Document Type: Framework
Status: Active
Version: v1.0
Authority: MWMS HeadOffice
Applies To: Experimentation Brain, Affiliate Brain, Finance Brain, Research Brain
Parent: Experimentation Brain
Last Reviewed: 2026-04-03
Purpose
This page defines the minimum learning value required for an experiment to justify capital deployment.
Experiments consume capital.
Experiments must therefore produce meaningful learning regardless of outcome.
Learning value thresholds ensure:
- capital is used productively
- weak tests are avoided
- experiment design remains intentional
- insights accumulate systematically
- scaling decisions are based on knowledge rather than assumptions
Core Principle
A failed experiment that produces learning has value.
An experiment that produces no learning has cost but no return.
Learning Value Categories
Category 1 — Directional Learning
Provides insight into whether a hypothesis direction may be promising.
Examples:
hook direction viability
audience responsiveness indication
mechanism interest signal
creative format preference indication
Directional learning reduces uncertainty.
Category 2 — Comparative Learning
Provides insight through comparison between variations.
Examples:
creative A vs creative B
hook variation comparison
audience segment comparison
format comparison
Comparative learning improves decision clarity.
Category 3 — Behavioural Learning
Provides insight into how audiences respond.
Examples:
engagement behaviour patterns
click behaviour patterns
scroll behaviour patterns
view duration behaviour
Behavioural learning improves targeting and messaging decisions.
Category 4 — Structural Learning
Provides insight into funnel or experience effectiveness.
Examples:
landing page structure impact
content sequencing impact
call-to-action positioning impact
information order effectiveness
Structural learning improves conversion environment.
Category 5 — Signal Sensitivity Learning
Provides insight into how responsive the environment is to change.
Examples:
small changes producing measurable behavioural shifts
creative adjustments influencing engagement
message clarity affecting interaction behaviour
Signal sensitivity improves optimisation efficiency.
Category 6 — Boundary Learning
Provides insight into limitations.
Examples:
identifying non-responsive audience segments
identifying ineffective hook angles
identifying weak creative formats
identifying offer mismatch signals
Boundary learning prevents wasted scaling attempts.
Learning Value Evaluation Questions
Before capital deployment, Experimentation Brain should be able to answer:
What will we learn if this test succeeds?
What will we learn if this test fails?
What decision will this learning support?
How will this learning reduce uncertainty?
Low Learning Value Warning Signals
Experiments may produce weak learning value when:
hypothesis unclear
variables poorly defined
outcomes difficult to interpret
no clear decision path from results
test design lacks structure
Learning Value Discipline
Experiments should prioritise:
clarity of insight
interpretability of outcomes
reduction of uncertainty
decision support value
Relationship to Other Pages
Experimentation Evidence Validation Criteria
Experimentation Signal Strength Classification
Experimentation Confidence Progression Model
Finance Experimentation Alignment Model
Experimentation Brain Canon
Architectural Role
This page ensures capital deployed for experimentation produces meaningful insight that supports future decision-making.
Future Expansion
Future versions may include:
learning value scoring models
experiment prioritisation frameworks
learning dashboards
uncertainty reduction metrics
Change Log
Version: v1.0
Date: 2026-04-03
Author: MWMS HeadOffice
Change: Initial creation of Experimentation Learning Value Thresholds defining minimum insight value expectations for experiments.