging fruit and validate your thinking better.”
Test early and often on a small group of users. Some designers wait until they have a high fidelity design and then do usability testing on a large group of people. This can be risky as you're investing a lot of time and effort into unvalidated design concepts which may not meet your user's needs.
In my experience, the opposite approach works a lot better. Test your designs as early as possible on a small number of people, make changes based on the findings and then test again. This basically keeps things fluid and avoids big surprises later on in the design process. Testing on a small number of users at a time will usually highlight most of your main usability issues.
Let's say you had a budget to do usability testing on 20 users. Rather than testing a single design concept on 20 users, you'd get much bigger bang for your buck by testing your concept on 5 users at a time and iterating your designs after each test. Don't be scared to test design concepts early as you can get surprisingly useful insights.
It’s really important to get the right participants for the study. While there is no need to exactly match the demographics of the user base at large, it is important that participants have the background and knowledge to conceptually understand what the website or app does. In fact, each participant should really be someone who could potentially be using the product or something similar in the future.
They should be “middle of the road” participants – not so experienced with the product or similar products that it’s impossible to get a first look type of experience while simultaneously not being so inexperienced that they don’t, for example, have a reasonable understanding of the platform on which the product is being used. Careful consideration should go into the participant screener so that whomever is doing the recruit will understand exactly who is needed for project success.
Generic measures such as error rate, time on task, and user satisfaction can only tell you so much. In order to measure the true return on investment of UX, make sure the usability metrics you choose are vital to the success of your specific product or service.
Remember that you are not testing the user, the user is testing the design.
That means the user cannot fail, you should never take any result personally (especially good results), and you should not be concerned with "confirming" that you were right. It's not about you.
Look for patterns among the users. Don't focus on the individual mistakes or problems of one user. Each user is only a piece of the puzzle. To see the whole picture, you need many users.
I have one tip for you and it includes two things - perform usability testing yourself and take notice. If someone else is doing it instead of you, they won't notice the devil in the details that can make all the difference.
Usability testing is not about gathering opinions from people, it is about studying their behavior. Instead of asking people what they think about our design, we should give them tasks to accomplish, be (mostly) silent, observe what is going on, and take notes. Our goal is to discover problems with our designs so that we can correct them, possibly early in the process.
Usability testing should be an integral part of every digital product during its entire lifetime. More often than not, usability testing is performed in the early days of a product lifecycle while, over time, changes accumulate to render the initial testing results obsolete. Things change, habits and expectations of people change, so make sure you are always one step ahead by periodically conducting usability testing.