A multi-day diary study asks people to record their experiences over several days instead of answering once. On an online bulletin board, this means they log in each day, respond to a short set of questions or tasks, and often share photos or video. Done well, you see real routines and changes over time. Done badly, people quit by Day 3. This guide covers how to design, run, and support a diary that people actually finish.
When a diary is the right method
Use a diary when the thing you want to understand unfolds over time or happens in places you cannot observe: a morning routine, a week of grocery decisions, how a new product fits into a household after the novelty fades. Memory is unreliable for these. People forget, summarize, and tidy their stories. A diary captures the event close to when it happens, which is its main advantage over a one-off interview.
Why diaries fail
- Days ask too much, so people feel behind by the second morning.
- Media tasks break on camera permissions and nobody is there to help.
- Platforms treat Day 3 like another survey page, with no sense of progress.
- Instructions are long and unclear, so people guess what you want.
- Incentives and expectations are not explained up front.
Choosing 3, 5, or 7 days
| Length | Best for | Watch out for |
|---|---|---|
| 3 days | Focused topics, quick turnarounds, concept reactions | Less time to see changes in behavior |
| 5 days | Typical routines, usage, and shopping patterns | Needs a clear middle-day purpose |
| 7 days | Full weekly cycles such as weekday versus weekend | Fatigue; keep daily effort light |
A simple day architecture
- 1Day 1: context, trust, light tasks, and an NDA if needed.
- 2Middle days: lived experience, stimuli, and media captured in the moment.
- 3Final day: tradeoffs, wrap-up, and future intent.
Aim for roughly ten to twenty minutes per day. If a day is heavy, move some questions to a lighter day rather than hoping people will push through. A useful rule: if a task would feel annoying to you on a tired Wednesday, shorten it.
Writing prompts people can answer
- Ask about specific recent events (“Think of the last time you cooked dinner…”) rather than generalizations (“How do you usually cook?”).
- Give one task per prompt, so it is clear when they are done.
- Tell people what a good answer looks like: a sentence or two, three photos, thirty seconds of video.
- Use neutral language that does not hint at the answer you want.
- Add a final “anything else we should know?” on the last day to catch what you did not think to ask.
Media without misery
Separate “upload an image” from “take a photo.” Require camera capture when authenticity matters, for example when you need a picture of what is in the fridge right now. Give one clear task per prompt, show an example when possible, and expect some permission failures. The fix is to put support inside the board, not in an email thread.
Keep people engaged
- Explain the schedule and incentive on Day 1.
- Send timely reminders when a day unlocks.
- Show progress so people know how much remains.
- Use follow-up questions on later days to show you read what they wrote.
- Watch daily completion so you can nudge before people drift away.
Moderating across days
A diary is not “set and forget.” Plan time each day to read new answers, reply to a handful of respondents with a specific follow-up, and note which topics need deeper probing tomorrow. Personalized follow-ups are the strongest retention tool you have, because they tell respondents their answers matter.
Example: a 5-day study on morning routines uses Day 1 for habits, Day 2 for a photo of breakfast prep, Day 3 for a short video of leaving home, Day 4 for reactions to a new product, and Day 5 for a wrap-up. Each day is under fifteen minutes.
FAQ