Session instrument
Experience sampling
ESM
Measures: The experience at the instant it happens, in the real world
It is how you watch the peak–end rule happening in your product: comparing what the person answered in the moment with what they tell you at the end of the week. When the two diverge, what decides whether they come back is memory, not the sum of the days.
The method was created to study exactly what this site calls flow state: instead of asking afterwards, the person is interrupted at randomly drawn moments of the day to find out what they are doing and how they are feeling (Mihaly CsikszentmihalyiDescribed the flow state and created the experience sampling method that measures it outside the laboratory.2 references in this work:1990Flow: the psychology of optimal experience1987Validity and reliability of the experience-sampling methodSee on Wikipedia ↗See in the bibliography → and Larson, 1987). For design there is a second use, which may be worth more: the divergence between the account in the moment and the retrospective account is the signature of peak–end at work. It is also the only instrument on this page that leaves the laboratory and the moderated test and goes to where the product is actually used, with traffic, a crying child and the battery at 4%.
How to run itA notification at random times, over a few days, with two or three short questions. At the end of the period, ask for the overall account and compare it with what was answered live.
SeesIt captures use in the real context and exposes the difference between living and remembering.
Does not seeIntrusive and laborious: the response rate drops over the days, and whoever drops out midway biases the rest.