> For the complete documentation index, see [llms.txt](https://evidano.gitbook.io/ailyze-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://evidano.gitbook.io/ailyze-docs/common-research-workflows/analyze-interview-transcripts.md). # Analyze Interview Transcripts A step-by-step guide to turning your interview transcripts into clear themes, quotes, and charts you can share. This page shows researchers how to turn interview transcripts into a structured report of [themes](/ailyze-docs/reference/glossary.md#theme), answers to research questions, quotes, and charts. Use it for any interviews or long-form text you want organized. Every control on these screens is listed in [Document Analysis Settings](/ailyze-docs/reference/settings/document-analysis.md). **When to use this** instead of the spreadsheet workflow: your data is in documents, not rows in a sheet. For rows, use [Analyze Open-Ended Survey Responses](/ailyze-docs/common-research-workflows/analyze-open-ended-survey-responses.md). The examples use the guide's invented sample: three staff interviews and a parents' focus group evaluating a mobile health clinic. Download them to follow along: {% file src="/files/6LrqKuFHOzp7JvBDC5UC" %} Sample interview 1: a community health worker, North Valley (invented) {% endfile %} {% file src="/files/strsb0lwSD9ke569Sse4" %} Sample interview 2: a clinic manager, South Ridge (invented) {% endfile %} {% file src="/files/t7GvfggiFtDFd5gCwBoy" %} Sample interview 3: a village health volunteer, North Valley (invented) {% endfile %} {% file src="/files/yEKie6SIa3nZWhcL9Lus" %} Sample focus group: four parents, North Valley (invented) {% endfile %} ## Before you start * A Free or Pro plan, and your interviews as DOCX, PDF, or TXT, one in each file. Clear speaker labels help. * Only have audio or video? [Transcribe it first](/ailyze-docs/prepare-research-data/transcribe-audio-and-video.md), then click **Analyze transcript** in the editor. * Optional: a proposal, interview guide, or codebook to give the AI context. ## Step 1: Pick the transcripts 1. From your account page, on the **Document analysis** card, click **New document project**. 2. Click **Upload files** and add your transcripts, or tick files you uploaded before. You should see each file listed under **Files to analyze**. Leave **Focus groups:** on **Auto-detect**. ## Step 2: Give the AI some context Context helps the AI fit the draft to your project. Under **Supporting files**: * Drop a proposal, interview guide, terms of reference, or codebook. An attached codebook is imported word for word as your themes. A ticked file moves there with its arrow (**Use as context instead**). * Or type a short brief under **Input manually**, for example: "Evaluate the mobile clinic programme in North Valley and South Ridge. I need the main themes, barriers to using the clinic, and a comparison by district." Click **Suggest a method from my files and notes**, or **Continue**; either way the AI suggests a methodology for you to keep or change.
Four sample transcripts under Files to analyze, a typed brief, the method suggestion button and Continue

Remove a file here with its x; it stays in your library.

You should see **Methodology**, opened on the AI's pick under **Suggested for your files**, with its reason. ## Step 3: Choose a methodology Keep the suggestion, or click **See other methods** and pick your approach, for example under **Browse all methods by field** → **Academia** → **Thematic and content approaches**, or search. The info icon explains each method. Click **Next**, or **No specific method** to let the AI design the table.
Methodology step opened on the AI's pick, Framework analysis, with its reason, See other methods and Next

The methodology becomes the AI's starting point for the table it drafts.

## Step 4: Confirm the focus-group participants If any file is a focus group, the **Focus groups** step lists the files the AI detected and the participants it found. Untick files that are one-to-one interviews, choose "Yes, one row per participant" to count opinions per person, fix any names, and click **Continue to output configuration**. See [Analyze Focus Groups](/ailyze-docs/common-research-workflows/analyze-focus-groups.md). You should then see **Drafting your project** work through its stages, usually in under a minute.
Drafting your project dialog with five progress stages

Nothing is created at this point; you review everything next.

## Step 5: Review the draft The draft opens with **You will receive a report**, the unit of analysis, **You will receive themes generated from the entire dataset**, and **You will receive a table of answers and metadata**. Work through it: 1. Read **Why this draft:** to see how the AI read your material. 2. Under **What is your unit of analysis?**, check **Each row is one** (usually a document; a participant for focus groups; "document or focus-group participant" when you mix them). 3. Make sure the themes section has themes. If it shows **Configuration required**, choose **Generate draft themes and sub-themes now**, or **Add manually** to bring your own codebook. **Edit themes and sub-themes** opens them for editing. 4. In the table of answers and metadata, click each column to tighten its description, and remove columns you will not use. 5. Add research questions with **Add** › **Add question**.
Themes section with numbered draft themes and sub-themes and Edit themes and sub-themes

Rename, merge, or delete draft themes now; it is quicker than editing after the run.

