generated: '2026-08-17'
method: derived
source: https://nap.launchmetrics.com/docs
note: >-
  Derived by reading the example request/response bodies published in the NAP reference. Launchmetrics
  publishes no OpenAPI, so there are no components.schemas or $ref links to walk — every entity, field
  and relationship below was read off a worked example in the HTML docs and is therefore a partial
  view of the real model, not a complete schema. Field types are omitted where the docs do not state
  them; nothing is guessed.
entities:
- name: document
  description: >-
    A publication — an article, post, video or other indexed piece of coverage. The central entity of
    the Launchmetrics data model and the unit returned by both Search and Documents.
  identifier: numeric string id (example ids are 20-digit values such as 14973462950408689474)
  retrieved_by:
  - GET /documents/{ids} (Documents v1, Documents v2)
  - GET|POST /documents (Search)
  observed_fields:
  - provider
  - date
  - content
  - title
  - summary
  - link
  - url
  - image
  - original_image
  - video_url
  - language
  - country_code
  - location
  - place
  - coords
  - keywords
  - type_publication
  - id_from_provider
  - date_from_provider
  - feeling
  - cluster
  - social
  - author
  - influencer_ids
  representations:
    parameter: response_type
    known_values:
    - analytics
    note: >-
      The reference demonstrates response_type=analytics, which returns a flattened analytics-shaped
      document. Other accepted values are referenced but not enumerated in the docs.
- name: author
  description: >-
    The account or byline behind a document. Embedded inside document rather than addressable on its
    own endpoint.
  observed_fields:
  - id
  - name
  - url
  - age
  - contact_count
  - subscriber_count
  - subscription_count
  - video_watch_count
  - favourites_count
  - tweets_count
- name: social_metrics
  description: Engagement counters attached to a document.
  observed_fields:
  - likes
  - shares
  - replies
  - retweet_to_user
  - user_mentions
- name: media
  description: >-
    A media outlet — the publication vehicle rather than the piece of coverage. Split into online and
    print corpora with separate search endpoints.
  identifier: media id
  retrieved_by:
  - GET /medias/{ids}
  - GET /medias/search
  - GET /medias/print/search
- name: influencer
  description: >-
    An influencer profile. Referenced from document via influencer_ids, and the Search API's
    /search/v1/influencers endpoint is named in the Documents overview as the source whose results are
    served from an Amazon-backed cache. The influencers endpoint itself is not separately documented.
  identifier: influencer id
- name: audit_log
  description: An audit record scoped to a team, a group and an object.
  identifier: composite (team, group, id)
  retrieved_by:
  - GET /auditlogs/{team}{group}{id}
- name: facet
  description: >-
    An aggregation bucket over a document search — counts by channel type, topic path, language,
    country or key phrase. Returned by /documents/facets, not independently addressable.
relationships:
- from: document
  to: author
  kind: has_one
  via: author
  evidence: author object embedded in the document response example
- from: document
  to: social_metrics
  kind: has_one
  via: social
  evidence: social object embedded in the document response example
- from: document
  to: influencer
  kind: has_many
  via: influencer_ids
  evidence: influencer_ids array on the document response example
- from: document
  to: media
  kind: belongs_to
  via: provider
  evidence: >-
    provider names the outlet or platform (e.g. "YouTube") on each document; the Medias API is the
    outlet-level service. The join is by name/provider, not by an explicit media id on the document in
    the published examples.
- from: audit_log
  to: document
  kind: belongs_to
  via: id
  confidence: low
  evidence: >-
    The auditlogs path is /auditlogs/{team}{group}{id} where id is the audited object. The reference
    does not say which object types are audited, so this edge is inferred from the path shape only.
- from: facet
  to: document
  kind: aggregates
  via: query
  evidence: /documents/facets aggregates over the same query as /documents
filter_dimensions:
  note: >-
    The Search API exposes these as *_filter parameters, which is the clearest published statement of
    how documents are dimensioned.
  dimensions:
  - channel_type_id
  - topic_paths
  - language_id
  - country_path
  - key_phrases
id_prefixes: []
render: null
gaps:
- No schema, type or requiredness information is published for any field.
- The full response_type value set is not enumerated.
- The influencers endpoint is referenced but has no documentation page.
- No relationship between document and the Insights/Media Impact Value scoring model is documented.