UX research

Magical guide about product management. Article #38

This article will help to give a systematic answer to pressing business issues: “Why and How to fix it”, “How much and How much”, “How people act in this situation”, “What people say”.

The UX research?approach should be used both before launching a new business and for analyzing and improving an existing one. He will help the first to test hypotheses and refine them in accordance with the needs of a potential audience. The second is to find out the problems of users and solve them by adjusting or adding new functions.

Stages of UX research

An ideal UX study can be divided into the following stages:

• collecting business requirements;
• market analysis;
• target audience analysis;
• generating solutions;
• the result is in the form of a list of functionality.

Research usually consists of a set of exercises corresponding to the design task. The client may know his market and competitors well, but he does not understand the target audience well, and vice versa. He may also have no knowledge at all.

Collecting business requirements

This is the starting stage that every project goes through. To begin with, the designer must find out what goals the client wants to achieve. In fact, it is to find out and evaluate how the design will help bring money to the business.

This stage consists of briefing the client and building a UX research strategy. The latter will help determine how best to achieve your business goal. Often, at the stage of collecting business requirements, the client already puts forward some hypotheses that eventually need to be confirmed or refuted.

Market Analysis

A UX audit is being done for existing products. This is a detailed analysis of errors in usability and visual part. The product is evaluated for compliance with the laws of UX, composition, color, generally accepted standards, for example, online stores. But that’s not enough. Such an analysis will not reveal the real problem and will not help to delve into it. This is a risk that the business will continue to exist, but no one will use it.

What is usually the easiest thing to do without communicating with users? Find a similar business and analyze it. For a clear analysis, we do an analysis of competitors — not only direct, but also indirect.

Target audience analysis

You can analyze the target audience through questionnaires or interviews. At this stage of UX research, it is important to find out what tasks the user needs to perform (jobs), benefits (gains) and problems (pains) of users. The list of questions for each activity helps to test hypotheses and understand the user’s online and offline experience.

For example, we ask about a person’s experience in the application, how he interacts with similar systems, his impressions, emotions, failures. And what is very important — we find out the context (that is, the circumstances and processes in the physical world). This is especially important for products that automate physical processes.

From the information received, it is possible to distinguish insights — problems or desires that are popular among the target audience.

In addition to questionnaires and interviews, the Value Proposition Canvas and Customer Journey Map are compiled:

• Value Proposition Canvas?— это закрепление jobs, pains и gains и выбор решения со стороны бизнеса для каждой job, pain and gain.
• Customer Journey Map?— это разбор пути пользователя по сервису с анализом его ощущений и мыслей на каждом этапе. Она позволяет понять, на каком этапе появляется проблема и как ее решить.

For example, we did a UX study for a startup in a fairly narrow and new area — a marketplace for the sale of marijuana in the United States (legalized in some states since 2018). There were no users of their own yet. To form hypotheses, they searched for information about the marijuana market everywhere: they interviewed “potential users” in other countries, searched for information on news sites, in Internet reviews.

It turned out that the goods are sometimes difficult to get because of the influx of buyers. This is the pain of the user. Accordingly, they came up with a pre-order function (for example, by the date of the party).

Another finding — according to the law, the buyer must present an ID card upon receipt of the goods, confirming the age — from 21 years. In the USA, where marijuana is legalized, people often come to rest who may not be aware of this requirement. Therefore, we came up with a reminder: to present an ID card upon receipt of the goods.

Generating solutions

The design team can generate solutions both independently and jointly with the client. In the latter case, workshops are often held.

Consider the generation of solutions on the design side. In fact, we compare all the previously obtained data. For example, we are looking for an answer to the pain of users in Feature Matrix. For those problems that cannot be solved by analyzing competitors, we come up with and offer a solution ourselves.

For sorting, you can use activities by the type of priority matrix (prioritization matrix). By y is the value to the user, and by x is the effort to create this function. This is how we select the features that must be added.

List of functions

Based on the results of the UX study, we compile a list of functionality. It may look like this:

Types of UX research and what is their specificity

The answer to the key question, ideally, should help to get out of the comfort zone. The question sounds like this: “Which method of user research should I use?” To answer this question, let’s look at the graph of methods in 3 dimensions:

1. Behavioral and Attitude Studies (Behavioral — Attitudinal)
2. Qualitative and quantitative research (Qualitative — Quantitative)
3. The context in which the user is located during the research (Context of Use)

On the graph, we have distributed the 20 most popular UX research methods by these dimensions.

