> ## Documentation Index > Fetch the complete documentation index at: https://docs.vistaly.com/llms.txt > Use this file to discover all available pages before exploring further. # Get interview insights > Retrieve the insights captured on an interview, together with the interview's metadata. Each insight is a highlighted passage: `source` says whether it was taken from the notes or from the transcript, `quote` is the highlighted passage itself (the words to quote), `text` is the same content formatted (for a recorded interview, one timestamped, speaker-attributed line per segment), `note` is the annotation the author gave it, and `cardIds` are the cards it is linked to. Insights on a recorded interview also carry `segments` (each covered segment with its speaker and timings) and a `transcriptRange`; when such an insight's excerpt can no longer be resolved it carries `excerptUnavailable: true` instead, with empty `text` and `quote`. AI suggestions the author has not accepted are not included. ## OpenAPI ````yaml https://api.vistaly.com/v1/swagger.json get /v1/interviews/{interviewId}/insights openapi: 3.1.0 info: contact: email: support@vistaly.com name: Vistaly Support Team description: >- Vistaly's public API for interacting with our platform, enabling developers to integrate with various features such as feedback, cards, and system health. termsOfService: https://vistaly.com/terms-of-service title: Vistaly API version: '2026-09-05' servers: - description: Production server url: https://api.vistaly.com security: [] tags: - description: System-related endpoints to interact with the overall service. name: system - description: >- Endpoints related to card management, including the creation and modification of card metrics and related data. externalDocs: description: Learn more about Cards url: https://docs.vistaly.com/en/card-types/overview name: cards - description: >- Endpoints for handling insights data, such as user feedback and interview records. externalDocs: description: Learn more about Insights url: https://docs.vistaly.com/en/insights/overview name: insights paths: /v1/interviews/{interviewId}/insights: get: tags: - interviews summary: Get interview insights description: >- Retrieve the insights captured on an interview, together with the interview's metadata. Each insight is a highlighted passage: `source` says whether it was taken from the notes or from the transcript, `quote` is the highlighted passage itself (the words to quote), `text` is the same content formatted (for a recorded interview, one timestamped, speaker-attributed line per segment), `note` is the annotation the author gave it, and `cardIds` are the cards it is linked to. Insights on a recorded interview also carry `segments` (each covered segment with its speaker and timings) and a `transcriptRange`; when such an insight's excerpt can no longer be resolved it carries `excerptUnavailable: true` instead, with empty `text` and `quote`. AI suggestions the author has not accepted are not included. parameters: - description: The unique identifier for the interview in: path name: interviewId required: true schema: type: string responses: '200': content: application/json: schema: $ref: '#/components/schemas/GetInterviewInsightsResponse' description: Interview insights retrieved successfully '400': description: Bad request content: application/json: example: message: Bad request schema: $ref: '#/components/schemas/MessageResponse' '401': description: Unauthorized content: application/json: example: message: Unauthorized schema: $ref: '#/components/schemas/MessageResponse' '403': description: Forbidden content: application/json: example: message: Forbidden schema: $ref: '#/components/schemas/MessageResponse' '500': description: Internal Server Error content: application/json: example: message: Internal server error schema: $ref: '#/components/schemas/MessageResponse' security: - bearerAuth: [] components: schemas: GetInterviewInsightsResponse: description: An interview with its insights. allOf: - $ref: '#/components/schemas/InterviewItem' - type: object properties: insights: description: >- The insights captured on the interview, notes insights first, then transcript insights, each in document order. items: $ref: '#/components/schemas/InterviewInsight' type: array required: - insights MessageResponse: additionalProperties: false properties: message: example: System provided message type: string type: object InterviewItem: additionalProperties: false description: A single interview. properties: id: description: The unique identifier for the interview. type: string capturedAt: description: ISO 8601 timestamp of when the interview was conducted. format: date-time type: string createdAt: description: ISO 8601 timestamp of when the interview record was created. format: date-time type: string source: description: Where the interview originated (e.g. "vistaly-api", "direct"). type: string sourceUrl: description: Optional link to where the interview originated. type: string textPreview: description: A short preview of the interview's summary or notes. May be empty. type: string cardIds: description: IDs of the cards this interview is linked to via insights. items: type: string type: array customerUserIds: description: IDs of the customer users who were interviewed. items: type: string type: array interviewerUserIds: description: IDs of the organization members who conducted the interview. items: type: string type: array insightCount: description: >- The number of accepted insights on the interview: what GET /v1/interviews/{interviewId}/insights returns. AI suggestions the author has not accepted are counted in `suggestedInsightCount` instead. Normally equal to the length of that endpoint's `insights` array, but it can exceed it in an edge case: the count is derived from every fragment of the interview document, while the endpoint reads highlights from the notes and transcript only. Do not assume the two are equal. type: