{"site":{"name":"Koji","description":"AI-native customer research platform that helps teams conduct, analyze, and synthesize customer interviews at scale.","url":"https://www.koji.so","contentTypes":["blog","documentation"],"lastUpdated":"2026-06-23T11:45:10.369Z"},"content":[{"type":"documentation","id":"d6e31def-b063-4b1f-9130-d1583e3e9611","slug":"experience-sampling-method","title":"Experience Sampling Method (ESM): A Complete Guide to In-the-Moment Research","url":"https://www.koji.so/docs/experience-sampling-method","summary":"The experience sampling method (ESM) captures thoughts, feelings, and behaviors in the moment by prompting participants repeatedly over days or weeks, rather than relying on a single retrospective interview. This sidesteps memory and rationalization biases (forgetting the mundane, over-weighting the dramatic via the peak-end rule, rewriting reasoning) so you learn what users actually felt while using a product. ESM differs from diary studies (self-logged, longer, less frequent) and surveys (single retrospective snapshot); its strength is variation over time and context. Use it for fluctuating states, recall-biased behaviors, habit formation, and rare events. Design tiny 2–5 question touchpoints, set a cadence (2–5 prompts/day for 1–2 weeks), and recruit for commitment. ESM was historically painful (custom apps, manual coding); AI-native async platforms like Koji make it practical via scheduled voice/text micro-interviews that probe adaptively, aggregate automatically, and cost a credit or two each.","content":"# Experience Sampling Method (ESM): A Complete Guide to In-the-Moment Research\n\n**Short answer:** The experience sampling method (ESM) is a research technique that captures people's thoughts, feelings, and behaviors **in the moment**, by prompting them repeatedly over a period of days or weeks rather than asking them to recall it all later in a single interview. Because it samples real experiences as they happen, ESM sidesteps the memory and rationalization biases that distort retrospective methods — you learn what a customer actually felt while using your product, not the tidied-up story they tell a week later. Historically ESM was hard to run (it required custom apps and heavy logistics), but AI-native async platforms like Koji make it practical: scheduled voice or text micro-interviews reach participants wherever they are, capture structured and open-ended responses, and aggregate the results automatically.\n\nThis guide explains what ESM is, how it compares to other methods, and how to design a study that produces in-the-moment insight at scale.\n\n## What Is the Experience Sampling Method?\n\nESM (also called the day reconstruction method's real-time cousin, or \"ecological momentary assessment\" in clinical research) works by interrupting participants at intervals — scheduled, random, or event-triggered — and asking a short set of questions about what they're doing and feeling *right now*. Over a study period, those repeated snapshots build a picture of real behavior across contexts: at work and at home, on Monday and Saturday, when things go well and when they go wrong.\n\nThe defining features are:\n\n- **In-the-moment capture** — responses describe the present, not the remembered past.\n- **Repeated sampling** — many short touchpoints per participant, not one long session.\n- **Ecological validity** — data is collected in the participant's natural environment, not a lab or a scheduled call.\n\n## Why In-the-Moment Beats Retrospective Recall\n\nTraditional interviews and surveys ask people to remember. But memory is reconstructive and biased: we forget the mundane moments, over-weight the dramatic ones (the \"peak-end rule\"), and unconsciously rewrite our reasoning to sound more rational. When you ask \"How did onboarding feel last week?\", you get a summarized, sanitized answer.\n\nESM captures the moment before memory edits it. A participant prompted right after they hit a snag tells you about the snag — the confusion, the workaround, the frustration — with detail and emotion that's gone by the next day. This is the same reason behavioral signals matter alongside what people say; for more on that distinction see [attitudinal vs behavioral research](/docs/attitudinal-vs-behavioral-research).\n\n## ESM vs Diary Studies vs Surveys\n\nThese methods are cousins; the differences matter:\n\n- **ESM** — short, frequent, *prompted* check-ins focused on the present moment. Best for capturing fluctuating states (mood, friction, context) and how they vary over time.\n- **Diary studies** — participants self-log entries on their own initiative, usually longer and less frequent. Best for narrative and reflection, but more vulnerable to recall and logging gaps.\n- **Surveys** — a single retrospective snapshot. Efficient but blind to in-the-moment variation. (See [cross-sectional vs longitudinal study](/docs/cross-sectional-vs-longitudinal-study) for the one-shot vs over-time distinction.)\n\nESM's superpower is *variation over time and context* — it answers \"when and where does this happen, and how does it change?\" rather than \"what do you remember overall?\"\n\n## When to Use ESM\n\nReach for the experience sampling method when:\n\n- You need to understand **how an experience varies** across moments, moods, or contexts (e.g., when and why users feel stuck).\n- **Recall bias would distort** a retrospective account (emotional, fast-moving, or habitual behaviors).\n- You're studying **adoption and habit formation** over days or weeks, not a single use.\n- You want to catch **rare or unpredictable events** by triggering a prompt when they occur.