This week is fieldwork at a mid-size design studio — five semi-structured interviews across a team, each person talking about their own workflow and how decisions get made in the space. The research is for a dissertation on collaboration in design practice. All five interviews are being recorded and transcribed on the Mac; each transcript is searchable; and nothing leaves the machine until it is time to analyze, which means IRB requirements stay simple and participant confidentiality stays in the researcher's hands rather than a server's.
Monday 10:00am — before the first interview, the Mac is ready
The Mac is already recording. Open the video call with the first participant, press record, and two audio tracks begin: the microphone in the researcher's office and the participant's audio coming through the screen. The participant sees no new person in the call, gets no permission prompt, and has no idea the recording began until the researcher tells them it started — which happens exactly when the informed-consent conversation ends. By 10:00, the track is rolling and the participant is giving informed consent.
10:00am to 10:40am — the first interview: a designer on the daily workflow
The first participant describes their own process — what they do first when they arrive at the office, what they do before a design critique, what they do when a decision has to be made across the team. They talk about tools they reach for, moments when the team's tools break down, and a few times when they wish the workflow was different. The interview transcript is being built in real time on the Mac.
- 00:14:22ParticipantWe have three different Slack channels, and I never know which one gets checked first.
- 00:14:31ResearcherWalk me through what happens when you need a decision from the lead.
- 00:14:40ParticipantDepends on the time of day, honestly.
10:45am — reading what the first participant said before talking to the second
The first interview has ended, and the Mac has already transcribed most of it. The researcher reviews the summary the Mac produced: the key points this participant raised, the moments they mentioned tools, the specific workflows they described. Before moving to the second interview at 11:00, the researcher has read the summary and already knows what the first person said about decision-making.
11:00am to 11:45am — the second interview: a different role, a different perspective
The second participant does a different job at the studio — part design, part project management — and describes their own workflow. They mention some of the same tools the first participant talked about, but from a different angle: the project manager sees the tool as a blocker; the designer sees it as a starting point. Both descriptions are now transcribed on the Mac in two separate files.
Noon to 1:00pm — comparing how two roles describe the same moment
Between the second and third interviews, the researcher searches both transcripts for the word "critique." The first participant mentioned critique as a moment of uncertainty. The second participant mentioned critique as a place to align. Same practice, two different experiences. The researcher makes a note: this split in how roles experience the same moment is probably important. That note gets written into the running research log on the Mac; both transcripts are available to quote from if the analysis needs them later.
1:00pm to 1:35pm — the third interview: the team lead's perspective
The third participant is the design lead — the person the first two participants said they check with for decisions. The lead describes how they try to see what both individual contributors are saying, how they manage when the team's priorities disagree, and what they wish the team communication tools could do. The Mac has now transcribed three interviews; all three are searchable by phrase on the Mac.
Tuesday 9:00am — reading across three days' interviews before the fourth one
The researcher reviews what all three interviews have produced: a summary for each, and three full transcripts on the Mac. The pattern from the first day becomes clearer: every participant mentioned the same collaboration moment, but each person described it with different frustrations. The researcher searches all three transcripts for the phrase "decision-making" and finds it used three times across the three interviews, each in a different context. Before the fourth interview, the researcher has a clearer picture of where the team agrees and where the roles diverge.
Tuesday 10:00am to 10:50am — the fourth interview: an admin perspective
The fourth participant handles operations and legal — the parts of the studio that sit outside design but enable it to happen. They describe their own workflow and also describe, from the outside, how they see the design team's process breaking down. The fourth transcript lands on the Mac. The researcher can now search across four participants' accounts of the same organization.
Wednesday 2:00pm — the fifth interview, and what it will complete
The researcher has scheduled the fifth interview for Wednesday afternoon — the last participant, who works part-time as a researcher and part-time as a designer. By now, four transcripts are searchable on the Mac; the fifth will complete the set. The research has moved from individual interviews to seeing patterns across interviews.
Wednesday 2:00pm to 2:45pm — the fifth interview: a practitioner-researcher's view
The fifth participant brings both the designer's perspective and the researcher's eye for the process itself. They describe their own workflow and also reflect on what they see when they step back from the work. The fifth transcript completes the set.
- 00:18:50ParticipantWhat I notice is that we are all describing collaboration, but we are all experiencing isolation.
- 00:18:58ResearcherSay more about that.
- 00:19:05ParticipantWe want the same thing but do not have a way to say it at the same time.
Thursday — analyzing across five interviews, on the Mac
The researcher's work now shifts from fieldwork to analysis. On the Mac are five interview transcripts, each searchable by phrase, speaker and timestamp. The research question was about how collaboration decisions happen in design teams. All five interviews touched that question from different angles.
