> ## Documentation Index > Fetch the complete documentation index at: https://docs.aitasker.co/llms.txt > Use this file to discover all available pages before exploring further. # Review Group Insights Aspect Diagnostics Route > Per-aspect rating impact and sub-theme rollups for one brand. Reads only columns the classification pass already wrote — no provider call and no model participates, so every figure here is counted. ## OpenAPI ````yaml /openapi.json post /api/v1/brands/{brand_id}/brand-intelligence/review-group-insights/aspect-diagnostics openapi: 3.1.0 info: title: AITasker version: 0.1.0 servers: [] security: [] paths: /api/v1/brands/{brand_id}/brand-intelligence/review-group-insights/aspect-diagnostics: post: tags: - Brand Reviews - Brand Reviews summary: Review Group Insights Aspect Diagnostics Route description: |- Per-aspect rating impact and sub-theme rollups for one brand. Reads only columns the classification pass already wrote — no provider call and no model participates, so every figure here is counted. operationId: >- review_group_insights_aspect_diagnostics_route_api_v1_brands__brand_id__brand_intelligence_review_group_insights_aspect_diagnostics_post parameters: - name: brand_id in: path required: true schema: type: string format: uuid title: Brand Id requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/ReviewAspectDiagnosticsRequest' responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/ReviewAspectDiagnostics' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' security: - HTTPBearer: [] components: schemas: ReviewAspectDiagnosticsRequest: properties: advertiser_id: type: string format: uuid title: Advertiser Id platform: type: string enum: - trustpilot - google title: Platform type: object required: - advertiser_id - platform title: ReviewAspectDiagnosticsRequest description: >- Per-aspect diagnostics for ONE brand on ONE platform. Platform is required for the same reason the matrix requires it: reply and aspect behaviour differ enough between Trustpilot and Google that a pooled figure describes neither. ReviewAspectDiagnostics: properties: advertiser_id: type: string format: uuid title: Advertiser Id advertiser_name: anyOf: - type: string - type: 'null' title: Advertiser Name platform: type: string enum: - trustpilot - google title: Platform classified_reviews: type: integer minimum: 0 title: Classified Reviews default: 0 examined_reviews: type: integer minimum: 0 title: Examined Reviews default: 0 sample_avg_rating: anyOf: - type: number maximum: 5 minimum: 1 - type: 'null' title: Sample Avg Rating min_mentions: type: integer minimum: 0 title: Min Mentions default: 0 sample_is_platform_total: type: boolean const: false title: Sample Is Platform Total default: false sample_basis_notice: type: string title: Sample Basis Notice default: >- Computed over AITasker's depth-capped sample of collected reviews, and within that only the reviews we classified — not the platform's full review population. Unclassified and uncollected reviews are excluded from every figure rather than counted as zero. aspects: items: $ref: '#/components/schemas/ReviewAspectRatingImpact' type: array title: Aspects type: object required: - advertiser_id - platform title: ReviewAspectDiagnostics description: The per-brand teardown layer under the cohort matrix. HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError ReviewAspectRatingImpact: properties: aspect: type: string maxLength: 100 minLength: 1 title: Aspect aspect_label: type: string maxLength: 100 minLength: 1 title: Aspect Label negative_reviews: type: integer minimum: 0 title: Negative Reviews default: 0 avg_rating_with_negative: anyOf: - type: number maximum: 5 minimum: 1 - type: 'null' title: Avg Rating With Negative avg_rating_without: anyOf: - type: number maximum: 5 minimum: 1 - type: 'null' title: Avg Rating Without max_sample_uplift: anyOf: - type: number maximum: 4 minimum: 0 - type: 'null' title: Max Sample Uplift below_min_mentions: type: boolean title: Below Min Mentions default: false sub_themes: items: $ref: '#/components/schemas/ReviewSubTheme' type: array maxItems: 8 title: Sub Themes type: object required: - aspect - aspect_label title: ReviewAspectRatingImpact description: >- What complaints about ONE aspect coincide with, in stars. **This is an association, not a counterfactual.** ``max_sample_uplift`` is the arithmetic identity "if every sampled review that complained about this aspect had instead rated at the average of the reviews that did not, the sample mean would become ``avg_rating_without``" — so the uplift is exactly ``avg_rating_without - sample_avg_rating``. It is an upper bound over the reviews we hold, NOT a prediction of what fixing the aspect would do: reviewers who complain about Delivery may also be unhappy for reasons this field cannot see, and the sample is depth-capped. Every figure is None rather than 0 when it could not be measured — a brand with no complaints about an aspect has no ``avg_rating_with_negative``, and rendering that as 0 stars would invert the finding. ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError ReviewSubTheme: properties: sub_theme: type: string maxLength: 60 minLength: 1 title: Sub Theme reviews: type: integer minimum: 1 title: Reviews negative: type: integer minimum: 0 title: Negative default: 0 positive: type: integer minimum: 0 title: Positive default: 0 type: object required: - sub_theme - reviews title: ReviewSubTheme description: >- One emergent sub-theme under an aspect — "the diagnostic tier". The aspect vocabulary is closed (the classifier is offered an enum), but ``sub_theme`` is free text the model wrote, so this is the only layer that can say *what about Delivery* customers are talking about. It is therefore also the only layer whose label was never validated against a vocabulary — it is shown as a grouping, never as a measured category. securitySchemes: HTTPBearer: type: http scheme: bearer ````