StrategizequalitativeBeginner

Observation: Ethnography

Direct observation of users in their natural context to understand real behaviors.

What it is

Main objective

Understand user behavior in their natural environment and usage context.

Detailed description

Ethnographic Research is a qualitative method that involves direct and prolonged observation of users in their natural environment, allowing deep understanding of their behaviors, motivations, and usage contexts. The researcher integrates into the user's environment, documenting daily activities, interactions, and routines, which facilitates the identification of latent needs and design opportunities. Scientific literature supports its effectiveness for discovering deep insights and improving user experience (Interaction Design Foundation; ACM). It is advisable to complement with interviews and data analysis to obtain a comprehensive view.

When it fits

Initial discovery stages or to understand real usage in context.

•When there are many unknowns
•In early product stages
•For new or innovative products

Use cases

WebMobile appsDesktop applicationsIn-person servicesLive interactions (banks, clinics, retail)

Effort level

High

Recommended number of users

5–10 users

What you gain and what you give up

Advantages

Ecological validity: By studying users in their natural environment, the data reflect reality far better than artificial lab studies.
Uncovers the unarticulated: It identifies needs users do not know they have, or cannot put into words because they are part of their subconscious routine.
Holistic understanding: It sees the user as a whole (social, physical, cultural influences), not just as an operator of a machine.
Minimizes social desirability bias: By observing the actual action, it reduces the risk that the participant lies or exaggerates to look good.

Disadvantages

Time and cost: Traditional ethnography takes a lot of time (months/years). Even "design ethnography" (days/weeks) is slower and more expensive than surveys or usability tests.
Observer effect (Hawthorne): The researcher's presence can alter the behavior of the people being studied, who may act differently once they know they are being observed.
Difficult analysis: It generates a massive amount of unstructured data. Analyzing and interpreting this data requires specialized skills and is prone to the researcher's interpretive bias.
Generalization: Because it works with small samples, it is hard to claim the findings are statistically representative of the whole population (though it aims for transferability, not statistical generalization).