The researcher opens a spreadsheet and begins coding: going through all five transcripts and marking the moments when each participant talked about a decision. The Mac's search function finds every mention of the keyword "decision" across all five interviews, and the researcher can see, at a glance, that the lead participant talked about decision-making eight times, the two individual contributors three times each, and the administrator twice. The practitioner-researcher talked about it five times, often reflecting on the team's pattern rather than their own moment. But beyond the frequency lies the substance: the lead describes decisions as moments of alignment; the administrators describe them as moments of constraint; the individual contributors describe them as moments of uncertainty. Same practice, three different emotional valences. Those differences would be nearly invisible if the researcher were working from memory or handwritten notes during fieldwork. On the Mac, searchable, they are immediate and concrete. The pattern that surprised the researcher — that roles perceive the same moment differently — is now supported by five quotable instances across five interviews, each with a timestamp.
Friday — member checking, from the Mac
The researcher wants to fact-check an analysis with one of the participants. They export just that participant's transcript from the Mac — the fifth interview — and send it to the participant to review. The export shows what the participant said; the researcher is ready to ask "Is this accurate?" or "Can I quote you on this?" without sharing the other four participants' interviews. All five recordings stay on the Mac; only the one transcript leaves it, and only with the researcher's choice.
Saturday — writing the first analysis memo
The researcher writes a three-page memo to their advisor: here is what all five participants said about decision-making, here is how those accounts differ by role, here is the pattern that surprised me, here is what I need to understand better in the next round of fieldwork. Every claim in the memo is backed by a searchable transcript on the Mac. The advisor reads the memo and asks: "In the second interview, did they really say the project manager never sees the studio's calendar?" The researcher searches the second transcript for "calendar" and finds the exact sentence, the timestamp, and pastes it into a message back to the advisor. No voicemail backs up the interview; no memory reconstructs what was said.
What this workflow keeps under the researcher's control
The researcher, not a service, decides when interviews get moved off the Mac. The researcher decides who sees what — a single participant can review their own interview without seeing anyone else's data. The researcher stays complicit with their IRB protocol because nothing leaves the Mac without an explicit decision to export. Five interviews on the Mac are five data-security boundaries the researcher maintains rather than five uploaded files in a system someone else administers.
What this workflow is not
This is not a platform for sharing research across a lab or for collaborative coding with advisors — nothing here syncs across researchers' machines or puts interviews on a shared workspace. This is not a team transcription service; it works for individual researchers on their own Macs. This is not a tool for participants to access their own interviews; that is done by the researcher exporting and sharing the file outside the tool. Dissertation research that happens on one machine, analyzed by one researcher and one advisor, fits this workflow. Larger projects or lab-based research with shared coding across a dozen graduate students, or research that needs to be archived institutionally for a decade, has a different set of needs this tool was not built for.
A week like this one — five interviews at one site — is common in dissertation fieldwork. Research unfolds faster than tools can keep up. A researcher arrives with a question, not an archive. The first interview answers something unexpected; the second clarifies it; by the fifth, the pattern is clear. Having transcripts searchable on site, rather than after you leave, means research can tighten while gathering data still matters.
Why recordings stay on the researcher's machine until they move them
A participant consents to be recorded in a specific context — a one-hour interview about their workflow — and the recordings that context produces belong to that specific relationship. Automatic uploads to a university server, a cloud service, or an archive changes the boundaries of consent retroactively, even if the terms said consent was broad. A researcher who keeps recordings on their Mac until they have a plan for them — a plan that may be "export to a secure server," "transcribe and delete the audio," or "keep indefinitely" — stays in control of the shape that consent takes.
Fieldwork often happens in contexts where institutional approval moves slowly. A researcher might have IRB approval to record interviews but not yet have secured a secure server or determined which data belongs to which project. A Mac that keeps recordings until the researcher is ready is the only way to honor the boundary between recording for research and uploading without knowing where or for how long.
Five interviews on a Mac are five opportunities to notice you misunderstood. If the fourth interview makes a participant's meaning from the second interview clear, the researcher can revisit and re-listen with fresh ears before writing. Once uploaded, that clarity is harder to find, and your ability to access it depends on systems you do not control.
What it costs, what it runs on
Coii AudioNotes is $19, paid once, covering three of a researcher's own Macs, with a 30-day trial needing no card and no account. It runs on macOS 13 Ventura or later, Apple Silicon or Intel — the same setup used for all five of this week's interviews. The full case for academic researchers covers the broader argument, and interview transcription on a Mac walks through the specific setup. For the particular constraint of fieldwork that needs to stay local, best private meeting recorders covers the wider alternatives, and how to keep recordings confidential walks through the mechanics of export and data minimization. To find what was said in past interviews, searching old meetings and writing minutes from a transcript both walk through the research techniques this page showed in practice. A journalist's interview workflow and a consultant's workflow both face similar back-to-back-interview days but use the same recording setup.