Design a diary study
The hardest part of a diary study isn't building it—it's deciding what to ask for, how often, and over how long. This page covers those decisions. For how to actually configure them, see Create a diary study.
Shape each activity around one of five patterns
Most good diary activities look like one of five patterns. Picking one before you start writing questions keeps an activity focused on a single kind of answer—which is usually the difference between entries you can analyze and entries you can't.
Moments
Participants capture recurring experiences as they happen, in real time. This usually sounds like: "Show me every time you…"
Process
Participants document the key moments of a process as they move through it. This usually sounds like: "Show me your process as you go about…"
Inventory
Participants show you everything they use or own within a topic. This usually sounds like: "Show me all the ways you…"
Reflection
Participants explain their preferences, memories, and motivations. This usually sounds like: "Show me how you feel about…"
Ideation
Participants present their own ideas and suggestions. This usually sounds like: "Show me your ideas for…"
Decide your days and entries
Every activity has a number of days and a number of entries. Together they set the rhythm of your study, and the combination matters more than either number on its own. A few worked examples:
"Show me every time you eat a meal on-the-go this week. Submit a new entry for each meal, explaining where you bought it and how much you liked it."
7
4
"Show me your morning beauty routine by breaking it down into steps. Submit a new entry for each step."
1
3
"Show me your child's 5 favorite mobile apps. Submit a new entry for each app, and tell me what you like or dislike about each one."
3
5
Things worth thinking through before you commit:
Multiply, don't add. What participants feel is activities × entries × questions. Four entries of six questions is twenty-four answers, and it lands very differently than one entry of twenty-four.
Match the window to the behavior. If you're studying something that happens twice a week, a one-day activity asking for four entries can't be completed honestly—people will batch or invent.
Leave room for real life. Participants miss days. A window slightly longer than strictly necessary gets you better entries than one that forces a scramble.
Your activity days should add up to roughly your study length. The study-level duration and the per-activity days are set separately, so it's easy to end up with a seven-day study containing fourteen days of activities.
Write instructions that earn good entries
An activity's instructions do more work than its questions. They're what participants read before deciding what to film and when.
Say when to submit, not just what. "Submit an entry each time you make coffee" gets you moments; "tell us about making coffee" gets you one summary.
Name the media you expect. If you want the room in the shot, or the screen rather than the face, say so—people default to a selfie.
Put deadlines in the instructions as well as the due date field. Participants read instructions; they don't always notice dates.
Say what a good entry looks like. One concrete example of the level of detail you want does more than three sentences of guidance.
Pick the right device for each activity
Because you set the device per activity, you can match each one to where the behavior actually happens—phone for in-the-moment capture, computer for anything involving a website or a larger screen. This also shapes what participants need to have installed. For more information, see Device options in a diary study.
Before you launch
Test your study on a real device before it goes out. It's the only reliable way to find instructions that read clearly in the builder and confusingly in the app. For more information, see Test a diary study.
For a full worked example, see Example of an effective diary study.