Document Type: Framework
Status: Active
Version: v1.1
Authority: Data Brain
Parent: Data Brain Architecture
Applies To: Ads Brain, Experimentation Brain, Affiliate Brain, Conversion Brain, Research Brain
Last Reviewed: 2026-04-22
Purpose
The Data Brain Paid Media Measurement Framework defines the measurement structure required to evaluate paid media performance accurately inside MWMS.
The framework ensures paid media decision-making is based on:
reliable signal interpretation
consistent performance measurement
cross-platform comparability
structured attribution logic
stable optimisation signals
interpretable scaling indicators
behavioural signal progression clarity
Paid media platforms often provide incomplete or biased performance metrics.
This framework establishes a structured measurement layer independent of platform reporting limitations.
Reliable paid media interpretation requires alignment between:
event structure
signal hierarchy
conversion definitions
attribution logic
financial relevance
Measurement discipline protects decision reliability.
Core Principle
Platform dashboards are not the source of truth.
They are signal interfaces.
MWMS must interpret platform data through structured measurement logic.
Measurement must prioritise:
decision relevance
behavioural signal clarity
conversion reliability
cost stability
signal hierarchy consistency
Metrics must support decision-making, not vanity reporting.
Platform-reported signals must be interpreted within behavioural context.
Behavioural Signal Hierarchy
Paid media measurement must align to structured signal hierarchy.
Hierarchy example:
Business Outcome Signals
revenue impact
profit impact
customer acquisition quality
customer lifetime value
repeat purchase behaviour
Highest confidence signals.
Primary scaling reference.
Conversion Signals
sales conversions
qualified leads
booked calls
purchases
confirmed applications
Primary decision-stage outcomes.
Used for ROI interpretation.
Progression Signals
checkout initiation
form initiation
trial initiation
application initiation
lead qualification steps
Indicate movement toward conversion.
Important for diagnosing funnel friction.
Intent Signals
CTA clicks
offer exploration behaviour
pricing interaction
product detail interaction
Indicate directional decision movement.
Strong early signal indicators.
Engagement Signals
click-through rate
scroll depth
video completion
multi-page interaction
time on page
Indicate behavioural interest strength.
Useful for creative filtering.
Exposure Signals
impressions
reach
frequency
Indicate visibility level.
Weak indicators when used alone.
Lower-layer signals support interpretation of higher-layer signals.
Primary decisions must not rely on exposure signals alone.
Signal hierarchy improves interpretation discipline.
Source of Truth Rule
Primary performance interpretation must rely on verifiable signals where possible.
Examples:
server-side tracking
CRM conversion records
verified transaction data
confirmed lead qualification data
validated behavioural events
Platform-reported metrics may contain distortion due to:
optimisation bias
attribution modelling
reporting delays
metric inflation
platform-defined conversions
inferred signals
MWMS measurement must prioritise verifiable signals.
Higher-confidence signals should influence scaling decisions.
Lower-confidence signals should guide hypothesis development.
Event Structure Dependency Rule
Paid media metrics depend on underlying event structure.
Examples:
conversion rate depends on conversion event definition
CPA depends on conversion integrity
CTR depends on click definition consistency
engagement metrics depend on event capture consistency
If event structure is weak:
paid media interpretation becomes unreliable.
Event measurement quality precedes media optimisation quality.
Stable event definitions improve comparability across tests.
Attribution Structure
Paid media attribution must be defined intentionally.
Attribution logic determines how conversion credit is assigned across touchpoints.
Common attribution considerations:
click attribution window
view attribution window
multi-touch influence
last-touch influence
assisted conversions
delayed conversions
Attribution settings influence reported performance.
Attribution settings must be documented for interpretability.
Unclear attribution logic reduces comparability across tests.
Attribution structure influences signal meaning.
Conversion Value Assignment
Where possible, conversion events should include defined value weighting.
Example conversion value hierarchy:
purchase = high value
qualified lead = medium value
email signup = low value
page engagement = diagnostic value
Assigning value improves optimisation signal quality.
