generated: '2026-08-13'
method: derived
source: openapi/lytics-api-v2-openapi.json (components.schemas, 506 schemas) + openapi/lytics-api-v1-openapi.json;
  entity descriptions from https://docs.lytics.com/reference and vocabulary/lytics-vocabulary.yml
note: 'Entity-relationship graph for the Lytics CDP. relationships[] are DERIVED mechanically from $ref links between
  components.schemas in the v2 OpenAPI. id_reference_fields[] are derived from property-name frequency across all
  506 schemas and show how the estate is wired: segment_id / segment_ids appear in 162 schema properties combined,
  because virtually every integration job config in the API points back at an audience. Most of the remaining *_id
  fields (advertiser_id, list_id, ad_account_id, adwords_account_id) are THIRD-PARTY identifiers carried inside
  connector job configs, not Lytics entities — that distinction is recorded rather than flattened into the graph.'
core_entities:
- entity: Account
  id: account_id (aid)
  description: A Lytics account. account_id is the most common parameter in the v2 API (1,321 of 1,467 operations);
    one token can reach several accounts.
  paths:
  - /account
  - /account/{id}
  schema: models.AccountResp
- entity: User
  id: id | email | "me"
  description: A human user of the Lytics account, with roles and granular permissions.
  paths:
  - /user
  - /user/{id}
  - /user/{id}/roles
  schema: models.UserResp
- entity: Profile (Entity)
  id: _uid + identity fields
  description: The unified customer profile — the core object of the CDP. Looked up by any identity field/value
    on a table (usually `user`).
  paths:
  - /api/entity/{table}/{fieldname}/{fieldval}
  - /v2/identity/{table}/{field}/{value}
  schema: models.IdentityResponse
- entity: Segment (Audience)
  id: id | slug_name
  description: A behavioural audience defined in FilterQL. The hub of the model — segments are referenced by jobs,
    flows, ML models and content recommendations.
  paths:
  - /segment
  - /segment/{id}
  - /api/segment/{segId}/scan
  schema: models.Segment
- entity: Schema / Field
  id: table + field id
  description: 'Profile schema for a table: fields, mappings, identity config (idconfig) and field ranks.'
  paths:
  - /schema
  - /schema/{table}/field
  - /schema/{table}/idconfig
  - /schema/{table}/mapping
  schema: models.SchemaResponse
- entity: Stream
  id: stream name
  description: A named inbound data stream. Events are posted to /collect/json/{stream} or /collect/bulk/{stream}.
  paths:
  - /stream
  - /stream/{id}/stats
  - /collect/json/{stream}
  schema: models.StreamMetadata
- entity: Connection
  id: id
  description: A configured link to an external system, built on an Auth.
  paths:
  - /connection
  - /connection/{id}
  - /connection/{id}/schema
  schema: models.ConnectionResp
- entity: Auth
  id: id
  description: A stored credential for a provider. The v2 API exposes one path family PER auth type — 379 operations
    across ~126 provider-specific auth types.
  paths:
  - /auth
  - /auth/{type}
  - /auth/{type}/{id}
  schema: models.Auth
- entity: Job / Work
  id: id
  description: 'An import or export workflow instance. The largest surface in the API: 831 operations, one path
    family per integration workflow.'
  paths:
  - /job
  - /job/{id}
  - /job/{id}/pause
  - /job/{id}/resume
  - /job/{id}/kill
  - /job/{id}/logs
  schema: models.JobResponse
- entity: DataModel (Cloud Connect)
  id: id
  description: A warehouse-backed model (BigQuery, Snowflake, Databricks) synced into profiles, with a schedule,
    health and sync errors.
  paths:
  - /datamodel
  - /datamodel/{id}
  schema: models.DataModelResp
- entity: Flow
  id: id
  description: A multi-step campaign/journey with delays, conditional splits, A/B probabilities and export steps.
