Research Brain Cohort-Based Revenue Forecasting Framework

Document Type: Framework
Status: Active
Version: v1.0
Authority: MWMS HeadOffice
Parent: Research Brain Canon
Slug: research-brain-cohort-based-revenue-forecasting-framework
Last Reviewed: 2026-04-13


Purpose

The Research Brain Cohort-Based Revenue Forecasting Framework defines how MWMS projects revenue using customer cohort behaviour rather than relying solely on top-line growth assumptions.

Traditional forecasting often assumes revenue grows proportionally with traffic or spend.

Cohort-based forecasting models how groups of customers behave over time.

Customer behaviour determines revenue durability.

Durable revenue improves strategic decision confidence.

Cohort modelling improves:

• revenue predictability
• growth planning accuracy
• acquisition budget calibration
• lifecycle optimisation prioritisation
• risk-adjusted scaling decisions
• capital allocation efficiency

Forecasting improves decision clarity.

Decision clarity improves scaling confidence.


Scope

This framework applies to:

• cohort-based revenue projections
• repeat purchase behaviour modelling
• customer file decay analysis
• customer lifetime value trend interpretation
• acquisition sensitivity modelling
• retention sensitivity modelling
• growth durability analysis
• forecast confidence band estimation

This framework governs how MWMS estimates future revenue based on behavioural patterns rather than assumptions.

It does not govern:

• accounting revenue recognition rules
• financial reporting compliance
• pricing strategy decisions
• budgeting governance processes

Those remain governed by Finance Brain systems.


Definition / Rules

Core Principle

Customers acquired at different times behave differently.

Cohorts represent groups of customers acquired within the same period.

Each cohort develops its own behavioural pattern.

Behaviour patterns influence revenue generation over time.

Forecasting accuracy improves when behavioural patterns are modelled explicitly.


Cohort Structure

Cohorts are typically defined by acquisition period.

Examples:

monthly cohorts
weekly cohorts
campaign-based cohorts
seasonal cohorts

Each cohort is tracked over time to observe:

repeat purchase rate
revenue contribution
engagement persistence
behavioural decay patterns

Behaviour over time provides predictive insight.


Customer File Growth Model

Total revenue is influenced by growth of the active customer file.

Active customer file size depends on:

new customer acquisition rate
repeat purchase frequency
customer churn rate
reactivation behaviour

Revenue durability improves when customer file growth is stable.

Customer file instability increases forecasting risk.


Customer File Decay Logic

Customer engagement decreases over time without reinforcement.

Decay patterns indicate how quickly customers stop purchasing.

Understanding decay rate improves forecast accuracy.

Decay insight informs lifecycle optimisation priorities.

Faster decay increases acquisition dependency.

Slower decay improves revenue stability.


Repeat Purchase Behaviour

Repeat purchase behaviour significantly influences lifetime value.

Key behavioural patterns include:

second purchase conversion rate
time between purchases
purchase frequency distribution
seasonal repeat patterns

Second purchase behaviour strongly predicts long-term customer value.

Improving second purchase rate increases cohort value durability.


Revenue Projection Layers

Revenue projections typically include multiple layers:

new customer revenue contribution
repeat customer revenue contribution
reactivated customer contribution
seasonal revenue variation effects

Layer separation improves forecast clarity.

Clarity improves decision quality.


Percentile-Based Forecast Ranges

Forecasting uncertainty should be explicitly represented.

Forecast outputs should include percentile ranges.

Examples:

conservative projection
expected projection
optimistic projection

Percentile modelling improves risk visibility.

Risk visibility improves capital allocation decisions.


Sensitivity Analysis

Forecast outputs should account for sensitivity variables.

Examples:

acquisition rate variation
repeat purchase rate variation
retention improvement effects
AOV variation effects

Sensitivity awareness improves planning resilience.


Relationship to Lifecycle Optimization Framework

Lifecycle improvements influence cohort behaviour durability.

Improved lifecycle performance improves repeat purchase rates.

Improved repeat purchase rates improve revenue predictability.

Lifecycle optimisation improves forecast stability.


Relationship to Unit Economics Framework

Customer lifetime value influences acceptable acquisition cost thresholds.

Higher lifetime value improves acquisition flexibility.

Improved flexibility improves scaling options.

Unit economics strength influences growth viability.


Relationship to Acquisition Systems

Customer acquisition quality influences cohort durability.

High-quality acquisition sources produce stronger cohorts.

Weak acquisition sources produce fragile cohorts.

Traffic quality influences forecast stability.


Relationship to Experimentation Systems

Experimentation improves behavioural performance.

Improved performance improves cohort strength.

Stronger cohorts improve revenue predictability.

Experimentation contributes to forecast reliability.


Failure Modes Prevented

This framework prevents:

overestimating revenue durability
assuming linear growth behaviour
ignoring customer file decay effects
relying solely on short-term revenue trends
underestimating repeat purchase influence
treating revenue growth as traffic-only function

Cohort modelling improves realism in growth planning.


Drift Protection

The system must prevent:

using single-point forecasts without uncertainty range
assuming lifetime value remains constant
ignoring retention variation across acquisition sources
extrapolating short-term behaviour indefinitely
overconfidence in simplified growth projections

Forecasts must remain behaviour-informed.


Architectural Intent

Research Brain Cohort-Based Revenue Forecasting Framework ensures MWMS bases growth expectations on observed behavioural patterns rather than simplified growth assumptions.

Customer behaviour determines revenue durability.

Revenue durability determines growth resilience.

Improved forecasting improves strategic decision stability.

Stable decisions improve scaling efficiency.


Change Log

Version: v1.0
Date: 2026-04-13
Author: MWMS HeadOffice

Change:

Initial creation of cohort-based revenue forecasting framework defining customer file growth logic, behavioural decay modelling, repeat purchase sensitivity analysis, and percentile-based forecast structuring.


END – RESEARCH BRAIN COHORT-BASED REVENUE FORECASTING FRAMEWORK v1.0