Document Type: Specification
Status: Active
Version: v1.1
Authority: Affiliate Brain (Audience Infrastructure Layer)
Applies To: Audience Engine event-category definitions, metadata expectations, state-influence rules, and anti-drift measurement controls
Parent: Supabase Event Schema
Last Reviewed: 2026-03-14
Purpose
This document defines the canonical meaning of each Audience Engine event_category.
Its purpose is to prevent metric drift by enforcing:
• category boundaries
• required minimum metadata expectations
• prohibited misuse patterns
• mapping rules to the Audience State Progression Schema
This layer is definitional only.
It does not execute.
It does not allocate capital.
It does not change Velocity.
Scope
This specification applies to:
• Audience Engine event-category definitions
• valid and invalid category usage
• minimum metadata expectations for audience events
• state-influence interpretation rules
• anti-inflation controls for engagement, subscription, advocacy, retention, attribution, and system events
This document governs what each allowed event category means and how it must be interpreted.
It does not govern:
• direct campaign execution
• paid traffic allocation
• Velocity decisions
• audience capital approval
• live automation rollout
• scoring logic beyond defined influence boundaries
Those remain governed by Affiliate Brain, HeadOffice, Finance Brain, and related audience infrastructure documents.
Definition / Rules
Canonical Event Categories (Allowed List)
Only the following categories are permitted:
• engagement
• subscription
• advocacy
• retention
• attribution
• system
Any new category requires canon update.
Category 1 – Engagement
Definition:
Signals that a user has meaningfully interacted with content beyond passive exposure.
Valid examples (event_type):
• content_view
• scroll_depth_75
• video_completion_75
• multi_internal_click
• long_form_completion
• repeat_session_30d
• high_frequency_session
Minimum metadata expectations:
• content_id or page_path
• content_type (article, video, tool, page)
• duration_seconds or scroll_depth_percent
• referrer or utm_campaign, if present
Prohibited misuse:
• counting page load as engagement without a threshold
• using impressions as engagement
• treating a single click as meaningful engagement
State influence:
• enables Anonymous → Engaged Viewer
• contributes to engagement_score
• supports Repeat Consumer qualification
Category 2 – Subscription
Definition:
Signals that a user has explicitly entered a permissioned channel.
Valid examples (event_type):
• email_subscribe
• asset_download
Minimum metadata expectations:
• channel (email, sms, community, push)
• asset_id, if download
• consent_recorded = true (boolean)
Prohibited misuse:
• auto-subscribing without explicit consent
• counting account creation as subscription unless it grants permissioned contact
• logging download without confirming completion
State influence:
• enables Engaged Viewer → Subscriber
• establishes durable identity anchor
• upgrades retention expectations
Category 3 – Advocacy
Definition:
Signals that a user has amplified, endorsed, or publicly interacted in a way that spreads reach.
Valid examples (event_type):
• content_share
• referral_click
• comment_repeat
• mention_tag
• advocacy_repeat
Minimum metadata expectations:
• platform (x, facebook, linkedin, youtube, reddit, email, other)
• share_type (link, share, comment, tag)
• outbound_target, if referral
Prohibited misuse:
• treating likes as advocacy
• treating a single comment as advocacy without repeat behavior
• counting internal navigation as referral
State influence:
• enables Repeat Consumer → Advocate
• strengthens Narrative Momentum signals
• increases authority_density over time
Category 4 – Retention
Definition:
Signals that a user’s relationship with the hub has weakened or strengthened over time based on inactivity or return patterns.
Valid examples (event_type):
• downgrade_state
• inactivity_threshold
Minimum metadata expectations:
• prior_state
• new_state
• inactivity_days
• rule_id (which retention rule triggered downgrade)
Prohibited misuse:
• manual downgrades without rule trace
• punishment downgrades during temporary campaign pauses
• silent state changes without audit trail
State influence:
• allows accurate decay logic
• preserves honest audience health metrics
• prevents inflated audience totals
Category 5 – Attribution
Definition:
Captures measurement context used to attribute sessions and actions to traffic sources.
Valid examples (event_type):
• utm_recorded
• session_start
Minimum metadata expectations:
• utm_source
• utm_medium
• utm_campaign
• session_id
• landing_page
Prohibited misuse:
• missing utm_campaign for paid traffic
• free-text campaign naming outside taxonomy
• overwriting session attribution after engagement begins
State influence:
• no state upgrades directly
• enables accurate acquisition mapping
• supports integrity reconciliation
Category 6 – System
Definition:
Events generated by the system for validation, enforcement, and audit traceability.
Valid examples (event_type):
• event_validation_pass
• event_validation_fail
Minimum metadata expectations:
• validation_rule_id
• validation_result
• rejection_reason, if fail
• source_system (wp, supabase, edge_function, other)
Prohibited misuse:
• using system events to create engagement
• logging system failures without reasons
• silent rejection of events without audit trail
State influence:
• no state upgrades
• can block state upgrades if integrity fails
• enables governance enforcement
Category Boundary Enforcement Rules
Engagement vs Attribution
Attribution captures source context.
Engagement captures behavior.
Subscription requires explicit consent
No consent means it is not subscription.
Advocacy requires outward amplification
Internal actions do not qualify.
Retention requires rule trace
No rule_id means the retention event is invalid.
System events never affect score directly
They only validate or invalidate.
Compliance Safety Note
This schema does not store sensitive personal data by default.
When identity data is captured:
• store minimal identifiers only
• avoid health, political, or religious inference
• keep consent proof separate from behavior metadata
Drift Protection
The system must block:
• new categories without canon update
• engagement inflation via impressions
• advocacy inflation via likes
• subscription inflation via auto opt-ins
• retention changes without rule trace
• system events being counted as audience growth
Audience event definitions must remain strict, auditable, and non-inflationary.
Architectural Intent
Audience Event Category Definitions exists to ensure that audience infrastructure inside MWMS remains measurement-clean, state-safe, and resistant to metric inflation.
Its role is to make event interpretation consistent across logging, state progression, integrity validation, and future audience-system analysis.
Change Log
Version: v1.1
Date: 2026-03-14
Author: MWMS HeadOffice / Affiliate Brain
Change: Rebuilt page to align with MWMS document standards. Added standardised document header, replaced legacy metadata with compliant structure, introduced Purpose / Scope / Definition / Rules format, normalised category formatting, and preserved the original canonical event categories, metadata expectations, state-influence rules, and drift protections.
Version: v1.0
Date: 2026-02-25
Author: Affiliate Brain
Change: Initial creation of Audience Event Category Definitions defining canonical event categories, category boundaries, metadata expectations, state-influence mapping, compliance safety note, and drift-protection rules.
END – AUDIENCE EVENT CATEGORY DEFINITIONS v1.1