  paths:
  - /flow/ui
  - /flow/ui/{id}
  - /flow/state
  schema: models.Flow
- entity: ML Model
  id: id
  description: A lookalike / predictive model built on a segment, with a feature-importance and confusion-matrix
    summary.
  paths:
  - /ml
  - /ml/{id}/summary
  schema: models.MLModel
- entity: Template
  id: id
  description: A webhook payload transform, authored in Jsonnet (`jsonnet`) or JavaScript (`js1`).
  paths:
  - /template
  - /template/{id}
  - /template/{id}/test
  schema: models.TemplateTestRequest
- entity: Content Topic / Document
  id: topic label | doc id
  description: 'Content classification and affinity: crawled documents, topics, rollups, taxonomy and recommendations.'
  paths:
  - /api/content/doc
  - /api/content/topic
  - /api/content/recommend/user/{fieldname}/{fieldval}
  schema: null
- entity: Report
  id: id
  description: A saved report composed of components.
  paths:
  - /report
  - /report/{id}
  schema: models.Report
relationships:
- from: models.AccountResp
  to: models.EntitlementPlan
  kind: has_many
  via: entitlement_plans
- from: models.AccountResp
  to: models.RolePolicy
  kind: has_many
  via: roles_available
- from: models.AdhocTemplateTestRequest
  to: models.SampleSegmentsInput
  kind: has_one
  via: sample_segments
- from: models.AnomalyRuleResp
  to: models.RuleCheckConfigReq
  kind: has_many
  via: checks
- from: models.ConnectionCreateReq
  to: models.ConnectionConfig
  kind: has_one
  via: config
- from: models.ConnectionResp
  to: models.ConnectionConfig
  kind: has_one
  via: config
- from: models.ConnectionScanResp
  to: models.ResultSetColumn
  kind: has_many
  via: columns
- from: models.ContentOpportunityResp
  to: models.ContentOpportunityTopic
  kind: has_many
  via: topics
- from: models.ContentOpportunityTopic
  to: models.ContentOpportunityDimension
  kind: has_many
  via: dimensions
- from: models.DataField
  to: models.Column
  kind: has_many
  via: columns
- from: models.DataModelCreateReq
  to: models.DataModelJoin
  kind: has_one
  via: joined_to
- from: models.DataModelResp
  to: models.AnomalyRuleResp
  kind: has_one
  via: anomaly_rule
- from: models.DataModelResp
  to: models.DataModelHealth
  kind: has_one
  via: health
- from: models.DataModelResp
  to: models.DataModelJoin
  kind: has_one
  via: joined_to
- from: models.DataModelResp
  to: models.SyncSchedule
  kind: has_one
  via: schedule
- from: models.DataModelResp
  to: models.SyncError
  kind: has_many
  via: sync_errors
- from: models.Feature
  to: models.FieldPrevalence
  kind: has_one
  via: field_prevalence
- from: models.Feature
  to: models.Impact
  kind: has_one
  via: impact
- from: models.Field
  to: models.Assertions
  kind: has_one
  via: assertions
- from: models.Field
  to: models.IDConfig
  kind: has_one
  via: id_config
- from: models.Flow
  to: models.FlowStep
  kind: has_many
  via: steps
- from: models.FlowStep
  to: models.DelayUntil
  kind: has_one
  via: delay_until
- from: models.FlowStep
  to: models.SplitAffinityKV
  kind: has_many
  via: split_affinities
- from: models.FlowStep
  to: models.SplitConditionKV
  kind: has_many
  via: split_conditions
- from: models.FlowStep
  to: models.SplitProbKV
  kind: has_many
  via: split_probabilities
- from: models.IdentityFragment
  to: models.Kv
  kind: has_many
  via: keys
- from: models.IdentityFragment
  to: models.Kv
  kind: has_many
  via: neighbors
- from: models.IdentityResponse
  to: models.IdentityFragment
  kind: has_many
  via: fragments
- from: models.IdentityResponse
  to: models.Kv
  kind: has_many
  via: keys
- from: models.JobResponse
  to: models.AnomalyRuleResp