Value assignment improves:
budget allocation decisions
scaling prioritisation
experiment comparison clarity
traffic quality interpretation
Conversion value structure must reflect behavioural significance.
Value weighting improves signal prioritisation.
Pre-Conversion Signal Interpretation Layer
Paid media learning often begins before primary conversion volume accumulates.
Pre-conversion signals provide early behavioural intelligence.
Examples:
high engagement without intent signals may indicate curiosity without relevance.
high intent signals without progression signals may indicate friction.
strong progression signals may indicate conversion readiness.
Pre-conversion signals support early traffic filtering.
Pre-conversion signals accelerate learning cycles.
Pre-conversion signals must not be mistaken for final outcomes.
Signal-layer clarity protects optimisation discipline.
Custom Metric Requirement
Standard platform metrics may not provide sufficient interpretability.
Custom metric definitions may be required.
Examples:
cost per qualified lead
cost per high-intent visit
cost per engaged user
cost per completed funnel stage
effective cost per acquisition
cost per progression event
Custom metrics must align with behavioural interpretation needs.
Custom metrics improve cross-platform comparability.
Custom metrics improve signal interpretability.
Reporting Structure
Paid media performance must be exportable for structured analysis.
External reporting environments may include:
spreadsheet environments
data warehouses
dashboards
modelling environments
Exported data enables:
deeper analysis
cross-campaign comparison
segmented performance analysis
custom metric construction
signal-layer comparison
Platform interfaces alone may limit analytical depth.
Structured export improves interpretability.
Naming Convention Structure
Campaign naming structures must support measurement clarity.
Naming conventions may include:
traffic temperature
audience segment identifier
geography identifier
offer identifier
funnel stage identifier
experiment identifier
Consistent naming improves:
report readability
segmentation accuracy
signal grouping clarity
analysis efficiency
Naming conventions act as measurement metadata.
Naming improves interpretability.
Signal Segmentation Capability
Measurement structures should allow segmentation across:
audience clusters
creative variants
funnel steps
geographic segments
device environments
traffic temperature layers
behavioural signal tiers
Segmentation improves interpretability.
Aggregated metrics may hide signal variation.
Segment-level measurement improves optimisation clarity.
Frequency Interpretation Layer
Frequency influences behavioural response patterns.
Frequency may impact:
message recall
brand familiarity
decision confidence
message fatigue
Measurement must monitor frequency-related signals.
Indicators of frequency imbalance:
declining engagement
rising cost per result
reduced incremental conversion gain
declining progression signals
Frequency interpretation improves scaling discipline.
Measurement Integrity Protection
Measurement structures must minimise:
duplicate conversion counting
attribution overlap distortion
metric misalignment
signal misinterpretation
weak event definition influence
behavioural stage confusion
Measurement must remain consistent across experiment comparisons.
Changing measurement logic mid-test invalidates comparability.
Measurement stability supports cumulative learning.
Relationship to Other MWMS Frameworks
Supports:
Experimentation Brain Paid Media Experiment Framework
Ads Brain Audience Experimentation Framework
Conversion Brain Funnel Structure Framework
Research Brain Behaviour Signal Framework
Data Brain Signal Integrity Framework
Data Brain Event Value Classification Framework
Data Brain Conversion Definition Framework
Provides measurement structure for paid traffic decision-making.
Signal structure consistency improves cross-Brain interpretation.
Architectural Intent
Paid media platforms optimise for internal objectives.
MWMS must optimise for business objectives.
Measurement structure protects MWMS from:
platform metric bias
incomplete performance interpretation
scaling misjudgements
misleading signal interpretation
premature optimisation decisions
Stable measurement logic enables cumulative learning across campaigns.
Measurement discipline strengthens system intelligence over time.
Change Log
Version: v1.1
Date: 2026-04-22
Author: Data Brain
Change:
Aligned paid media measurement hierarchy with Event Value Classification Framework.
Added pre-conversion signal interpretation layer for early-stage testing environments.
Strengthened linkage between event structure and paid media signal reliability.
Improved attribution interpretation clarity.
Improved segmentation compatibility with Behavioural Event Analysis Framework.
Strengthened protection against platform metric bias.