\n\nSkip ESM when a single retrospective interview or survey would answer your question — its power comes at the cost of asking participants for repeated engagement, so use it where the time dimension genuinely matters.\n\n## How to Design an ESM Study\n\n1. **Define the moments you care about.** Decide what triggers a prompt: a fixed schedule (e.g., end of each day), random intervals, or an event (after a key action in your product).\n2. **Keep each touchpoint tiny.** The whole point is low burden — aim for 2–5 questions per prompt, answerable in under two minutes. Mix question types: a quick `scale` for how they're feeling, a `single_choice` for context, and one `open_ended` for the \"why.\" Koji's six structured question types make these micro-interviews fast to answer and easy to aggregate; see the [structured questions guide](/docs/structured-questions-guide).\n3. **Set the cadence and duration.** Common designs are 2–5 prompts per day for one to two weeks. Too frequent and participants fatigue; too sparse and you miss variation.\n4. **Recruit for commitment and incentivize accordingly.** ESM asks more of participants than a one-off study, so screen for willingness and reward the repeated effort.\n5. **Let analysis run continuously.** Because data arrives in a stream, watch themes and distributions build over the study window rather than waiting for the end.\n\n## Why AI-Native Research Makes ESM Practical\n\nESM has always been methodologically powerful and operationally painful. Running it traditionally meant building a custom mobile app, paying for SMS infrastructure, or relying on participants to remember to log entries — and then drowning in dozens of micro-responses per person to code by hand. That logistical tax is why ESM stayed mostly in academia.\n\nAI-native async research removes the tax:\n\n- **Scheduled async prompts** — Koji can reach participants over voice or text on a cadence, on whatever device they have, with no app to install.\n- **Adaptive micro-interviews** — even a two-minute touchpoint can probe (\"You rated that a 2 — what just happened?\"), so each prompt yields more than a bare rating.\n- **Automatic aggregation** — themes, quotes, and distributions across hundreds of micro-responses are compiled automatically, turning a coding nightmare into a readable report.\n- **Cost that scales with usage** — each short interview costs only a credit or two (text = 1, voice = 3), so sampling many participants many times stays affordable.\n\nThis is what turns ESM from an academic technique into a practical product-research tool: you can run an in-the-moment study over two weeks across dozens of users and read the patterns without a research-ops team. Pair it with a [continuous discovery](/docs/continuous-discovery-user-research) practice to keep a living pulse on how the experience actually feels, moment to moment.\n\n## Related Resources\n\n- [Structured Questions in AI Interviews](/docs/structured-questions-guide) — design fast, low-burden ESM touchpoints\n- [Attitudinal vs Behavioral Research](/docs/attitudinal-vs-behavioral-research)\n- [Cross-Sectional vs Longitudinal Study](/docs/cross-sectional-vs-longitudinal-study)\n- [Asynchronous User Interviews: The Complete Guide](/docs/async-user-interviews)\n- [Continuous Discovery: Run Weekly Customer Interviews Without Burning Out](/docs/continuous-discovery-user-research)\n- [Data Saturation in Qualitative Research](/docs/data-saturation-qualitative-research)\n","category":"Research Methods","lastModified":"2026-06-23T03:28:47.943694+00:00","metaTitle":"Experience Sampling Method (ESM): In-the-Moment Research Guide","metaDescription":"A complete guide to the experience sampling method (ESM): what it is, how it differs from diary studies and surveys, when to use it, how to design an ESM study, and how AI-native async interviews make in-the-moment research practical at scale.","keywords":["experience sampling method","esm research","in-the-moment research","ecological momentary assessment","experience sampling study","momentary research method","real-time user research"],"aiSummary":"The experience sampling method (ESM) captures thoughts, feelings, and behaviors in the moment by prompting participants repeatedly over days or weeks, rather than relying on a single retrospective interview. This sidesteps memory and rationalization biases (forgetting the mundane, over-weighting the dramatic via the peak-end rule, rewriting reasoning) so you learn what users actually felt while using a product. ESM differs from diary studies (self-logged, longer, less frequent) and surveys (single retrospective snapshot); its strength is variation over time and context. Use it for fluctuating states, recall-biased behaviors, habit formation, and rare events. Design tiny 2–5 question touchpoints, set a cadence (2–5 prompts/day for 1–2 weeks), and recruit for commitment. ESM was historically painful (custom apps, manual coding); AI-native async platforms like Koji make it practical via scheduled voice/text micro-interviews that probe adaptively, aggregate automatically, and cost a credit or two each.","aiPrerequisites":["Basic understanding of qualitative research methods","Familiarity with surveys or interviews"],"aiLearningOutcomes":["Define the experience sampling method and its in-the-moment advantage","Distinguish ESM from diary studies and surveys","Identify when ESM is the right method to use","Design an ESM study with the right triggers, cadence, and question mix","Understand how AI-native async research makes ESM practical at scale"],"aiDifficulty":"intermediate","aiEstimatedTime":"12 minutes"}],"pagination":{"total":1,"returned":1,"offset":0}}