Research Brain LTV Signal Framework

Document Type: Framework
Status: Active
Version: v1.0
Authority: MWMS HeadOffice
Parent: Research Brain Canon
Last Reviewed: 2026-04-12


Purpose

The Research Brain LTV Signal Framework defines how MWMS interprets behavioural and transactional indicators to estimate expected customer value and guide acquisition, lifecycle, and forecasting decisions.

Customer Lifetime Value (LTV) is not a fixed number.

LTV is a probabilistic estimate derived from observed behaviour patterns.

LTV signals improve:

CAC allocation decisions
channel prioritisation decisions
lifecycle investment decisions
retention strategy prioritisation
growth sustainability evaluation

Improved LTV signal clarity improves capital efficiency.

Capital efficiency improves growth stability.


Scope

This framework governs:

interpretation of value signals
probabilistic estimation logic
signal weighting inputs
cohort-informed value expectations
LTV uncertainty handling
directional decision usage

Applies to:

customer acquisition evaluation
lifecycle investment decisions
offer economics evaluation
channel prioritisation
cohort performance analysis

Does not govern:

financial accounting definitions
revenue recognition policy
accounting system calculations

These are governed by Accounting Canon.


Core Principle

LTV is an estimate derived from behavioural evidence.

Observed behaviour informs expected future behaviour.

Expected future behaviour informs expected economic value.

Higher behavioural stability improves LTV predictability.

LTV must be interpreted directionally, not deterministically.

LTV reduces uncertainty but does not eliminate uncertainty.


LTV Signal Categories

Repeat Purchase Signals

indicate ongoing relationship probability.

examples:

second purchase occurrence
repeat purchase frequency
purchase interval consistency
repeat purchase acceleration

repeat purchase behaviour strongly influences expected lifetime value.

second purchase probability significantly increases expected LTV.

cohort analysis often shows second purchase as highest drop-off point.


Order Value Signals

indicate economic intensity of relationship.

examples:

average order value
order value growth over time
bundle purchase frequency
cross-category purchasing behaviour

higher order value increases expected economic contribution.

consistent order value improves predictability.


Retention Duration Signals

indicate relationship longevity.

examples:

time between first and most recent purchase
duration of engagement activity
persistence of interaction behaviour

longer retention duration increases cumulative expected value.

longer duration improves forecast confidence.


Product Expansion Signals

indicate increasing brand integration.

examples:

additional category purchases
accessory purchases
complementary product adoption

category expansion indicates deeper customer relationship formation.

expanded relationships often produce higher LTV trajectories.


Promotion Sensitivity Signals

indicate margin risk patterns.

examples:

discount-only purchasing behaviour
sale-only engagement patterns
promotion-driven purchasing frequency

high promotion sensitivity may reduce profit-adjusted LTV.

promotion sensitivity affects acquisition bidding logic.


Engagement Persistence Signals

indicate continued brand interaction.

examples:

email engagement consistency
site revisit frequency
content interaction depth

continued engagement indicates ongoing relationship probability.

engagement persistence supports retention durability assumptions.


Directional Use Principle

LTV should guide decision direction rather than define rigid acquisition ceilings.

example:

if estimated LTV = 100

acceptable CAC may be set at conservative fraction of LTV estimate.

fractional interpretation reduces forecasting risk.

uncertainty buffer protects growth stability.


Cohort-Informed Estimation

LTV estimation should consider cohort-level behavioural patterns.

cohort retention curves inform expected purchase frequency.

cohort decay patterns influence expected duration of relationship.

cohort stability improves estimate reliability.

unstable cohorts increase estimation uncertainty.

cohort-informed estimation improves forecasting accuracy.


Time Horizon Sensitivity

longer projection horizons increase uncertainty.

short-term LTV estimates are more reliable than long-term projections.

decision logic should consider confidence interval width.

higher uncertainty requires conservative CAC allocation.

uncertainty-aware decisions improve capital protection.


Relationship to Cohort Behaviour Framework

cohort behaviour provides empirical evidence for LTV estimation.

repeat purchase timing informs revenue accumulation expectations.

cohort decay rate influences expected value trajectory.

cohort analysis improves signal reliability.


Relationship to Acquisition Framework

LTV signals influence channel investment prioritisation.

example:

channel A produces lower CPA but weak retention
channel B produces higher CPA but stronger retention

channel B may produce higher long-term value.

LTV interpretation improves acquisition allocation decisions.


Relationship to Lifecycle Framework

lifecycle optimisation influences realised LTV.

improved onboarding improves second purchase probability.

improved engagement improves repeat purchase persistence.

lifecycle optimisation increases realised value from existing customers.


Relationship to Forecasting Framework

forecasting requires estimation of expected revenue per customer.

LTV signals inform revenue projections.

cohort-informed LTV improves forecast accuracy.

stable LTV signals reduce projection volatility.


Signal Confidence Principle

LTV estimation confidence increases with behavioural evidence accumulation.

early-stage customers have wider value uncertainty range.

confidence improves after repeat behavioural confirmation.

more behavioural data improves estimate reliability.

confidence weighting improves decision accuracy.


Drift Protection

system must prevent:

treating LTV as guaranteed value
ignoring uncertainty range
assuming stable behaviour across all segments
over-allocating acquisition budget based on optimistic projections
using outdated cohort behaviour assumptions

LTV must remain adaptive.

estimates must adjust as behaviour evolves.


Architectural Intent

Research Brain LTV Signal Framework enables MWMS to interpret behavioural patterns as probabilistic value indicators that improve acquisition efficiency and forecasting stability.

value signals guide capital allocation decisions.

capital allocation influences growth sustainability.

sustainable growth depends on accurate value interpretation.


Future Expansion

predictive LTV modelling
dynamic cohort-adjusted value estimation
probabilistic value range modelling
margin-adjusted LTV estimation
CAC:LTV adaptive bidding logic
retention-adjusted forecasting models

future models improve estimation precision.


Final Rule

LTV is directional intelligence.

MWMS uses LTV signals to improve decision quality, not to assume certainty.


Change Log

Version: v1.0
Date: 2026-04-12
Author: MWMS HeadOffice

Change: Initial creation of Research Brain LTV Signal Framework defining behavioural value indicators, probabilistic estimation logic, cohort-informed signal interpretation, and uncertainty-aware decision guidance structure.


CHANGE IMPACT

Pages Created:

Research Brain LTV Signal Framework

Pages Updated:

None

Pages Deprecated:

None

Registries Requiring Update:

MWMS Architecture Registry
MWMS Brain Registry
MWMS Intelligence Layer Map
MWMS Canon Hierarchy Map

Canon Version Update Required: No
Change Log Entry Required: Yes