Why users fail. The most frequent usability issues we observed when testing with users.


At TeaCup Lab, we are constantly running usability tests on all kind of devices and interfaces for companies of all sizes and sectors.
Throughout my career as a UX consultant I’ve been involved — as a researcher or project manager — in more than 200 user research studies with thousands of participants.

I had the privilege to take a close look into the problems users had when dealing with an interface and I’ve been able to learn what works and what doesn’t work for them.

This ongoing list of the most frequent issues is not intended to be exhaustive nor based on a thorough collection or classification method.

Jakob Nielsen’s tremendous longitudinal study is a far better source of information for quantifying the frequency and impact of usability issues.

This article just highlights some of the difficulties we have repeatedly seen over and over in most usability tests, regardless of the business sector, device used or participants’ skills.


Lack of information

Probably, the most common cause of failure I’ve observed is the lack of relevant information to support the task the participant is trying to accomplish.

Designers often make wrong assumptions regarding how much information is needed in an interface, and apply the “less is more” rule too freely.

Users always need clear and relevant information when they are dealing with complex features, when they are trying a new product or service or when they could likely suffer negative consequences from a wrong choice.

Making a purchase, booking a flight or managing their bank account are three common situations where the lack of information can make participants abandon their task or indicate they would rather make a phone call to the support center.


Your website does not provide enough confidence

When testing e-commerce websites, participants often abandon a purchase because they don’t trust the website or are concerned about the security of their data.

Many would say that this should not occur in a controlled environment such as a usability lab, where participants are generally using fake data to complete a purchase or a registration.

However, it happens. And it happens more often than you can expect.

Incoherent or confusing information, technical errors, the lack of a contact option or simply a shabby look are the most frequent factors that made users mistrust a website and abandon their task.


There are some pieces of information users still don’t want to provide (without any good reason)

Users’ concerns about providing their personal information in a website are not only a matter of confidence.

Phone numbers, national ID cards, license plates, or even email addresses… they should be requested only when it is strictly necessary for accomplishing the task.

This is a widely known usability best practice that, however, gets ignored sometimes.
Users still abandon a registration or a purchase just because they consider that the request for some personal data is not justified.

Most of the times they are true.

Asking for their phone number or home address when they are only trying to get a quote for their new car insurance is not acceptable for many of them.

Corporations do not need their phone number to calculate a price. What they are “saying” with their phone request is that they are going to call them if they leave without purchasing an insurance.

On the other hand, asking for the same information when you’re filling in your delivery address is fine. However, you could still abandon the task if I don’t explain you why I need your phone number (for example, to send you updates about the status and delivery of your order)


Confusing or incoherent information

When the same sensitive information is presented in different ways during the journey, test participants, as real life users, tend to become suspicious and doubt about whether they can trust the website.

A couple of years ago I tested a large website specialized in hotel room reservations.

Unlike other travel websites I’ve been testing, my client was doing a good job in explaining the cancellation policy clearly and in every step of the booking process.

Most participants were able to find an hotel and to select a room without major errors. However, some of those who paid attention to the cancellation terms in the hotel detail page noticed that the same information had been written in a slightly different way during the payment step.

The differences were insignificant and the conditions did not change at all, but that was enough to make participants feel insecure. They believed the website was not reliable and they felt they were no longer in control of the process.

The result was that some of them decided to end their task without booking.


Hard to read content

Small texts, sometimes placed over complex images or with insufficient contrast compared to their background, often proved to be hard to read for a number of participants of any age.

Some of them won’t explicitly complain about the poor readability of some sentences. Nevertheless, the issue is easily detectable by observing your participants leaning towards the screen in the attempt of reading what is written on the page.

I’ve been spotting this issue on both PCs and mobile devices, even if it is more common on small screens.

In a few cases, while I was recording the interaction of the participants on a mobile phone with a document camera, the content was so difficult to read that they even placed their heads between the camera and the device.

When the content that is hard to read is trivial, this issue has no or little impact on the users’ experience. On the contrary, hiding some relevant content may often result in confused participants that are not able to end the task correctly or decide to abandon it.

Usability tests are only one of the many user research techniques we master in TeaCup Lab.
Tell us about your project and discover how we can help you.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.