# Field Notes by Voice: Capture Research Observations Faster > Field notes decay within hours of leaving the field. Here's how to capture observations and reflexive memos by voice, before the detail disappears. - **Author**: Sami AZ - **Published**: 2026-08-10 - **URL**: https://klu.so/blog/field-notes-by-voice --- Field notes and reflexive memos are supposed to be written right after each session, while impressions are still sharp, but in practice they get deferred to the evening or the end of the week and lose most of their detail. Voice fixes the timing problem: you speak your observations into your phone in the car, on the walk back, or outside the building, minutes after leaving the field, and an app turns them into dated, structured, searchable notes. Flint does this in one press and keeps notes on your device, which matters when observations concern participants. Recording participants themselves is a separate matter governed by your ethics approval and consent, this is about capturing your own observations. Every qualitative methods text says the same thing: write your field notes as soon as possible after the session, not at the end of the study. Every researcher who has actually done fieldwork knows why that advice is so often ignored. You finish three hours of observation, you are tired, you have a drive ahead of you, and the laptop stays shut. By the time you write, the thick description has thinned into a summary of what you can still remember. Why Field Notes Decay So Fast Observational detail is perishable in a way that is easy to underestimate. What fades first is not the broad narrative, which you will remember for weeks, but the specifics that give field notes their analytic value: the exact phrasing someone used, the pause before an answer, the arrangement of a room, the thing that surprised you and the reason it surprised you. Reflexive material is even more fragile. Your emotional reaction to a session, the discomfort, the assumption you caught yourself making, the moment you realised your question had been leading, is precisely what a reflexive journal is supposed to capture, and it is gone within hours. Methodological guidance is explicit that you should record emotional reactions, surprises, and discomfort rather than only logistical details, and those are exactly the things a tired evening write-up flattens out. There is also a rigour dimension. Memos and journals form part of the audit trail that makes a study transparent and auditable, and a trail reconstructed from memory a week later is a weaker trail. The dates, the sequence of decisions, and the reasoning behind an emerging interpretation matter as evidence of how the analysis actually developed. The Bottleneck Is Typing, Not Discipline It is tempting to treat deferred field notes as a discipline problem. It is really a mechanics problem. Writing detailed notes requires a surface, a device, a chunk of uninterrupted time, and enough energy to compose prose, and fieldwork rarely leaves you with all four at once. Speaking requires none of them. You can produce a rich, detailed account far faster by talking than by typing, and you can do it standing outside a building, sitting in a car, or walking to a station. That difference is what moves the note from "later, when I get home" to "now, ten minutes after I left," which is the entire ballgame for detail retention. Talking also produces better reflexive material than typing does. Writing invites you to compose and tidy; speaking lets you think aloud, follow a tangent, admit confusion, and say the thing you have not fully formed yet. For a reflexive journal, that unpolished quality is a feature, since the point is to document your thinking honestly rather than to present it well. Making Spoken Observations Usable A folder of audio files is not a set of field notes, so the tool has to close the gap between talking and a usable record. Each recording should be turned into a clean, structured note automatically rather than left as a raw transcript, so you can read it back at a glance during analysis. Every note needs to be dated, because chronology is part of the audit trail. And the whole set should be searchable and queryable across sessions, which is where the real analytic payoff shows up: partway through a study you can ask what you have observed about a particular theme and get your own accumulated thinking back across many sessions, which is exactly the move that surfaces patterns and helps you notice when your interpretation shifted. How to Do It With Flint Flint fits the workflow because it removes the delay between leaving the field and recording what you saw. Capture is one press. On iPhone you can start from the Action Button or Lock Screen widget without unlocking, so you can begin speaking within a minute of walking out, when the detail is still intact. Talk through what happened, what struck you, and how you reacted, and Flint turns it into a structured, dated note rather than a wall of transcript. You can pick the shape you want, a descriptive account, a bulleted list of observations, or a first-person reflexive entry, and refine it with a follow-up instruction. Because every note is transcribed and searchable, and you can ask questions across the whole set, a study's worth of field notes becomes something you can interrogate rather than reread. The original audio is kept alongside each note, so you can go back and check exactly how you described something. And Flint is local-first, so notes stay on your device rather than in a vendor's archive, which is the right default when your observations concern real people. It is a one-time $12rather than a subscription, which is a genuine advantage on a student or small project budget. The honest limits. Flint is not qualitative analysis software: there is no coding, no theme building, no codebook, and no integration with NVivo, Atlas.ti, Delve, or Quirkos, so it complements a CAQDAS workflow rather than replacing it. It is local-first rather than fully offline, meaning some AI processing happens in the cloud, so if your ethics approval or data management plan requires that participant-related data never leaves the device or must reside in a specific jurisdiction, check that carefully before using it, and consider using pseudonyms and participant IDs in spoken notes rather than names. And recording participants themselves is an entirely separate question governed by your consent procedures and ethics approval; this workflow is about capturing your own observations after the fact. Flint is available on the App Store and on Google Play. A Field Notes Routine That Survives Real Fieldwork Build the habit around leaving, not arriving home. The moment you are out of the setting and somewhere you can speak freely, record for two to five minutes before you do anything else. Cover three things: what happened descriptively, what struck you or surprised you, and how you were reacting, including anything that felt uncomfortable or that you might be assuming. Use participant IDs rather than names. Then let it go. Later, when you sit down to write formally or to code, you are working from a rich contemporaneous account rather than an eroded memory, and during analysis you can search across every session to trace how your interpretation developed. Frequently Asked Questions When should field notes be written? As soon as possible after each session, not at the end of a study. Observational and reflexive detail decays within hours, and it is the specific detail, exact phrasing, surprises, your own reactions, that disappears first and matters most. Can I record field notes as voice memos? Yes, and it solves the timing problem that causes most field notes to lose detail. The key is using a tool that converts each recording into a dated, structured, searchable note, otherwise you accumulate audio you never revisit. What is a reflexive journal? A running, dated record of your decisions, reactions, and assumptions kept throughout a study. It supports transparency and forms part of the audit trail, and guidance recommends writing entries after fieldwork sessions and revisiting them during analysis. Does dictating field notes replace qualitative analysis software? No. Voice capture handles the recording of observations and memos; coding, theme development, and formal analysis still belong in dedicated qualitative software. The two work together. Is it ethical to record participants for field notes? Recording participants is governed by your ethics approval and consent procedures and is separate from capturing your own observations afterward. Dictating your own notes after leaving the setting does not record anyone else. Where should voice field notes be stored? Somewhere consistent with your data management plan. Notes about participants are personal data, so check retention, access, and location requirements, prefer tools that keep data on your device, and use pseudonyms or participant IDs when speaking. The best field notes are the ones you record before the detail fades. With Flint, one press captures your observations minutes after you leave the setting, and you get a dated, structured, searchable note kept on your own device. One-time $12, no subscription. Download Flint on the App Store or Google Play. --- - [All articles](https://klu.so/blog)