User Research in Product Interaction Design - Observation


In any stage of product design, observation is a useful data collection technology. In the early stage of design, observation helps designers understand user situations, goals and tasks, while in the evaluation or development stage, it can be used to study whether the prototype can effectively and efficiently support the completion of tasks.

This paper first explains the definition of observation method, then introduces the direct observation method and indirect observation method in detail, and analyzes their advantages and disadvantages, which is helpful for product design newcomers to learn the theory. You need to read all kinds of articles and books for details.

1. Observation method

1. Definitions

User observation method is derived from ethnographic research. It refers to a method in which researchers observe the studied object according to certain research objectives, research outline, their own senses and auxiliary tools, so as to obtain data.

Ethnographic research: Seeing everything as a stranger, not imposing a frame, just making direct observations. It has now become a popular interaction design method because it allows designers to gain detailed insights into people's behavior and the use of technology that are otherwise unavailable. 


Observation can be divided into two dimensions - direct or indirect, two environments - field or controllable environment.

2. The method of direct observation

1. Classification by setting environment

(1) Artificial setting / controllable environment

It is carried out in the laboratory, or in a place similar to the real environment, which is carefully arranged and transformed by the researcher. This kind of setting facilitates the control of experimental conditions (variable conditions), saves a lot of cost, and avoids the influence of accidental factors.

Direct observation in a controlled environment:

A survey method in which researchers observe users performing specific tasks in a controlled environment. Typically done in the lab, the focus is on user action details, capturing activity details.

It should be noted that one of the observation problems is that the researcher does not know what the user is thinking. In a controlled environment, the user can tolerate more disturbances. Thinking aloud is a useful method for understanding human thoughts.

Think Aloud Method: Externalize their thought process by asking people to say aloud everything they are thinking and doing.
However, the situation of silence is one of the biggest problems of thinking aloud. The researcher can remind participation saying aloud, but it is intrusive. Constructive interaction (pair participation/collaboration) can be used, that is, two users communicate collaboratively, which makes many subconscious and self-examination mental state manifested

Observation points:

Observe whether the user has completed the task independently (effectiveness problem), whether the user has performed ineffective operations (efficiency problem), and whether there is dissatisfaction (satisfaction problem).


  • Prepare the interface involved in the test
  • According to the test goal, design the task (task description is presented to the user in the form of script)
  • Invite suitable test users (2-8 test users produce the most effective results)
  • Start testing
  • Organize and analyze data

(2) Natural scenery/field environment

It is the real environment for real users to use the product, and it is easy to stimulate the natural behavior of users.

2. Direct field observation

A survey method in which researchers observe users performing everyday tasks in a natural environment. It can observe the user's actions and usage in the real situation, and focus on the user's interaction with surrounding factors.

Note that there is no absolute end criterion for direct observation in the field, and the current assessment criterion is that similar behavioral patterns recur repeatedly.

3. Advantages and disadvantages of field observation and laboratory observation

The advantage of field observation is that the observer sees how to use technology to solve real problems in real situations, and can observe the real process of users' use.

The disadvantage is that researcher records can be subjective.

The advantages of lab observations are that they are easier to replicate, so several users can perform the same task, specific usability problems can be identified, users' performance can be compared, and the average time and number of errors to complete a particular task can be calculated.

The disadvantage is that the research environment is artificial and cannot account for the usage in the actual environment.

4. When two observations are useful

Depends on research goals.

Lab observations help examine interaction details and uncover usability issues; field observations reveal how technology is used in real-world settings and how it affects user behavior.

2. Classification by structure

(1) Open observation (unstructured observation)

In qualitative research, the second observation is made after gradually screening the recorded data, and the valuable data is deeply and carefully observed.

This type of approach can yield massive amounts of data, open up access to comprehensive and complex data for researchers, uncover previously unknown data, and lead to innovative designs. Typically used for innovation-focused projects.

However, the cost of sorting is huge, and it is necessary to purposefully screen and summarize valuable data.