Each dimension allows you to differentiate methods by the questions they answer and the goals for which they are most suitable.

1. Attitude and behavior studies

The main difference between attitude research and behavior research is that in the first case we study “what people say”, and in the second — “what people do”. At the same time, very often the opinions and words of users differ greatly from their actions.

Attitude studies help to identify and analyze people’s beliefs, which is why these studies are so popular with marketers.

It is clear that the work on improving usability should be based on the results of behavioral research, but the methods of relationship research can also be useful for designers. Let ‘s look at the examples:

• Sorting cards allows you to understand the mental model of the user’s information space, as well as determine the best version of the information architecture for a product, application or website;
• Surveys help to identify and group user perception patterns. In surveys, people give their own answers. Such surveys can help identify weaknesses, and working out weaknesses in the product will improve usability.;
• Focus groups for various reasons are not very effective in terms of improving usability, but this research method gives a general idea of how people perceive your brand or product concept;
• Behavioral research, in turn, is designed to help determine WHAT people do with a product or service, and HOW they behave when working with it.

Let’s look at the examples:

1. During A/b testing, various changes in the design of the site are presented to random samples of users so that it can be assessed how any change will affect the perception and behavior of the user;
2. Eytracking also serves to ensure that we understand how the user visually interacts with the interface: in which direction his gaze moves and in what order he looks at objects;

Between these two poles of behavior and attitude research lie the two most popular methods that are used most often: usability testing and field research. They provide researchers with a mixed type of data at the output and allow them to shift the focus of research in the direction of attitude or behavior (but it’s worth remembering that the behavioral direction is still more preferable).

2. Qualitative research and quantitative research: what is the difference

The difference between qualitative and quantitative studies is very large.

Qualitative research generates data on the behavior or attitude of the user, which is based on direct observation of the person.

In quantitative studies, data is collected indirectly: using a questionnaire or any analytical tools.

For example, in field research and usability testing (these are qualitative studies, if you look at the graph), the researcher directly observes how a person uses technology (or does not use it at all) to meet their needs. This allows you to ask additional questions to the user during the study and even adjust the session protocol so that the process better meets the goals set. The received data is usually processed in a non-mathematical way.

Conversely, the data obtained in quantitative studies are usually processed using mathematical analysis, since data collection tools in such methods (questionnaires or logs from servers) collect a large amount of information that is easy to translate into numerical values.

Thus, qualitative research is much better suited for identifying usability problems and finding solutions to them, and quantitative research is suitable for obtaining statistical information. The availability of statistical information collected by quantitative methods helps to prioritize the solution of project tasks and focus on the main thing.

3. Product usage context

This measurement determines the conditions under which participants get to know and use the product or service. Types of context:

1. Natural conditions (uses the product in a familiar environment)
2. Uses the product according to a given scenario
3. Does not use the product at all during the study
4. Mixed type: a hybrid of all of the above

The main goal of user research in natural conditions is to minimize external interference in order to understand the behavior and attitude of the user to the product as close to reality as possible.

Such a context gives high reliability of the results, but less control over what exactly you want to study. Usually, ethnographic field studies are carried out in natural conditions to obtain the most reliable results, but still there may always be some biases that will lead to inaccuracies.

User research with a script is great for focusing on a specific aspect of the product, for example, a redesign of the interface or a separate feature. The degree of detail of the scenario is always different and depends on the objectives of the study. For example, scenarios in benchmarking studies are usually prescribed very strictly and are of a qualitative nature, so we will have reliable usability indicators at the output.

Studies that do not involve the use of products are conducted to study more global issues than usability. For example, it can be brand research or even models of cultural behavior.

Hybrid methods suggest non-standard approaches in the study:

Collaborative design allows users not only to interact with the interface, but also to rearrange its elements themselves. At the same time, in the process of research, the user’s decisions are immediately discussed: why did he change the interface in this way, and how will it help him?

The methods of conceptual testing of a product help to understand whether this product/service is needed in principle? Will consumers use it??

Most of the methods on the graph can move along the measurement axis, for example, from qualitative to quantitative and vice versa. This can even happen within the framework of a single study in order to achieve several goals at once. For example:

1. Collaborative design allows users not only to interact with the interface, but also to rearrange its elements themselves. At the same time, in the process of research, the user’s decisions are immediately discussed: why did he change the interface in this way, and how will it help him?
2. The methods of conceptual testing of a product help to understand whether this product/service is needed in principle? Will consumers use it?

