Techniques for Requirements Analysis with Users

The first step in designing an interactive system is a careful analysis of its requirements. This is not a trivial phase. It takes real effort to understand the problem before turning proposed solutions into software.

Start by identifying the people who will actually use the system. Research should concentrate on them, not only on the person who commissioned the project. The customer and the user may be different people with different needs.

Several research techniques can involve users directly. Each reveals a different kind of evidence.

Interviews

An interview uses questions to explore a user's needs, work, and problems. It is particularly useful early in a project, when the team is still learning the domain.

An observer can help by taking notes and suggesting follow-up questions. The conversation may expose unexpected and valuable topics, but the interviewer must also be able to guide it back when it drifts too far from the research goal.

Interviews demand time both during the session and afterwards. The data is qualitative: recordings or notes must be reviewed, interpreted, and coded before patterns become visible. A quiet, comfortable setting helps participants speak openly.

There are three common formats:

• Structured interviews use a fixed sequence of questions. They fit a clear research goal, such as comparing two competing systems or identifying strengths and weaknesses before a redesign.
• Unstructured or flexible interviews follow a topic rather than a fixed questionnaire. They are useful for exploring an unfamiliar domain and discovering needs the team did not know to ask about.
• Semi-structured interviews prepare questions in advance but allow their order to change. Answers can lead to new questions or deeper investigation of an unexpected issue.

Focus groups

A focus group is a moderated interview involving several people. Participants respond not only to the moderator but also to one another, creating discussion and exposing different perspectives in a short time.

The moderator must prevent a few confident voices from controlling the conversation and actively make room for quieter participants. Focus groups can reach more people than individual interviews, but they are harder to organise and the social setting can influence what people are willing to say.

Questionnaires

A questionnaire presents open or closed questions in a form. It is useful when a project needs input from many users with limited time and resources, especially when the form can be distributed online.

Unlike an interview, an unclear question cannot be immediately rephrased. The wording therefore needs careful design, and a pilot test should be conducted before broad distribution. A short trial often reveals ambiguous questions, missing answers, and assumptions that made sense only to the researchers.

Closed questions produce quantitative data that is easier to aggregate and compare. Open questions add context but require qualitative analysis. Questionnaires work best when the research goal is already reasonably clear; they are less useful for exploring a domain or asking spontaneous follow-up questions.

Contextual inquiry

Contextual inquiry moves the research into the user's real environment. The researcher observes the work as it happens, looking at the tools, interruptions, constraints, and informal practices that an interview may never uncover.

The relationship is often described as an apprenticeship: the participant is the expert in their work, while the researcher watches, asks questions, and learns.

The method is guided by four principles:

• Context: go to the place where the activity actually happens.
• Partnership: researcher and participant collaborate to build a correct understanding of the work.
• Interpretation: observations are checked with the participant before becoming design conclusions.
• Focus: the researcher keeps the project's goal in view without ignoring important surprises.

Field studies reveal real behaviours that may remain invisible in interviews, questionnaires, or laboratory sessions. They are also expensive. Activities overlap, interruptions occur, and qualitative observations take time to analyse.

A laboratory makes the environment easier to control and allows rare situations to be reproduced. At the same time, people know they are being observed, which can change their behaviour and performance. No technique is neutral.

The strongest requirements analysis usually combines methods. Interviews explain what people believe they do, observation shows what actually happens, and questionnaires help test whether a pattern extends to a larger group. The goal is not to choose the most sophisticated technique, but to collect the evidence the design decision genuinely needs.