(2) Structured observation

In quantitative research, pre observation is first carried out, and then the observed behaviors are coded and classified. In the formal observation, according to the clear observation factors, the observation phenomenon is preset according to the grade, which can collect clear data, which is very conducive to quantitative analysis and reduce the deviation of researchers' subjective differences. Usually used for improved design. 

Observation steps:

  • Clarify the research direction: clarify the purpose of the research and the theme of the research (object; problem; specific situational conditions)
  • Develop an observation plan: make the observation specific and indexed, clarify the observation object and observation method, and determine the sampling factors
  • Specify the observation content and determine the observation frame. POEMS framework: guide observers when designing observations, what to observe, how to record and consider later analysis. P stands for people, that is, the observer; O stands for Object, which refers to the object seen during observation, the product itself or related objects; E stands for environment, which refers to the environment in which the observation content is located; M stands for message, which refers to the process of the observed event. , possibly related information; S stands for service, which refers to the service that may be involved in the event of the observed.
  • Pre-observation: Eliminate the problem
  • Observation: make comprehensive, accurate and orderly records, and respect objective facts
  • Sorting and analysis

3. Precautions

In direct observation, researchers especially need to take into account the "observer effect" to avoid affecting the results of the study.

Observer effect: Whether the observed objects are aware that they are being observed sometimes affects the results of the observation. If they don't know they are being observed, then their behavior will naturally be true.

Such methods will not miss relevant information, can grasp the true meaning of the results and effectively verify the views of members, but it is inevitable that researchers need to intervene in real events to observe the integrity of things (participatory observation). Once the observed person knows the true identity of the researcher, it may affect the natural behavior and even the life of the observed person, so it is necessary to carefully consider the specific implementation scheme.

3. Indirect observation

1. Diary

Asking participants to keep regular diaries about their activities does not take up much of the researcher's time and requires no special equipment, suitable for long-term studies, but depending on the reliability of the user participant, participants may exaggerate memories or forget details.

2. Empirical sampling ESM

Relying on participants to record information about their daily lives, participants answered specific questions at random times, helping to gather information immediately.

3. Interaction log

Install software on devices that record user activity, capture actions, help analyze and understand how users complete tasks, undisturbed, but designed for privacy.

4. Web Analytics

User behavior is closely tracked, behavioral data is collected, and user goals are assessed, often for business and market research.

The advantages and disadvantages of the observation method

1. Advantages

The data is complete and true, showing the actual situation in an all-round way; it can observe the research objects that cannot answer the questions.

2. Disadvantages

High cost; records are subject to subjective influence of researchers; Privacy issues.

4. Supplementary knowledge points

1. Fieldwork

It is the most representative method in anthropological research, requiring the investigator to live together with the respondents for a period of time, from which to observe and understand their society and culture.

Investigators conduct overall research on people, occasions and systems, investigate specific regional cultures, and use the observation results to find the needs of target users.

It is often used for latent demand discovery, using observations to verify that what users say is what they actually do.

2. Field experiment method

Also known as field or situational experimentation, field experimentation. It is one of the research methods of social psychology, which refers to the research activities of social psychology carried out outside the laboratory, in the actual and natural social life and situation. The field test method has timeliness, authenticity and validity.

Categories: Artificial Field Trials (Stakeholders Participate), Scenario Field Trials (Subject Knows Trials), Natural Field Trials (Randomness and Authenticity)

Advantages: subjects do not know that they are subjects, so they will not have reaction bias; Since the independent variables are controlled, the causal relationship between the variables to be studied can be seen.

Disadvantages: The degree of control over independent variables is low, and the influence possibility of irrelevant factors is high; High cost.

5. Ending

Good design should exceed user expectations and stimulate latent demand. 

Nowadays, the homogeneity of similar products is serious. Designers, product managers and user researchers need to carefully observe the target users, find the emotional fluctuation points and operation problems, and use them as design points to widen the gap.

If you like the article, please share it with others with page link, thanks for your supporting! ❤


1 comment

  • Ted Pepper

    Dear admin, Your posts are always well researched and well written.

Leave a comment

Please note, comments must be approved before they are published

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

This section doesn’t currently include any content. Add content to this section using the sidebar.