Diary studies: when UX research needs time, not a lab

A diary study is the method where participants log what they do, think, and feel across days or weeks, in their own context. Use it when the question doesn’t fit a one-hour session. Habits, journeys, shifts in perception, and channel use over time need this format. NN/g treats the diary as a remote, asynchronous context method, complementary to field studies.

What the method answers

The Nielsen Norman Group article on diary studies defines the study as qualitative research that collects behavior, activities, and experiences over time and in context. Participants report interactions as they happen, for a few days, weeks, or longer.

The method covers questions a single session can’t hold:

Habits and frequency. When people open the product, and how often.
Attitudes and motivation. What drives the task, and how mood shifts across days.
Change. Learnability, loyalty, and brand perception after many interactions.
Journeys. The full path across several touchpoints.
Channels and devices. Which surface people pick, and why.
External factors. Collaboration, competitors, and side tools in the same flow.

Focus can be broad (general behavior), tied to one product, or tied to a long activity such as researching and buying a plan.

Diary versus usability testing and versus field study

Moderated usability tests and interviews pull people out of context and run short. The diary-studies guide is clear: contextual, longitudinal insight doesn’t come from one session.

The article on context methods groups diary studies, field studies, and contextual inquiry together. All three study people in real settings. Field work and inquiry observe on site, often for hours. The diary runs remote and async, for days or weeks, at lower cost and with reach to spread-out participants.

Pick the diary when the question needs time: repeated use, a long journey, a habit. Pick the field when the question needs the physical setup and the strategy in the moment of the task.

Three logging triggers

The same diary-studies guide splits three ways to ask for an entry:

Event-based. People log when the event of interest happens. Fit when every interaction matters and volume won’t overwhelm them.
Interval-based. Entries at a fixed cadence, such as end of day. Fit when events are too frequent to log one by one.
Signal-based. A prompt (text, push) asks for a response on the spot. Fit when a fixed interval would skew the time of day, or when forgetting is likely.

Many studies mix all three. The research question drives the mix, not a liking for one format.

Length and how many participants

Length follows how often the behavior happens. Daily use may fit in one week. Weekly use needs about three weeks. A purchase that takes two weeks of research needs about two weeks of logging.

For sample size, the guide gives rough bands until saturation:

Small discovery, fairly homogeneous group: about 5 to 12 people.
Broad discovery, varied group: about 12 to 30.
Large or academic work: about 30 to 50 (NN/g).

Recruit more people than you need. In a long study, many drop out along the way. Without enough data at the end, the themes never settle.

Short entries: five to ten minutes

The article on diary-study entries treats the entry as the core data unit. Mix closed, open-ended, and multimedia questions. Practical rule: put participant effort first. Close what you can close. Leave open what needs their own words. Ask for photo or video only when text isn’t enough.

Aim for five to ten minutes per entry. Past fifteen, people skip questions or quit logging. Pilot the form before the real study.

The engagement article repeats the same ceiling: more than fifteen minutes per entry is already too much. Fewer prompts, only what the research question needs.

How it fits Zero First

On Zero First, the platform covers the full path: connect sources, build audiences, read how each segment behaves, and push to ad destinations. Every screen was designed from scratch: sign-in, password recovery, the list with its empty state, simple-mode creation, and the detail view with insights.

People don’t finish with the product in one sitting. People return to the list, create another audience, and read insights again. That multi-day rhythm is what a diary study captures better than a one-hour test. There was no formal diary study on the case. The design, though, assumes repeated use and audience discovery on first contact. And that is just the kind of question the method answers.

Pilot, briefing, and wrap-up interview

Before the real study, run a short pilot. The pilot doesn’t produce analysis data. It tests the tool, the instructions, and entry time.

In the pre-study brief, align what to log, the minimum number of entries, and the incentive. Walk through the tool. Clear doubts before day one.

After the logging period, schedule a wrap-up interview with each person. Probe vague entries and ask for reflection on the whole period. Without that wrap-up, the log stays loose and the story is only half told.

For recruiting, the engagement article asks for a screener with an open-ended question. The answer shows whether the person writes with enough detail. Pay incentives in installments across long studies, not only at the end.

Where the sources diverge

What the diary collects. The context-methods article frames the diary as mostly attitudinal data, with self-reported behavior. The definition guide talks about collecting behavior, activities, and experiences in context.
Sample size. The definition guide publishes numeric bands until saturation. The engagement article barely fixes a number and leans on motivation, screener, and a light tool.
Entry format. The entries article balances closed, open, and multimedia, with a five-to-ten-minute target. The engagement piece cuts prompts to the minimum and treats fifteen minutes as a hard ceiling.

The common mistake

The mistake is treating the diary like a usability test stretched over weeks. Long entries, too many prompts, and a single payment at the end lead to dropout and made-up data. Another mistake is skipping the wrap-up interview. Without it, you read the log and never confirm the why. Build a longitudinal question, a short entry, a pilot, and a close. The diary answers what time does to use. The one-hour session answers something else.