Hyveminds/Products/Quali
Product 03 · Stable · v1.9

Listening,
at archive
scale.

Read every interview, see every theme, without losing the participant’s words.

Quali codes transcripts the way a careful researcher would: theme suggestions you can accept, edit or reject; sentiment that respects context; codebooks that travel across projects. Built so you can read the field and stay close to the speaker.

02 · What it is

Coding that respects
the speaker.

Quali doesn’t replace the researcher. It does the slow work, reading every line, suggesting where codes might apply, and gives you back hours so you can focus on the patterns and the people.

For coding

Suggest, don’t impose.

Quali proposes codes; you accept, edit, reject or add your own. A working codebook emerges as you read, staying yours, not a black-box label. Disagree once, and the model adjusts; disagree systematically, and the suggestion stops.

For analysis

Themes across the corpus.

Co-occurrence, frequency, theme matrices, and sentiment trajectories, computed across the corpus and back-traceable to the exact line that triggered them. No statistic without the quote it came from.

For trust

Inter-rater on tap.

Built-in second-coder mode: blind another researcher (or another instance of you) to the same passages, compute Cohen’s κ, surface where you diverge. Reliability becomes a daily habit, not a publication scramble.

03 · A walk through

Three views of one corpus.

Twenty-eight interviews from the Per Capita Cost-of-Living qualitative arm, seen as a transcript, a code-by-participant matrix, and a triangulation panel.

quali.app/projects/cost-of-living-qual/transcript/p07-helen.vtt
P07 · Helen, 58 · Outer-suburban Sydney Interview 04/28 · 47:11
INT02:14

Could you tell me a bit about how the last six months have felt, money-wise?

P0702:21

Honestly? It’s been the kind of tight where you start counting things you didn’t use to count. Bread. The good kind, I mean. I’m back to the home-brand because the difference is two-fifty a week now. That sounds… small, doesn’t it. But two-fifty here, four-eighty there, and suddenly the week doesn’t close.

INT02:48

And in terms of who you talk to about that, family, friends, advisors?

P0702:53

Mostly nobody. You don’t want to be the one at coffee who says ‘I can’t’. My sister knows, but only in the abstract way. If I called the bank I’d sit on hold for forty minutes and end up speaking to someone reading from a script. That’s not a person to confide in.

INT03:19

How do you make decisions about where to cut, when something has to give?

P0703:24

By the end of the day my brain is fried. I make the worst decisions at 9pm because that’s when the bills come into focus, and that’s also when I’m the most tired. I don’t feel like I’m choosing, I feel like I’m reacting. Then in the morning I think, why did I do it that way?

INT03:48

Has there been anything that’s helped, even a little?

P0703:53

The library. There’s a woman there, Sandra, who just knows me now. She’ll see me come in and put aside the magazines I like. That’s small. But it’s the only thing I get for free that doesn’t feel like a handout.

Codebook5 active
cost38
trust24
time19
agency14
stigma11
+add code
Sentiment · this turn
POS 46%NEU 31%NEG 23%

Across the full transcript, P07 shifts negative when describing money decisions (-0.42 mean) and warm when describing community (+0.61 mean).

Themes · forming
Hidden austerity22/28
Decision fatigue18/28
Trust in strangers11/28
Quiet shame9/28
quali.app/projects/cost-of-living-qual/matrix/codes-by-participant
Codes × participants
28 PARTICIPANTS · 5 CODES · κ = 0.83 (BLINDED 2ND CODER)
P01P02P03P04P05P06P07 P08P09P10P11P12P13P14 P15P16P17P18P19P20P21 P22P23P24P25P26P27P28
cost 7381649 3560345 2751354 6258343
trust 2313054 1204302 5124031 2403123
time 3025325 4023024 3052432 0423523
agency 0210204 2020400 2023024 0200203
stigma 2022023 0202020 2202020 2022020
quali.app/projects/cost-of-living-qual/themes/hidden-austerity
Theme · 22 of 28 participants · 78% saturation
Hidden austerity, cuts that aren’t announced.

