AEIOU: Five Buckets for Field Notes

WHY IT EXISTS: Qualitative field research produces messy, unstructured notes. Without a shared taxonomy, teams drown in raw observations and struggle to spot patterns during analysis. AEIOU was created by Rick E. Robinson as a categorizing heuristic to give researchers a lightweight, common language for sorting what they see in the field.

THE MENTAL MODEL: Think of AEIOU as a five-column spreadsheet you keep in your head. Every observation gets dropped into one of five buckets. Activities are the actions people take. Environments are the settings and contexts. Interactions are the exchanges between people, tools, and systems. Objects are the artifacts and details in the space. Users are the people themselves, their traits and behaviors. The goal is not perfect classification; it is forcing disparate notes into a structure that reveals gaps and themes.

HOW IT WORKS: Before a session, print or prepare a worksheet with five sections labeled A, E, I, O, U. During the observation, capture what you see, hear, and feel, routing each note to the relevant bucket. After the session, review the columns as a team. One column may be overflowing while another is empty, which tells you where to dig deeper or adjust your research questions. Because Robinson designed the framework as editable heuristics, teams should refine the definitions to match their project goals.

WHEN TO USE IT: Use AEIOU during contextual inquiries, usability tests, and ethnographic field studies where you need to convert live observations into categorized data quickly. It shines when multiple researchers are taking notes and you need to align everyone on what to capture. It also works as a cover-page reminder on discussion guides to keep observers focused.

WHEN NOT TO USE IT: Do not use AEIOU when you need strict statistical rigor or fully structured quantitative data. It is not a replacement for survey instruments or analytics. It also fails if the team treats the five categories as immutable laws; forcing observations into a bucket that does not fit distorts the data and hides context.

ONE CANONICAL EXAMPLE: A software team is observing nurses at a hospital station. Under Activities they note frequent password re-entry. Under Environments they record poor lighting and noise. Under Interactions they see nurses asking each other for login help. Under Objects they list sticky notes with credentials taped to monitors. Under Users they describe shift changes and varying tech comfort. Reviewing the worksheet later, the team realizes the security-object and interaction columns tell a single story: authentication friction is forcing unsafe workarounds.

Source: openpracticelibrary.com