01 · Account surfaceDocumenté
Public profile or page
A source-specific public account surface, not a verified real-world person.
- Source profile ID
- Handle and public name
- Profile URL and visible metadata
Les réseaux sociaux
Public social data with every post tied to its surface, time, and evidence.
Collect documented Twitter profile and tweet results, access eligible public pages through web access APIs, or pilot a maintained multi-source feed that keeps profiles, posts, media, threads, topics, and engagement snapshots interpretable.
Public social record model
A feed item is not the profile, conversation, or metric history around it.
Keep stable source objects separate from changing observations and derived labels so downstream teams can interpret what was public, where, and when.
02 · ContentDocumenté
Post
A public source post with source identity, visible text, publication cue, and URL.
- Post ID and URL
- Public text and time
- Author profile relationship
03 · Time-bound valueDocumenté
Engagement observation
The visible metrics attached to a specific collection time and context.
- Like or reaction count
- Reply and share count
- observed_at and availability state
04 · RelationshipLe pilote d'abord
Thread or reply
A source-exposed parent, reply, quote, or conversation relationship where the approved surface supports it.
- Parent post ID
- Relationship type
- Pagination and depth state
05 · AttachmentLe pilote d'abord
Media and topic
Public media references, hashtags, and source or derived topic labels kept distinct from the post.
- Attachment URL or ID
- Hashtag and mention
- Topic rule or model version
Object boundaryNot assumed
A public profile is not a verified person.
Source IDs, handles, posts, and visible engagement remain platform observations. Private access, sensitive-trait inference, and cross-platform identity conclusions are not implied.
Coverage contract
Confirm the public surface, object depth, and request context together.
A platform name alone does not define which pages, queries, fields, pagination depth, or history a workflow can use.
Engine or familyLes donnéesContextesÉtat
Record schema
Keep the source post and mutable metrics in separate layers.
Every observation should carry enough source and collection context to explain what the count, relationship, or availability state means.
Public source identity
- source_profile_id
- Platform identifier exposed by the source.
- handle
- Public account handle at observation time.
- profile_url
- Public source page associated with the profile.
Public content object
- source_post_id
- Platform identifier for the post.
- text
- Public text exposed on the approved surface.
- published_at
- Source-attributed post time, when exposed.
Time-bound engagement
- metrics
- Visible named counts with source semantics retained.
- observed_at
- Time the collection produced the snapshot.
- availability_state
- Observed, absent, unavailable, failed, excluded, or review.
Relationships and provenance
- parent_post_id
- Source-exposed reply or thread relationship.
- source_url
- Public page associated with the observation.
- schema_version
- Version of the agreed output shape.
{
"record_id": "social_obs_0832",
"object_type": "post_observation",
"profile": {
"source": "twitter",
"source_profile_id": "demo_204",
"handle": "example_studio"
},
"post": {
"source_post_id": "post_demo_17",
"published_at": "YYYY-MM-DDT07:55:00Z"
},
"metrics": { "likes": 148, "reposts": 21 },
"observed_at": "YYYY-MM-DDT08:32:00Z",
"collection_state": "observed",
"schema_version": "social.v1"
}Read the snapshot literally.
Engagement metrics are observations, not audited totals. The visible count can change, be rounded, or differ by source surface.
Identity and relationships
Preserve platform identity before adding any cross-source interpretation.
Profiles, posts, replies, and media references keep their source IDs and URLs. Derived relationships remain separately versioned and reviewable.
- 01
Surface key
platform + profile ID
Retained exactly as observed. - 02
Content key
source post ID
Joined only within source semantics. - 03
Relationship
author, parent, media, topic
Source links and derived links kept apart. - 04
Observation
object + observed_at
Mutable fields create a new snapshot.
Freshness semantics
Treat every mutable metric as a time-stamped observation.
Publication time, collection time, delivery time, and first or last seen answer different questions and should remain separate.
- Source timepublished_at
The post time attributed by the source, when exposed.
- Observéobserved_at
The profile, post, or metric was visible in the requested public context.
- Delivereddelivered_at
The snapshot became available to the customer.
- Changedfirst_seen / last_seen
The source object or metric state entered or left the monitored history.
Quality and missingness
A missing post, a zero metric, and an inaccessible surface are not equivalent.
Preserve the reason state so downstream analytics do not invent meaning where the source or collection supplied none.
Required structure present
The profile, post, or metric snapshot passed agreed checks.
Field not displayed
The approved public surface did not expose the requested value.
Relationship or layout ambiguous
A thread, source object, or parser state needs inspection.
No supported observation
Restricted, removed, failed, or excluded stays distinct from a visible zero.
Structural checks
Required IDs, URLs, types, metric names, timestamps, collection state, and schema version.
Field checks
Source-specific count types, relationship keys, pagination state, and public-surface evidence.
Acceptance boundary
Your team approves permitted sources, representative records, retention, and analytical thresholds; we operate the agreed technical contract.