integer transcriptType: description: >- How the transcript was captured: `media` for an uploaded recording that was transcribed, `text` for a pasted transcript. Absent when the interview has no transcript. enum: - media - text type: string processed: description: >- Whether the interview's processing — transcription and AI analysis — has finished. Note this differs from `FeedbackItem.processed`, which means the feedback was analyzed for insights. type: boolean appUrl: description: A link to view the interview in the Vistaly app. type: string interviewees: description: >- The interviewed customers, resolved to people. An id that appears in `customerUserIds` but not here could not be resolved: its record is gone, or the batched lookup came back short. Absence here is not proof the person was deleted. items: $ref: '#/components/schemas/InterviewInterviewee' type: array interviewers: description: >- The organization members who conducted the interview, resolved to people. An id that appears in `interviewerUserIds` but not here could not be resolved; see `interviewees`. items: $ref: '#/components/schemas/InterviewInterviewer' type: array suggestedInsightCount: description: >- The number of AI-suggested insights awaiting the author's decision. Not returned by the insights endpoint. type: integer required: - id - source - cardIds - customerUserIds - interviewerUserIds - interviewees - interviewers - insightCount - suggestedInsightCount - processed - appUrl type: object InterviewInsight: additionalProperties: false description: A single insight captured on an interview. properties: id: description: The unique identifier for the insight. type: string source: description: >- Whether the insight highlights the interview's notes or its transcript. enum: - notes - transcript type: string note: description: The annotation the author gave the insight, when they wrote one. type: string text: description: >- The highlighted content, formatted: the marked passage of the notes or text transcript verbatim, or — for a recorded interview — the covered transcript excerpt rendered one `[HH:MM:SS] Speaker (role): text` line per segment. Empty when `excerptUnavailable` is set. type: string cardIds: description: IDs of the cards the insight is linked to. items: type: string type: array createdAt: description: ISO 8601 timestamp of when the insight was created, when recorded. format: date-time type: string transcriptRange: $ref: '#/components/schemas/InterviewInsightTranscriptRange' quote: description: >- The highlighted passage alone, with no timestamps or speaker prefixes — the words to quote. For a recorded interview, the covered segments' text joined with newlines. Empty when `excerptUnavailable` is set. type: string segments: description: >- For insights on a recorded interview, the covered transcript segments in order, each with its resolved speaker. Absent for notes and text-transcript insights, and when `excerptUnavailable` is set. items: $ref: '#/components/schemas/InterviewInsightSegment' type: array excerptUnavailable: description: >- Present (and `true`) when a recorded interview's insight could not be resolved to its excerpt: the transcript could not be read, or its segment ids are no longer in it because the interview was re-transcribed. `text` and `quote` are empty then, which is not the same as an empty highlight — skip or flag the insight rather than treating it as empty. enum: - true type: boolean required: - id - source - text - quote - cardIds type: object InterviewInterviewee: additionalProperties: false description: A customer who was interviewed. properties: id: description: The customer user's ID. type: string name: description: 'Display name: full name, falling back to email, then id.' type: string email: description: The person's email, when recorded. type: string company: additionalProperties: false description: The company the person belongs to, when one is recorded. properties: id: description: The company's ID. type: string name: description: The company's name, when recorded. type: string website: description: The company's website, when recorded. type: string required: - id type: object required: - id - name type: object InterviewInterviewer: additionalProperties: false description: An organization member who conducted an interview. properties: id: description: The user's ID. type: string name: description: 'Display name: full name, falling back to email, then id.' type: string email: description: The user's email, when recorded. type: string required: - id - name type: object InterviewInsightTranscriptRange: additionalProperties: false description: Where in a recorded interview's transcript an insight sits. properties: startSegmentId: description: ID of the first transcript segment the insight covers. type: string endSegmentId: description: ID of the last transcript segment the insight covers. type: string startTime: description: Seconds into the recording at which the excerpt starts. type: number endTime: description: Seconds into the recording at which the excerpt ends. type: number required: - startSegmentId - endSegmentId type: object InterviewInsightSegment: additionalProperties: false description: One transcript segment covered by an insight on a recorded interview. properties: id: description: The transcript segment's ID. type: string speaker: description: >- The speaker: the mapped person's display name, or the raw transcript label when unmapped. type: string speakerRole: description: The speaker's role, when the label maps to a known person. enum: - interviewer - interviewee type: string startTime: description: Seconds into the recording at which the segment starts. type: number endTime: description: Seconds into the recording at which the segment ends. type: number text: description: The segment's text. type: string required: - id - speaker - text type: object securitySchemes: bearerAuth: scheme: bearer type: http ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.