Data Brain Signal Classification Framework

Document Type: Framework
Status: Active
Version: v1.2
Authority: MWMS HeadOffice
Parent: Data Brain Canon
Last Reviewed: 2026-04-22


Purpose

The Data Brain Signal Classification Framework defines how MWMS organises, labels, and interprets behavioural and performance signals so they can be used consistently across the ecosystem.

Signal classification improves:

clarity
comparability
measurement trust
decision speed
cross-brain interpretation quality

The framework exists to prevent different parts of MWMS from classifying the same type of signal differently.

Strong classification improves system coherence.

This framework distinguishes between:

structural signal type
behavioural signal meaning

Both dimensions are required for reliable interpretation.


Scope

This framework applies to:

• event signals
• parameter signals
• metric signals
• dimension signals
• behavioural signals
• attribution signals
• channel signals

This framework governs how signals are categorised and distinguished from one another.

It does not govern:

• final reporting layout
• campaign setup
• experiment approval
• capital allocation

Those remain governed by the relevant Brain systems.


Core Principle

Not all data elements represent the same type of signal.

MWMS must distinguish between:

event
parameter
dimension
metric

Structural classification defines signal format.

Behavioural classification defines signal meaning.

Both layers must remain consistent across the ecosystem.

If these layers are mixed together, interpretation quality declines.


Structural Signal Type Classification

Signal classification begins with structural type.


Event Signal Definition

An event signal represents that something happened.

Examples:

page_view
session_start
first_visit
form_submit
video_progress
purchase

Event signals are action records.

They describe occurrence.

Events should be treated as foundational behavioural units.

Events are the base layer of the signal stack.


Parameter Signal Definition

A parameter signal describes the context of an event.

Examples:

page location
page title
source
medium
product name
progress percentage
device category

Parameter signals do not stand alone well.

They gain meaning when attached to the correct event.

Parameters should be treated as contextual qualifiers.

Parameters enrich event interpretability.


Dimension Signal Definition

A dimension signal is a categorical field used to group and interpret data.

Examples:

source / medium
default channel group
device category
geography
content category
campaign
offer identifier

Dimensions answer:

which type
which source
which group
which category

Dimensions support segmentation and comparison.

Dimensions provide analysis structure.


Metric Signal Definition

A metric signal is a quantitative measure.

Examples:

sessions
users
engaged sessions
revenue
event count
purchases
conversion rate

Metrics answer:

how many
how much
how often

Metrics require correct definitions and correct context.

Metrics should not be interpreted without knowing the events and dimensions behind them.


Signal Stack Rule

Signals should be interpreted in layered order:

event

parameter

dimension

metric

interpretation

Example:

event = purchase
parameter = product_type
dimension = traffic_source
metric = purchase_count

Interpretation occurs only after structural clarity is established.

Structural clarity protects interpretation reliability.


Behavioural Signal Meaning Classification

Signals may also be classified by behavioural role.

Behavioural classification defines decision-stage relevance.


Exposure Signals

Indicate presence or visibility.

Examples:

impressions
reach
landing page view
session start

Low decision certainty.

Useful for visibility diagnostics.


Attention Signals

Indicate user awareness or content consumption.

Examples:

scroll depth
video start
time on page threshold
content expansion

Indicate attention but not commitment.

Useful for message resonance evaluation.


Engagement Signals

Indicate active interaction with environment.

Examples:

navigation interaction
multi-page visits
content exploration
tool interaction

Indicate relevance perception.

Moderate decision significance.


Intent Signals

Indicate directional movement toward decision.

Examples:

CTA click
outbound click
pricing page interaction
product detail interaction
feature exploration

Strong signal of potential conversion progression.


Progression Signals

Indicate commitment advancement.

Examples:

form start
add to cart
checkout initiation
lead submission initiation

High predictive value.

Critical for funnel diagnostics.


Outcome Signals

Indicate conversion completion.

Examples:

purchase
qualified lead
booking confirmation
subscription activation

Primary performance signals.

Used for ROI evaluation.


Monetization Signals

Indicate realized economic value.

Examples:

revenue recorded
order value captured
lifetime value increase
upsell acceptance

Highest certainty signal layer.

Supports financial evaluation.


Structural vs Behavioural Classification Distinction

Structural classification identifies signal format.

Behavioural classification identifies signal meaning.

Example:

Event Type:
form_submit

Structural classification:
event

Behavioural classification:
outcome signal

Both classifications must remain clear.


Signal Interpretation Dependency Rule

Higher-order signals depend on lower-order structure.

Example:

metric interpretation depends on correct dimension structure

dimension interpretation depends on correct parameter structure

parameter interpretation depends on correct event design

Classification errors at lower layers contaminate interpretation at higher layers.

Strong signal classification protects interpretive integrity.


Segmentation Rule

Dimensions should be used to segment metrics and event outputs into useful comparison groups.

Without segmentation:

important signal differences remain hidden.

Signal classification supports:

source-based segmentation
device-based segmentation
funnel-based segmentation
geography-based segmentation
campaign-based segmentation

Segmentation improves signal usefulness.


Relationship to Other Data Brain Frameworks

Supports:

Data Brain Event Measurement Framework
Data Brain Measurement Integrity Framework
Data Brain Traffic Source Interpretation Framework
Data Brain Attribution Reliability Framework
Data Brain Event Value Classification Framework
Data Brain Conversion Definition Framework

Signal classification provides structural language for Data Brain interpretation logic.


Drift Protection

The system must prevent:

events being treated as metrics
parameters being treated as standalone business outcomes
dimensions being confused with metrics
top-line metrics being interpreted without structural traceability
different Brains creating conflicting signal category definitions

Signal classification must remain stable, explicit, and shared.


Architectural Intent

The Data Brain Signal Classification Framework ensures MWMS uses a consistent structural language for interpreting data.

Clear signal classes improve:

analysis quality
handoff quality
experiment interpretation
attribution quality
strategic clarity

Strong classification makes the ecosystem easier to scale and safer to interpret.


Change Log

Version: v1.2
Date: 2026-04-22
Author: MWMS HeadOffice

Change:

Added behavioural signal meaning layer aligned with Event Value hierarchy.

Clarified distinction between structural signal type and behavioural signal meaning.

Improved compatibility with Conversion Definition Framework and Behavioural Event Analysis Framework.

Strengthened signal stack clarity for improved cross-brain interpretation consistency.