Participants describe small, private downgrades, the cheaper bread, the unrenewed magazine, the cancelled birthday plan, framed not as deprivation but as recalibration. The cuts are made invisibly, often before the financial pressure is acknowledged out loud.

P07 It’s the kind of tight where you start counting things you didn’t use to count. Bread. The good kind, I mean. I’m back to the home-brand because the difference is two-fifty a week now.”
P14 I cancelled my magazine subscription and didn’t tell my husband. I just didn’t want him worrying about something so small. He probably wouldn’t even notice for three months.”
P21 We didn’t do birthday this year, not really. I told her we’d do something special when the weather turned. The weather turned in October.”
P25 “You don’t make a decision. One day you just notice you’re not buying the things anymore.
Cohort agreement · κ = 0.86
Where the field converges.

Inter-rater agreement on this theme between Quali’s suggestions and the human second coder.

POS0.86
Adjacent themes
Decision fatigue0.74
Quiet shame0.62
Trust in strangers0.41
Theme proposed: 2026-02-18 · accepted by lead researcher 2026-02-19 · 4 quote exemplars confirmed · saturation reached at participant 19.
04 · Capabilities

What ships in the box.

Nine capabilities that together replace a stack of four legacy tools, and add an audit trail none of them had.

01 · Coding

Suggested in-line codes

Quali proposes codes as you read. Accept, edit, reject, or apply your own. Your codebook stays yours; the model adapts to your conventions.

02 · Sentiment

Context-aware affect

Sentiment respects discourse structure. “Honestly? It’s been tight” isn’t neutral; “not the worst” isn’t positive. Tuned on Australian English by default.

03 · Themes

Theme synthesis & saturation

Themes emerge from co-occurring codes; saturation tracking shows when adding interviews stops yielding new patterns. Reflexive thematic analysis (Braun & Clarke) by default.

04 · Reliability

Inter-rater on tap

Blind a second coder to the same passages. Cohen’s κ surfaces in real time; divergence drill-down shows exactly where you and they disagree.

05 · Matrix

Code × participant matrix

Frequency, presence-or-absence, and co-occurrence views. Filter by demographic; export to CSV, R, SPSS or Stata for downstream stats.

06 · Triangulate

Quote triangulation

Three–five exemplars per theme, surfaced automatically and back-traceable to the speaker. Useful for findings tables, briefings and policy submissions.

07 · Languages

31 languages

Coding and theme synthesis across English, Auslan-glossed, Mandarin, Cantonese, Vietnamese, Arabic, Greek and 24 others. Bilingual transcripts handled natively.

08 · Privacy

De-identification helper

Auto-flags names, places, employers, identifying details. Replaces with consistent pseudonyms across the corpus; the lookup stays encrypted and stays with you.

09 · Import

Audio, video, NVivo, Atlas

Import .vtt, .srt, .txt, .docx; or drop in audio/video and Quali transcribes via Phonos under the hood. Round-trip with NVivo .nvp and Atlas.ti .atlas12.

05 · Workflow

From a stack of interviews to findings.

How a 28-interview qualitative arm moves from raw .vtt files to a triangulated findings document.

01 · Import
i

Drop the corpus.

Drag .vtt, .srt, .docx, audio or video. Quali transcribes (via Phonos), de-identifies, and aligns speaker turns automatically.

02 · Code
ii

Read & code.

Quali proposes codes inline; you accept, edit or reject. The codebook adapts to your conventions in real time.

03 · Verify
iii

Second-coder pass.

Blind a colleague (or another instance of yourself) to the same passages. κ surfaces in real time; divergence is drillable.

04 · Synthesise
iv

Themes & saturation.

Themes emerge from co-occurring codes. Saturation tracking shows when the corpus stops surfacing new patterns.

05 · Hand off
v

Findings document.

Export a triangulated findings document, themes, exemplar quotes, agreement statistics, ready for the report or paper.