How to conduct a UX test during a face-to-face interview

Conducting a UX test as part of an interview, has certain advantages. You can for example add a more exploratory phase, to understand your users and learn about their expectations, before the test phase of your interface.

It is also a good solution for testing static mock-ups or simple prototypes, since the present moderator will be able to help the user imagine how things would interact.

But what are the keys to a successful user interview session?

User recruitment, the first decisive step

It is essential to have your website, your prototype, or even your simple idea, tested by users who are unfamiliar to your project and professional sector.

Indeed, it is sometimes tempting to test your project with your colleagues (less expensive, faster), but they will necessarily have a dangerously biased vision: vocabulary, prerequisite knowledge, personal involvement, etc.

It is therefore necessary to recruit people outside your environment who fall within your target.

How to determine this target and choose the profile of your users?

Whether it's for an exploratory interview, a design test or ergonomics, your users must know the subject well. This will be the condition for them having relevant things to tell you and to teach you. They will be questioned for a while, so you might as well to talk people who have the ability to answer!

Unlike quantitative studies, it is therefore necessary to have a more precise target meeting one or more recruitment criteria. Examples:

• brand customers
• age group
• users of a product or service
• etc ...

These criteria will allow you to have similar profiles and experiences, and thus be able to consolidate your insights.

In terms of sample size, 8 users is a good start if the profiles are similar.

It is also possible to diversify the targets (age, client vs non-client, etc.) for your user interviews, but you will have to ensure that you have a sufficient number of users of each profile to be able to consolidate the results. Indeed, if you find yourself with one or two people of each profile, it will be difficult to generate reliable insights.

It fluctuates depending on the target, but in general, a base of 5 people is a minimum. This can lead to building several series of interviews if you have many separate segments.

The content of your UX tests or user interviews

To properly prepare for your interview, it is important to write a guide beforehand.

This guide should cover the framework of the interview, all the questions you ask yourself and those you want to ask the user.

We advise you to divide it into several parts, according to the main themes addressed (current habits, perception, situation, etc.) and to allocate a defined time to each of these parts.

This will allow you to frame the discussion with the user, to address all the desired subjects without being overwhelmed.

Start with an introduction that will explain to the tester what the exchange is all about and which will help put him at ease. We then advise you to build this guide in a funnel-shape and always go from the most general, to the most specific, for example from the most exploratory to the ux test of your website.

This logic must also be used in the formulation of your questions. These should always be as open as possible, even if you need to specify when you are looking for an answer on a specific element. Prefer formulations like:

• "What do you like on this page" to "Do you like this page element?"
• Or "How do you find this site, what words come to mind?" rather than “Do you find that clear? is it easy to understand? ”

You can adapt a small part of the guide to a specific profile among all of your users, but don't forget once again about the consolidation of the results. If the guide differs from person to person, it will be difficult to find converging elements.

Finally, it is recommended to optimize your interview guide after the first interview, but it must then remain the same for the other users in order to avoid biasing the results of your study.

The course of the UX test or user interview

For an individual face-to-face user interview, it is recommended not to exceed one hour. Beyond that, your user will start to lose concentration and his feedback will therefore be less relevant.

For the same reasons as those mentioned above for recruitment, it is very important for the person who conducts the UX interviews and tests to be external to the project.

Indeed, if you know the prototype or the tested website, you will, even unconsciously, give indications and bias the user feedback. Example: "Did you understand that the button on the right is used to do this?" (and yes, this is a true example!).

Ideally, there are professional facilitators, also called moderators. It is also possible to have an additional person to take notes.

Apart from these two people and the user, it is important that there are no other speakers or even spectators present in the room. Rooms with one-way glass exist to allow other attendees without distracting the participant.

The user should feel comfortable and free to speak, if they feel watched, they may either speak less or not truly say what they think. This is even more true if the people present are linked to the tested project.

The project to be tested

Whether it's testing a live website, an animated prototype made with a tool like InVision, or a more raw wireframe, it's important that your material is ready for testing. It is essential not to change it along the way, again to avoid distorting the overall results.

Warning: if you are testing an online website or app, check with your developers that there is no maintenance planned at the time of the tests.

If it's an animated prototype, even if the content of your service is not yet ready, the elements must be consistent with each other. Remember that the user needs to imagine that they are living the experience to the fullest, so you have to make it as real as possible. For example, if you are testing a mobile banking application, make sure that the amounts displayed correspond from screen to screen, otherwise it may interfere with the user's understanding.

If you are short on time, it is still better to put "XXX" instead of the numbers or some "lorem ipsum", there will at least be no misinterpretation.

Even if you only want to test the design of a prototype or mock-up, be sure to address the details. It is difficult for the user to make sense of things and ignore certain malfunctions or inconsistencies.

The case of the international user test

Testing in different countries or in a country other than yours does not change any of the above principles. Yet, we strongly advise that you work closely with a local or native moderator on building the interview guide, conducting the interviews and having a first read through the results.

During each of those steps, there will be cultural differences which your moderator can help you grasp and understand. This person will probably write the guide or help you with translations, in case of multiple localities.

For the interviews themselves, you are better off requesting translation services. Many local actors provide instant translations which allows you to follow the interview live.

If you want to know more about face-to-face UX tests or exploratory interviews we conduct at Ferpection, you can find more details on our solution pages or contact us directly.