The growing use of tablets and mobiles is contributing greatly to changing the way we consume services, get information and build our professional relationships. In this context, tools such as user testing should enable us to better understand this trend and to offer services that are increasingly adapted to real-life uses and user expectations.
So here are some tips to efficiently organize a user test on a mobile or tablet.
Is it necessary to develop a complete prototype before a user test?
The design of a fully functional mobile prototype is not necessary. This design is often costly in terms of time and investment. In fact, the more complete the prototype is, the more difficult it will be to make changes later.
Only basic use cases should be made up. Complex or atypical use cases will be approached from an "exploratory" angle: if a user wishes to access a feature or content that has not been prototyped, the animator will ask the user about his expectations and what he imagines he can do in this non-developed part of the prototype.
Why not enjoy unlimited reading of UP'? Subscribe from €1.90 per week.
What is the support submitted to the user test?
We recommend testing prototypes at each stage of the project. The support tested can be :
- A "low definition" model, in the form of sketches or wireframes that can be clicked on to simulate screen sequences.
- A "high-definition" prototype, with realistic visuals and text, which can be presented on a computer screen or in an embedded version in a smartphone.
- A Beta version of the application or service before the official launch.
For each media, it is ideal to carry out one or two days of testing and to modify the tested media according to the first results. Once the corrections have been made, an additional day of testing can then be conducted to validate these modifications.
Is it better to perform the user test directly on the user's mobile phone or on a supplied device?
The user test should recreate the real user experience as much as possible. For example, when participants are asked to use a different phone model than their own, it is important that learning about the new device interferes as little as possible with the experience of the service being tested.
The ideal is to choose participants who are already using the same model of phone that will be used for the test. This avoids learning bias, and allows you to ensure in advance that the service to be tested will work well on the test phone.
Are there differences between iPhone and Android users?
To fight against disinformation and to favour analyses that decipher the news, join the circle of UP' subscribers.
Should both profiles be included in a study?
In general, and considering the results obtained in our numerous studies, the difference in task success is small when comparing the scores of iPhone or Android mobile users. However, it is important to be aware that users develop specific skills and usage strategies related to the regular use of a particular OS. This is the case, for example, for interactivity elements such as the "menu" and "back" keys on Android, the creation of folders on iOS, "drag" vs. "long press"... The same goes for habits related to the specific functioning of application download platforms (stores).
The real differences appear above all between users of mobile phones with a touch screen and those who have a phone with a "classic" screen: the time it takes to enter a text varies a lot and the manipulations are drastically different.
Therefore, in order to best determine what type of motive to include in a study, one must ask oneself the following question:
- What devices are owned by the target users of the service or application to be tested?
- Is there any interest in having the service tested on several mobile OSes?
- For an application under development, is the test media designed for a specific OS?
- For a mobile site, can the test prototype be used on a wide variety of platforms?
In cases where it seems interesting to carry out the tests with several types of mobiles, a representative sample of the OSes that current and future users will use should be included.
Are laboratory user tests more appropriate than "field" tests?
The choice between laboratory testing and field observation is made according to the objectives of each study. Indeed, "in situ" studies are an excellent way to understand the real context of use and the place that certain functionalities or applications can take in the daily life of users.
Data collection can then be carried out in different ways: interviews (at home, in the office, in a mobile situation...), telephone surveys, logbooks kept over several days or weeks, or technical devices that record the journey on the device.
However, this type of study is often inappropriate when simply assessing the ergonomics and usability of a mobile or tablet device, service or application. In this case, the laboratory is the ideal place to identify bottlenecks and to better understand how interfaces or services can be improved.
It is therefore important to be able to quickly raise the objectives of the study before defining the methodological outlines of the study. Because the test must be approached in a specific logic in order to understand the specificities related to mobiles and tablets.