Rows, columns, and types are explained in [Design Your Analysis Table](/ailyze-docs/analyze-data/design-your-analysis-table.md). ## Step 6: Set up comparisons and the report These options are in **You will receive a report**, at the top of the draft. 1. To compare groups (district, role, site), keep **Compare groups (e.g., by gender, year)** ticked. The AI tags each file as the draft opens; click **Review** in its message to check, or the pencil beside the option for files still untagged. See [Compare Groups (Segments)](/ailyze-docs/analyze-data/compare-groups-segments.md). 2. Set **Report in** for a report in a language other than English. 3. When your themes answer your research questions, tick **Research questions mapped to themes**.
Report options with summaries, group comparison, question mapping, report language, word cloud and the standard outputs

The outputs every report has are folded under one line; the rest are your choice.

## Step 7: Start the analysis Click **Review and start analysis**. If anything is missing, a message names it and highlights the place at fault. Otherwise a summary of what the AI will do with every document opens; type a **Project name** if you like, click **Confirm configuration and start analysis**, then **View project**.
Create project dialog with the Project name field, the rows, columns, themes and report, Keep editing and Confirm

Read the summary as a checklist; Keep editing goes back to the draft.

{% hint style="warning" %} **Costs money:** On a paid plan, each file beyond the included count is charged; the summary states the cost, and **Confirm charge** asks before the project is created. The Free plan unlocks only the first theme and question in the results. See [Plans, Limits & Credits](/ailyze-docs/reference/plans-limits-and-credits.md). {% endhint %} You should see the project open, marked in progress. You get an email when it is done. ## Step 8: Read your results The results page, from top to bottom (**Contents** on the left jumps to any section): 1. **Header**: **Back to projects**, the name, chips for the documents and the completion date, and **View changes**. 2. The action row: **Export**, **Ask AI**, **Share**, and **Edit project**. 3. **Executive Summary**, then the table of all rows (for example **All documents**) with a column per theme and question. 4. **Codebook**: themes and codes with counts and charts. 5. **Themes** and **Questions**: one block each, with a summary across files, quotes grouped by code, a comparison across groups, and a summary of each file. Each cell opens on its summary and first quote; **Show quotes (N)** lists the rest. 6. When each row is something the documents describe, such as an outcome: a section named after the unit (for example **Outcomes**). Its counts are units, not documents.
Results page with the Codebook and its themes, codes and counts

Start with the codebook to see which themes came up most, then open a theme for its quotes.

## Step 9: Do more with your results | Button | What it does | Plan | | ----------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------ | ----------- | | **Ask AI** | Ask questions; request summaries or charts. See [Chat With Your Data](/ailyze-docs/analyze-data/chat-with-your-data.md). | Pro to type | | **Export** | **Word report (DOCX, ZIP)** or **Excel workbooks (XLSX, ZIP)**. | Pro | | **Edit project** › **Add file** | Add transcripts and re-run the parts that change. | All plans | | **Edit project** › **Add theme**, **Add question**, **Delete files** | Extend or trim the analysis. | Pro | | **Edit** and ⋯ › **Calculate AI accuracy** (in the **Codebook** header) | Rename, merge, split, or move codes; check coding reliability. See [Build a Codebook](/ailyze-docs/analyze-data/build-a-codebook.md). | Pro | | **Share** | Invite a colleague by email, or create a view-only link. See [Export & Share Findings](/ailyze-docs/create-deliverables/export-and-share-findings.md). | Pro |
Optional: a worked example * Files: the three sample staff interviews and the parents' focus group. * Methodology: **Framework analysis**, as the AI suggested. * Brief: "Evaluate the mobile clinic programme… barriers to using the clinic, and a comparison by district." * Draft: one row per document or focus-group participant, a theme hierarchy, questions on relevance, delivery, and barriers, and segments for district and respondent type. * Result: themes such as "Clinic close to home replaces lost days" and "Costs borne by families and volunteers", each with quotes, and a comparison of North Valley and South Ridge.
## Next steps * [How to Read Your Results](/ailyze-docs/analyze-data/how-to-read-your-results.md) — a full tour of every section and chart. * [Build a Codebook](/ailyze-docs/analyze-data/build-a-codebook.md) — refine your themes and codes. * [Generate a Research Report](/ailyze-docs/create-deliverables/generate-a-research-report.md) — turn your results into a shareable deliverable.