Operating model
Choose how much of the public social data operation your team owns.
Compare infrastructure, request-time APIs, recurring feeds, and a managed program against the same record and quality contract.
Maximum control
Build the full public social workflow on proxy infrastructure.
Your team selects eligible surfaces, collects pages, parses objects, manages identities and snapshots, monitors changes, and delivers records.
ResponsabilitéPropriétaire
Source and field briefCustomer
Access infrastructureShared
Extraction and schemaCustomer
Quality and maintenanceCustomer
Storage and decisionsCustomer
Applications
Use one public-evidence layer across listening, research, and measurement.
The records show what was visible on an approved public surface. Your team defines audience models, sentiment, attribution, thresholds, and decisions.
Social listening
Track approved topics, posts, and public conversations with source context.
Posts · threads · hashtagsBrand monitoring
Surface new public mentions and engagement snapshots for review.
Queries · posts · observed_atCreator landscape research
Compare public account and content signals without treating them as identity proof.
Profiles · posts · public metricsCampaign visibility
Observe selected public content and source-displayed engagement over time.
Posts · media · metric snapshotsÉtude de marché
Build source-linked public corpora for approved topic and narrative analysis.
Content · topics · provenanceAI data preparation
Prepare governed public examples with time, source, and collection state attached.
Text · relationships · schema versionRepresentative pilot
Validate public access, object depth, and metric semantics before scale.
Use approved public profiles, posts, threads, and queries that represent ordinary content, missing fields, pagination, unavailable states, and repeated engagement snapshots.
- 01
Define
Public sources, objects, fields, cadence, retention, and permitted use.
- 02
Sample
Representative profiles, posts, threads, media, and missing states.
- 03
Inspect
Source IDs, metric times, relationships, collection states, and exceptions.
- 04
Accept
Approve the schema, public-source set, operating model, and thresholds.
Evaluation FAQ
Define public scope and interpretation before delivery begins.
Answers separate documented access from the source-specific work a representative pilot needs to validate.
What social media data is documented today?
The Twitter Search API documents profile and tweet result workflows. Generic Scraper API or Browser API access can also be used for eligible public pages when your team owns the extraction. Broader normalized feeds across Instagram, TikTok, YouTube, Reddit, LinkedIn, and other sources are pilot first rather than assumed standard coverage.
Which social objects can a scoped feed contain?
A scoped record model can include a public profile or page, post, engagement observation, media attachment reference, thread or reply relationship, and hashtag or topic context. Source identifiers and URLs remain attached so those objects can be interpreted without pretending every platform exposes the same fields.
Are engagement counts exact totals?
Engagement metrics are observations, not audited totals. A visible like, reply, share, repost, or view count can change after collection, may be rounded, and may not represent all platform activity. Each metric should retain the source, observation time, and availability state.
Can public profiles be matched to the same person across platforms?
Cross-platform person matching is not assumed. Source profile IDs, handles, public names, and URLs remain source-specific. Any entity-resolution pilot requires an agreed business purpose, permitted fields, explainable cues, explicit uncertainty, and a separate responsible-use review.
Can comments, replies, and threads be delivered?
Public reply or thread relationships can be evaluated where the selected documented endpoint or eligible public surface exposes them. Coverage depth, ordering, pagination, deleted states, and nested relationships vary by source, so a representative pilot defines the usable object and field contract.
How fresh can social observations be?
Documented APIs return observations when a request or asynchronous job is processed. Scheduled or managed collection uses an agreed cadence by surface, object, query, and source. The record retains observed_at, and repeated engagement snapshots remain separate rather than overwriting history.
How are deleted, unavailable, or missing posts represented?
The contract can distinguish a post not returned by the requested surface, a field not shown, a page unavailable, a collection failure, a changed layout, a deleted or removed cue exposed by the source, and an excluded surface. These states are not collapsed into a zero engagement count or empty post.
Are private profiles or messages included?
No. Public social surfaces only. Private profiles, private messages, member-only content, authenticated inboxes, and other restricted areas are not standard scope. Public accessibility also does not remove the need for a lawful purpose and appropriate retention and access controls.
Can WebScrapingAPI provide personal-data enrichment?
The standard scope does not promise personal-data enrichment, sensitive-trait inference, private contact discovery, or identity verification. Customers remain responsible for the lawful basis, minimization, retention, access controls, and decisions associated with any permitted public fields.
Who maintains social source changes?
Proxy customers maintain their own collectors and parsers. WebScrapingAPI maintains access within documented product boundaries. Scheduled feeds and Managed Web Data can include contracted extraction, normalization, technical quality monitoring, source-change maintenance, and delivery for the agreed public source set.
Les réseaux sociaux
Start with documented access or scope a governed public feed.
Bring the approved sources, objects, fields, cadence, and use case. We will help define the record and responsible-use boundary before scale.