  kind: has_one
  via: anomaly_rule
- from: models.LQLUploadResponse
  to: models.Field
  kind: has_many
  via: fields
- from: models.LQLUploadResponse
  to: models.Mapping
  kind: has_many
  via: mappings
- from: models.MLModel
  to: models.ModelConfig
  kind: has_one
  via: config
- from: models.MLSummary
  to: models.ConfusionMatrix
  kind: has_one
  via: error_matrix
- from: models.MLSummary
  to: models.ModelMessage
  kind: has_many
  via: msgs
- from: models.MLSummaryResponse
  to: models.Feature
  kind: has_many
  via: features
- from: models.MLSummaryResponse
  to: models.MLSummary
  kind: has_one
  via: summary
- from: models.MetricResult
  to: models.MetricTimeSeriesEntry
  kind: has_many
  via: values
- from: models.Ranks
  to: models.IDConfig
  kind: has_many
  via: id_config
- from: models.Ranks
  to: models.RankList
  kind: has_one
  via: id_ranks
- from: models.Report
  to: models.Component
  kind: has_many
  via: components
- from: models.SchemaResponse
  to: models.Field
  kind: has_many
  via: fields
- from: models.SchemaResponse
  to: models.Mapping
  kind: has_many
  via: mappings
- from: models.SchemaResponse
  to: models.Ranks
  kind: has_one
  via: ranks
- from: models.SchemaResponse
  to: models.RetentionSettings
  kind: has_one
  via: retention_settings
- from: models.Segment
  to: models.Expr
  kind: has_one
  via: ast
- from: models.SegmentDeleteResp
  to: models.SegmentDeleteStatus
  kind: has_many
  via: segments
- from: models.StreamMetadata
  to: models.AnomalyRuleResp
  kind: has_one
  via: anomaly_rule
- from: models.TemplateTestRequest
  to: models.SampleSegmentsInput
  kind: has_one
  via: sample_segments
- from: models.UserResp
  to: models.UserAccount
  kind: has_many
  via: accounts
- from: models.VersionDiff
  to: models.SchemaResponse
  kind: has_one
  via: from
- from: models.VersionDiff
  to: models.SchemaResponse
  kind: has_one
  via: to
id_reference_fields:
- field: segment_id
  schema_properties: 88
- field: segment_ids
  schema_properties: 74
- field: id
  schema_properties: 28
- field: account_id
  schema_properties: 15
- field: advertiser_id
  schema_properties: 11
- field: list_id
  schema_properties: 11
- field: client_id
  schema_properties: 10
- field: app_id
  schema_properties: 8
- field: ad_account_id
  schema_properties: 7
- field: profile_id
  schema_properties: 6
- field: event_id
  schema_properties: 6
- field: user_id
  schema_properties: 6
- field: auth_ids
  schema_properties: 6
- field: organization_id
  schema_properties: 5
- field: provider_id
  schema_properties: 5
- field: adwords_account_id
  schema_properties: 5
- field: dimension_id
  schema_properties: 5
- field: author_id
  schema_properties: 5
- field: partner_id
  schema_properties: 4
- field: subject_id
  schema_properties: 4
- field: request_id
  schema_properties: 4
- field: site_id
  schema_properties: 4
- field: collection_id
  schema_properties: 4
- field: ads_account_id
  schema_properties: 4
- field: template_id
  schema_properties: 4
- field: parent_id
  schema_properties: 3
- field: adwords_customer_id
  schema_properties: 3
- field: adwords_list_id
  schema_properties: 3
- field: external_user_id
  schema_properties: 3
- field: braze_id
  schema_properties: 3
id_reference_note: segment_id/segment_ids dominate because nearly every export job config targets an audience. advertiser_id,
  list_id, ad_account_id, adwords_account_id, app_id, client_id, organization_id and partner_id are EXTERNAL platform
  identifiers held inside connector configs (Google Ads, Meta, Amazon DSP, Airship, etc.), not Lytics entity keys.
render: null