In this post you will find a summary of different UX Research methods that can be used from the three different approaches through which we can meet a user research process, depending on what we want to observe in the study we are going to carry out. We are going to do so by basing our article on the Christian Roher’s one about research methods, which is almost already a classic.
Research points of view
As we mentioned earlier, there are three ways we can approach research, referring to user research regarding user experience, which are the following:
- Attitudinal versus conductual: On the first we will assess users’ attitudes, while on the second one we will assess their behaviour and conduct when they are taking a test.
- Qualitative versus quantitative: On the first one the results obtained will be subjective, and on the second these will be measurable and comparable.
- Context of use: On many occasions carrying out research with both previous points of view is not enough and we will need to test the digital product in the context of use where the company or the potential users will actually handle it so we can see possible changes or improvements we can make in the design purpose based on context.
Parting from these three points of view Norman Nielsen made this classification graphic of research methods:
Each symbol differentiates between contexts of use on each test.
The graphic divides different methods between four squares (behavioural, attitudinal, qualitative and quantitative). The further up, more conductual the method is and the further down, the more attitudinal. The same thing happens on the right-left axis for qualitative (the left side) and quantitative (right side).
If we want to arrange a good user research process for our digital product the recommendation from our Studio is to carry out a variety of them. We should not risk all of our research just in interviews, since this may be a nice trend anticipator and about what our users may want in the future regarding a digital product in our segment, but, for example, they are not useful to identify usability issues in a digital product that could negatively impact in the conversion rate, which a usability test in a lab or an A/B test will help us to appropriately identify.
Now, we will list the summary of 20 research methods which appear in Norman Nielsen’s graphic and article:
- Usability-Lab Studies: these are studies which are carried out on a rather controlled environment in a UX laboratory, where the moderator guides users on a one on one session through a series of different scenarios so they do different tasks as similar as possible to situations that want to be tested in order to detect usability issues on a given product.
- Ethnographic Field Studies: Researchers meet the study participant in their environment (either their workplace or home), where they use the product. This kind of study helps to have into account keys about user types and the environments in which our users move around.
- Participatory Design: Participants are given some basic design elements or creative materials and they build up their ideal experience with the given digital product, indicating which things are most important and why.
- Focus groups: Groups between 3 and 12 participants that are encouraged to talk about different topics related to the digital product and also to share their opinions and points of view in a vocal and strict way through different exercises.
- Interviews: A researcher debates one on one with the participant about topics that are related to the digital product which is object to the study. Read our Guide for user interviews. 2022 Edition.
- Eye Tracking: An eye tracking device correctly calibrated can observe where the participant is looking while they are executing the tasks that they were assigned or how they normally interact with a digital product.
- Usability Benchmarking: These are usability studies with several participants with different profiles where parameters are strictly scripted so they develop correctly and the digital product’s performance test can be as accurate as possible, and we can see where improvements can be made.
- Moderated Remote Usability Studies: Same as a usability study just moderated remotely, by using different tools such as screen sharing or remote control.
- Unmoderated Remote Panel Studies: A user panel is recorded with a camera and a data collection software while they use a digital product and comment aloud, which can be analysed at the moment or after.
- Concept Testing: A researcher shares a concept idea that they have that has a value proposition for users and clients and we try to determine if the idea has a potential market or not. This study can be carried out with a few or many participants, online or offline.
- Diary/Camera Studies: These are similar to a User Journey, where participants record with a camera or take down notes in a diary the impressions that they have while continuously using the digital product, and also the feelings they have in the meantime. These are usually longitudinal and have to be easily recordable data by the users. With the recorded information in the chosen medium an expert can arrange a user journey which synthesises all the experiences.
- Customer Feedback: Closed or open information- depending on the kind of question they are asked- of a selected group of customers usually through mail, a link or social media.
- Desirability Studies: Participants receive some visual alternatives and are asked about their preferences- which one would they prefer or which one do they find more appropriate-, and they are asked to rate each one, just as a justification about the emotions that each of the alternatives transfer from a given list.
- Card Sorting: A quantitative or qualitative method that asks participants to organise a series of concepts into groups and to assign them a category to each of them. This method helps to improve Information Architecture by exposing the user’s mental models.
- Clickstream Analysis: Analysing the screen records from those screens the users navigate through and analysing where they click on and what they look at while using a digital product. This requires that the site has the appropriate instruments or that the app has data telemetry in order to do so and be able to reach conclusions.
- A/B Testing: Also known as Multivariant Testing if it is carried out with more than one alternative, this is a scientific method through which different designs are tested, by randomly assigning each of them to a group of users so that they interact with them and measure the effectiveness of changes in users.
- Unmoderated UX Studies: This is an automatic method for qualitative and quantitative analysis which uses a specific research tool to record conducts (through an application on their computer) and attitudes (through embedded surveys), usually providing users with goals or objectives to reach in the digital product (or a prototype).
- True-Intent Studies: A method that asks randomly selected users what is their purpose for using the digital product and asks about their experience on the site when they achieve it before leaving.
- Intercept Survey: This survey is activated while the user navigates through the site so that they answer it.
- Email Survey: This survey is sent to participants that are recruited through email.
All these different methods that we just explained can be used on a user research process. Deciding which of these methods is best in order to define a digital product redesign is a joint task of the client and a good user experience designer or user researcher.
If you are thinking about learning all about UX Research, take a look at our Advanced Research Specialisation Programme (this training is in Spanish).
This is a translation of an article posted in our corporate website: