> ## Documentation Index > Fetch the complete documentation index at: https://docs.tied.com/llms.txt > Use this file to discover all available pages before exploring further. # Topic labeling health > Read-only labeling-health statistics through Current.account. Requires account membership, ReportPolicy view access and administrator membership. Topics need not be enabled: missing or disabled settings still returns HTTP 200 with actual statistics. Counts use current rank 1 topics; cleared and superseded assignments are excluded. Coverage and teams measure conversations; agents measure correction events; sources measure current labels. Fractions range 0–1 rounded to four decimals and are null with a zero denominator. Empty data has zero counts, empty agent/team arrays and null ratios/timestamps. UTC calendar weeks use the account weekday (default Monday); included inbox restrictions apply. ## OpenAPI ````yaml /api-reference/openapi.json get /api/v2/accounts/{account_id}/topics/labeling_health openapi: 3.0.4 info: title: Tied API description: >- The Tied REST API. Three families of endpoint, each with its own credential: **Application** for everything inside an account, **Platform** for provisioning accounts and users, and **Client** for acting as a contact from your own surface. version: 1.1.0 termsOfService: https://tied.com/legal contact: email: hello@tied.com license: name: MIT License url: https://opensource.org/licenses/MIT servers: - url: https://app.tied.com description: Tied cloud security: [] tags: - name: Account AgentBots description: Account-specific Agent Bots - name: Agents description: Agent management APIs - name: Canned Responses description: Pre-defined responses for common queries - name: Contacts description: Contact management APIs - name: Contact Labels description: Manage contact labels - name: Conversation Assignments description: Manage conversation assignments - name: Conversation Labels description: Manage conversation labels - name: Conversations description: Conversation management APIs - name: Custom Attributes description: Custom fields for contacts and conversations - name: Custom Filters description: Saved filters for conversations - name: Inboxes description: Communication channels setup - name: Integrations description: Third-party integrations - name: Messages description: Message management APIs - name: Profile description: User profile APIs - name: Reports description: Analytics and reporting APIs - name: Teams description: Team management APIs - name: Webhooks description: Event notification webhooks - name: Automation Rule description: Workflow automation rules - name: Help Center description: Knowledge base management - name: Contacts API description: Public contact APIs - name: Conversations API description: Public conversation APIs - name: Messages API description: Public message APIs - name: CSAT Survey Page description: Customer satisfaction survey - name: Account description: >- The authenticated user's own account — read and update account settings. Distinct from the Platform family's **Accounts**, which provisions accounts across an installation. - name: Audit Logs description: Read the account's audit trail - name: Inbox API description: Public inbox lookup by channel identifier - name: Labels description: >- Create and manage the account's labels. Applying one to a conversation or contact is a separate endpoint under those resources. - name: Macros description: >- Saved multi-step actions an agent — or an integration — can run against one or more conversations. - name: Campaigns description: Ongoing website campaigns and one-off SMS and WhatsApp sends. - name: Contact Notes description: Internal notes on a contact. Never visible to the contact. - name: Help Center Categories description: Categories group a portal's articles, one tree per allowed locale. - name: Help Center Articles description: The articles a portal publishes, and their draft and archived states. - name: Summary Reports description: Per-agent, per-team, per-inbox and per-label totals for a period, as JSON. - name: Live Reports description: Right-now queue counts. No date range — these describe the present. - name: CSAT description: >- The satisfaction ratings contacts have submitted, and the metrics computed from them. - name: Notifications description: >- The per-user notification feed, its read and snooze state, and the settings that decide what lands in it. - name: Conversation Participants description: >- The agents watching a conversation. Participants get notified about it without being assigned to it. - name: Draft Messages description: >- The unsent reply held for a conversation. Stored in Redis, one per conversation. - name: Search description: >- One search service across conversations, contacts, messages and Help Center articles. - name: Dashboard Apps description: Embedded pages that appear as a tab beside a conversation. - name: Assignment Policies description: How conversations are handed out, and which inbox uses which policy. - name: Files description: Storing a file so another endpoint can attach it. - name: Bot Flows description: >- Scripted conversation flows, their versions, and the publish and rollback that move between them. - name: Billing description: >- Credit balance, usage, invoices, spend controls and the provider-hosted payment flows. - name: AMB Templates description: >- Saved Apple Messages primitive payloads an agent can reuse from the composer. - name: AMB Onboarding description: The in-flight Apple Messages for Business registration for an account. - name: Widget Session description: Starting a widget session and reading the visitor it belongs to. - name: Widget Conversation description: The visitor's conversation and its messages. - name: Widget Context description: Campaigns, agents, labels, events and uploads for a widget session. - name: CSAT Survey description: The public, unauthenticated survey a contact answers from a link. - name: Workspace description: Scheduling and cancelling deletion of the whole workspace. - name: Widget Integrations description: Provider actions available to a visitor session in its own conversation. - name: Orchestration Studio description: >- Administrator-only read APIs for Orchestration Studio cases and knowledge sources. - name: Topic Analytics description: Account Topics readiness, labeling health and usage aggregates. - name: Analytics V2 description: >- Account-scoped metrics, anonymized benchmarks and audited conversation drilldown. paths: /api/v2/accounts/{account_id}/topics/labeling_health: parameters: - $ref: '#/components/parameters/account_id' get: tags: - Topic Analytics summary: Topic labeling health description: >- Read-only labeling-health statistics through Current.account. Requires account membership, ReportPolicy view access and administrator membership. Topics need not be enabled: missing or disabled settings still returns HTTP 200 with actual statistics. Counts use current rank 1 topics; cleared and superseded assignments are excluded. Coverage and teams measure conversations; agents measure correction events; sources measure current labels. Fractions range 0–1 rounded to four decimals and are null with a zero denominator. Empty data has zero counts, empty agent/team arrays and null ratios/timestamps. UTC calendar weeks use the account weekday (default Monday); included inbox restrictions apply. operationId: get-topic-labeling-health parameters: - name: weeks in: query description: Coverage and correction window in calendar weeks. schema: type: integer minimum: 1 maximum: 26 default: 12 - name: backlog_days in: query description: Unlabeled age window in days. schema: type: integer minimum: 1 maximum: 365 default: 90 responses: '200': description: Actual synthetic request response; see topic analytics evidence. content: application/json: schema: $ref: '#/components/schemas/topic_labeling_health' example: window: weeks: 2 from: '2026-09-28T00:00:00Z' to: '2026-10-07T12:00:00Z' weekStart: monday backlogDays: 90 conversations: 2 lowVolume: true minimumConversations: 30 coverage: weeks: - weekStart: '2026-09-28' conversations: 1 labeled: 0 coverage: 0 - weekStart: '2026-10-05' conversations: 1 labeled: 1 coverage: 1 conversations: 2 labeled: 1 coverage: 0.5 lowVolume: true minimumConversations: 30 backlog: days: 90 unlabeled: 1 buckets: zeroToOneDay: 0 twoToSevenDays: 1 eightToThirtyDays: 0 overThirtyDays: 0 oldestUnlabeledAt: '2026-10-04T12:00:00Z' agents: - agentId: 1 name: Demo Agent labeled: 1 cleared: 0 lastActiveAt: '2026-10-07T12:00:00Z' teams: - teamId: 1 name: Demo Support conversations: 2 labeled: 1 coverage: 0.5 sources: labeled: 1 counts: person: 1 rule: 0 ai: 0 shares: person: 1 rule: 0 ai: 0 lowVolume: true minimumConversations: 30 lowLabels: true minimumLabeled: 50 x-example-provenance: source: spec/contract/topic_analytics_spec.rb capture: >- Actual synthetic Rails integration response; full Content-Type and scrub details in docs/platform-api/topic-analytics-evidence/responses.json. scrubbed: true '400': description: >- The request presented both a Devise Token Auth session credential and an api_access_token. The request is refused before the action runs. content: application/json: schema: type: object required: - error properties: error: type: object required: - code properties: code: type: string enum: - multiple_credentials additionalProperties: false additionalProperties: false example: error: code: multiple_credentials '401': description: >- Unauthenticated or account is not accessible to the caller. Devise authentication responses may use an `errors` array; account access denials use `error`. content: application/json: schema: oneOf: - $ref: '#/components/schemas/error_message' - type: object required: - errors properties: errors: type: array items: type: string '422': description: Invalid period, window, grouping, metric, pagination or version. content: application/json: schema: $ref: '#/components/schemas/error_message' security: - userApiKey: [] components: parameters: account_id: in: path name: id schema: type: integer required: true description: The numeric ID of the account schemas: topic_labeling_health: type: object description: All five health panels; does not require Topics to be enabled. required: - window - coverage - backlog - agents - teams - sources properties: window: type: object description: Effective coverage and backlog windows. required: - weeks - from - to - weekStart - backlogDays - conversations - lowVolume - minimumConversations properties: weeks: type: integer description: Calendar week count. minimum: 1 maximum: 26 from: type: string description: Start of oldest calendar week in UTC. format: date-time to: type: string description: Current request time in UTC. format: date-time weekStart: type: string description: Configured weekday, default monday. enum: - sunday - monday - tuesday - wednesday - thursday - friday - saturday backlogDays: type: integer description: Backlog window in days. minimum: 1 maximum: 365 conversations: type: integer description: Conversations in coverage window. minimum: 0 lowVolume: type: boolean description: >- True below minimumConversations; counts and ratios are still returned. minimumConversations: type: integer description: Conversation warning threshold, currently 30. minimum: 0 coverage: $ref: '#/components/schemas/topic_health_coverage' backlog: $ref: '#/components/schemas/topic_health_backlog' agents: $ref: '#/components/schemas/topic_health_agents' teams: $ref: '#/components/schemas/topic_health_teams' sources: $ref: '#/components/schemas/topic_health_sources' error_message: type: object description: >- The shape returned by `render_unauthorized`, `render_not_found_error` and `render_could_not_create_error`. properties: error: type: string description: A single human-readable message. example: Resource could not be found required: - error topic_health_coverage: type: object description: Weekly and total main-topic coverage. required: - weeks - conversations - labeled - coverage - lowVolume - minimumConversations properties: weeks: type: array description: Every week in the window, oldest first. items: type: object description: One calendar week. required: - weekStart - conversations - labeled - coverage properties: weekStart: type: string description: UTC calendar week start date. format: date conversations: type: integer description: Conversations created in the window. minimum: 0 labeled: type: integer description: Conversations with a current, non-superseded rank 1 topic. minimum: 0 coverage: type: number description: >- Labeled / conversations rounded to four decimals; null with zero conversations. nullable: true minimum: 0 maximum: 1 conversations: type: integer description: Conversations created in the window. minimum: 0 labeled: type: integer description: Conversations with a current, non-superseded rank 1 topic. minimum: 0 coverage: type: number description: >- Labeled / conversations rounded to four decimals; null with zero conversations. nullable: true minimum: 0 maximum: 1 lowVolume: type: boolean description: >- True below minimumConversations; counts and ratios are still returned. minimumConversations: type: integer description: Conversation warning threshold, currently 30. minimum: 0 topic_health_backlog: type: object description: Unlabeled conversations in the backlog window. required: - days - unlabeled - buckets - oldestUnlabeledAt properties: days: type: integer description: Backlog age window in days. minimum: 1 maximum: 365 unlabeled: type: integer description: Conversations without a current main topic. minimum: 0 buckets: type: object description: Age buckets measured since conversation creation. required: - zeroToOneDay - twoToSevenDays - eightToThirtyDays - overThirtyDays properties: zeroToOneDay: type: integer description: Age 0–1 whole days. minimum: 0 twoToSevenDays: type: integer description: Age 2–7 whole days. minimum: 0 eightToThirtyDays: type: integer description: Age 8–30 whole days. minimum: 0 overThirtyDays: type: integer description: Age above 30 days, bounded by days. minimum: 0 oldestUnlabeledAt: type: string description: Oldest unlabeled creation time, or null. format: date-time nullable: true topic_health_agents: type: array description: At most 100 people, ordered by set count descending then user ID. items: type: object description: Human correction activity; rules and AI are excluded. required: - agentId - name - labeled - cleared - lastActiveAt properties: agentId: type: integer description: ID of the person who corrected labels. name: type: string description: Current account member display name; null if no longer a member. nullable: true labeled: type: integer description: >- Set correction events, including repeated sets on one conversation. minimum: 0 cleared: type: integer description: Clear correction events. minimum: 0 lastActiveAt: type: string description: Most recent set or clear timestamp in UTC. format: date-time maxItems: 100 topic_health_teams: type: array description: >- Top 100 teams by conversation count, then no-team, then optional overflow; totals include all conversations. items: type: object description: Team coverage; otherTeams is optional. required: - teamId - name - conversations - labeled - coverage properties: teamId: type: integer description: Team ID; null for no-team and otherTeams rows. nullable: true name: type: string description: Current team name; null for no-team and otherTeams rows. nullable: true conversations: type: integer description: Conversations created in the window. minimum: 0 labeled: type: integer description: Conversations with a current, non-superseded rank 1 topic. minimum: 0 coverage: type: number description: >- Labeled / conversations rounded to four decimals; null with zero conversations. nullable: true minimum: 0 maximum: 1 otherTeams: type: integer description: >- Number of teams merged into the final overflow row; present only on that row. minimum: 0 maxItems: 102 topic_health_sources: type: object description: Who assigned the current main topics. required: - labeled - counts - shares - lowVolume - minimumConversations - lowLabels - minimumLabeled properties: labeled: type: integer description: Current main topic count in the coverage window. minimum: 0 counts: type: object description: Source counts. required: - person - rule - ai properties: person: type: integer description: Current main topics assigned by person. minimum: 0 rule: type: integer description: Current main topics assigned by rule. minimum: 0 ai: type: integer description: Current main topics assigned by ai. minimum: 0 shares: type: object description: Source fractions. required: - person - rule - ai properties: person: type: number description: Source count / labeled, four decimals; null with no labels. nullable: true minimum: 0 maximum: 1 rule: type: number description: Source count / labeled, four decimals; null with no labels. nullable: true minimum: 0 maximum: 1 ai: type: number description: Source count / labeled, four decimals; null with no labels. nullable: true minimum: 0 maximum: 1 lowVolume: type: boolean description: >- True below minimumConversations; counts and ratios are still returned. minimumConversations: type: integer description: Conversation warning threshold, currently 30. minimum: 0 lowLabels: type: boolean description: True below minimumLabeled; returned shares remain populated. minimumLabeled: type: integer description: Label warning threshold, currently 50. minimum: 0 securitySchemes: userApiKey: type: apiKey in: header name: api_access_token description: >- This token can be obtained by visiting the profile page or via rails console. Provides access to endpoints based on the user permissions levels. This token can be saved by an external system when user is created via API, to perform activities on behalf of the user. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.