Field study guidelines
One board per participant for observing users in their real environment, with the synthesis structure built in.

- Time
- Half-day to full-day visit + 30 min synthesis
- People
- 1–2 researchers on site
Overview
Going to where the work happens — a plant, a depot, a dispatch desk — and watching. For industrial software this routinely surfaces the constraints that never appear in a meeting: gloves, glare, noise, no signal, and a screen that has to be read from two metres away.
The board is a template with a worked example attached (a utility dispatcher emergency-response study). You duplicate it once per participant, so each session ends up as its own annotated board rather than a pile of notes. Its real contribution is the synthesis structure: pain points and insights have a required shape, so an observation cannot quietly become a finding without evidence.
When to use it.
During concept validation, when the concept needs to survive contact with a real environment rather than a quiet room. Also in framing, when a protopersona needs testing against someone who actually does the job.
Who to involve.
One or two researchers on site — the board records who facilitated and who took notes, because one person cannot do both. Plus a host who can get you access and vouch for you. More than two observers changes the behaviour you came to see.
Effort.
High prep High engagement Site access, consent and a safety induction usually take longer to arrange than the visit itself. Synthesis immediately afterwards is part of the session, not optional.
Outcomes.
A validated or challenged protopersona, an annotated flow map, a prioritised pain-point list with evidence attached, and a set of observation-to-opportunity chains. Plus the workarounds people have built that nobody thought to mention.
Recommendations.
Ask “can I take a photo of your workspace?” every time — the board is blunt that first-time researchers skip environment capture and that it is the most valuable thing they miss. Never write while the participant is talking; look at them and note immediately after. And do the synthesis before you leave: the board’s line is that ten bullet points written straight afterwards beat detailed notes written a week later.
How to run it
Set the study goals before you go
Pin them at the top of the board so observations stay focused. The example separates primary goals (validate the protopersona, validate the flows) from secondary (understand the real environment, surface unmet needs and workarounds) and emerging (assess emotional and cognitive load, test language and terminology).
Duplicate the board, once per participant
One board per person. Fill in the anchor table at the top of each: study name, session date and time with timezone, participant, researchers present, location, planned versus actual duration, consent confirmed, and recording method. Recording consent is tracked separately from consent to take part.
Validate the protopersona
Paste the protopersona card onto the board with three columns beside it —
confirmed ,challenged ,new . Sort your observation stickies into them after the session. That sorted board is the persona validation artifact.Capture the environment
A wide shot of the setting, close-ups of the devices in use, a short walkthrough video if permitted, or a sketch. Record role and years of experience, current tools and which of them fail, how often emergencies happen, and tech comfort. Note the interruptions and the time pressure they normally work under.
Tag the flow live
Paste screenshots of each step side by side. As the participant moves through them, place a sticky directly on the step where something happened:
red for failures,green for successes,yellow for observations.Log the behavioural detail
For each tagged step, log the first action taken versus the expected one, any hesitation over three seconds, task completion and time, and a confidence rating out of five. You leave with an annotated flow map you can take into critique.
Record pain points in three columns
What you saw, what it costs the user, and what evidence you have. If you cannot fill all three, it is not a pain point yet — it is an observation. This replaced freeform notes, which were impossible to prioritise.
Photograph every workaround
Every sticky note on a monitor, every paper log, every manual timer exists because the system did not solve the problem. Photograph them and ask the person to explain each one, capturing the answer verbatim. These are the clearest design opportunities on the board.
Synthesise within 30 minutes
While memory is fresh. Write a messy “hot takes” cluster first and refine later. Then build the chains: from observation to pattern, insight and opportunity, each starting from something actually observed rather than a feeling. Two or three observations before you may call something a pattern, one chain per theme, three to five chains per session.


