Diary Study

Asking participants to record their own experience over days or weeks, capturing behaviour no single session can show.

Process & Methods

Some behaviour only exists over time: how a habit forms or fails in the first two weeks, what happens on the days somebody does not open the app, how a monthly process actually unfolds, when frustration accumulates rather than spiking. A single session cannot see any of it. A diary study asks participants to log entries as they go — prompted at intervals or triggered by events — producing a record of real moments instead of a recollection.

The main design problem is participation decay. Entries are rich on day one and thin by day five unless the burden is small and the prompts are specific. Ask for a photo or a voice note rather than an essay, prompt at the moment rather than at the end of the day, and plan a short interview afterwards to unpack what the entries mean — the diary tells you when something happened, the debrief tells you why.

In practice

A fintech app assumed users abandoned after onboarding because the product was confusing. A two-week diary study showed the drop was on days 3–5, when people were waiting for a bank verification that gave no status and no estimate. The app was not confusing; it was silent. Adding a progress state to the waiting period cut week-two abandonment by roughly a third.

Where teams get it wrong

  • Asking for too much per entry, which produces detailed day-one logs and nothing after.
  • Prompting at the end of the day, which collects recollection rather than the moment.
  • No follow-up interview, leaving entries without the reasoning behind them.
  • Running one over a period that does not contain the behaviour you care about.
  • Paying only on completion, which biases the sample towards the highly motivated.

( Learn more )

Related terms

All terms

Defined by Mara Bureau LLP. Last reviewed . Corrections: hello@mara.ink.

( You may ask )

Questions about Diary Study

Didn’t find your answer?

Ask us directly — a designer replies within one business day.

Contact us
  • Long enough to contain the behaviour — one to two weeks for onboarding and habit formation, a full cycle for anything monthly. Longer studies need lighter entries or participation collapses.

  • When the thing you want to understand happens across days rather than in one sitting: habit formation, waiting periods, intermittent use, or frustration that builds rather than spikes.

( Contact )

What problems are we solving?

At our core, we believe in delivering high-quality designs in days, not weeks.

What do you need?

( Contact )

What problems are we solving?

At our core, we believe in delivering high-quality designs in days, not weeks.

What do you need?