Most of the methods on the graph can move along the measurement axis, for example, from qualitative to quantitative and vice versa. This can even happen within the framework of a single study in order to achieve several goals at once. For example:

1. Field research can both focus on what people SAY (e.g. ethnographic interviews) and what they DO (extended observations);
2. Expediency testing and card sorting can be both qualitative and quantitative methods;
3. Tracking can be carried out according to the scenario and without it;

At what stage of product development, which method is best suited

Another very important point to consider when choosing user research methods is the development stage at which your product is located and the goals that are being pursued at this stage.

Stage 1: Creating a product strategy

At the start of product development, ideas and opportunities for the future are usually determined. Where are we going? What are we creating and for whom? The set of research methods in this case is very wide, but in no case should we refuse to conduct user research at this stage.

Stage 2: Implementation

At this stage, the process of creation and experimentation is in full swing, and user research helps to improve usability and avoid the risk of failure.

Stage 3: Evaluation

Users are working hard with the product, so it’s time to evaluate how successful it is. If it is a completely new product, then you can compare it with competitors, and if you have launched a new version of an existing product or service, then you can compare it with the previous version.

The 20 most popular methods of UX/UI research and how they are conducted

Here is a brief description of the methods shown in the graph above.

1. Laboratory studies (Usability-Lab Studies). Each respondent is one-on-one with a researcher in a special UX laboratory and performs a set of tasks following a given scenario;
2. Ethnographic Field Studies. Researchers observe users in a familiar environment for them, where they would most likely encounter the use of the tested product/service;
3. Joint design (Participatory Design). Participants are given the opportunity to design their own interface from the proposed elements in such a way that it is as convenient as possible for everyone. In this study, the most important thing is to get answers to the questions: what is the most important thing in the interface for each user and why;
4. Focus Groups. Groups of 3–12 participants who discuss a number of topics and/or perform specified exercises, then giving feedback;
5. Interviews. The researcher meets one-on-one with the respondent and they discuss in detail a particular question on the topic under study;
6. Eyetracking. The eytracking device tracks the user’s gaze to accurately measure where the user is looking when performing tasks or simply interacting with a site or product;
7. Usability Benchmarking. Usability research, which is performed by several participants according to a strictly prescribed scenario with clearly defined performance metrics in advance.
8. Moderated Remote Usability Studies. It is carried out remotely using programs for screen sharing and remote access.
9. Unmoderated Remote Panel Studies. A group of trained participants with video recording and data collection programs installed on their devices say out loud what they think when working with a website or application. The recorded data is immediately sent to the researchers on the company’s server for analysis.
10. Concept Testing. One or more test subjects are provided with a conceptual version of the product to determine whether the product meets the needs of the target audience or not.
11. Diary studies (Diary/Camera Studies). Participants are given cameras or diaries to record all aspects of their lives that in any way relate to the product or service under study, or are simply key to the target audience. Such studies are usually long-term and are suitable only for those data that users can easily record themselves.
12. Reviews (Customer Feedback). Closed or open information that users independently leave by clicking on the survey link, filling out a form or sending an email.
13. Evaluation of preferences (Desirability Studies). Participants are offered various visual design options, each of which they evaluate according to the proposed list of attributes. Such studies can be both qualitative and quantitative.
14. Card Sorting. Users sort items into groups, then assign a category to each group. This method helps to create or refine the information architecture of the site by understanding the mental model of the user. The method can be both qualitative and quantitative.
15. Clickstream Analysis. Analysis of the user activity map. This method requires that the site or application be equipped with special tools for data collection.
16. A/B Testing (A/B Testing). Random groups of users are offered one of several interface design options to interact with it. Then it evaluates how each design option has affected user behavior.
17. Unmoderated UX Studies. A quantitative/qualitative automated method that uses special tools to track and record the behavior of participants. Usually, programs are used for this, which are installed on the computers / smartphones / browsers of the subjects, and the users themselves are given scripts or tasks that need to be performed using the prototype or site under study.
18. The study of true intention (True-Intent Studies). This method polls random visitors of the site, what their goals and intentions are when they enter the site, measures their subsequent behavior, and when they exit the site asks if they managed to achieve their goal or not.
19. Intercept Surveys. A survey that is launched while using the site or application.
20. Email Surveys. Surveys, the participants of which are selected by email messages.

Conclusion

This is the end of the UX study. It is followed by the design stage. And most importantly, UX research can be considered successful only if you managed to find the user’s pain and come up with a solution.

Thank you for your attention ??