Ads Brain Cost Control And Bidding Governance Framework

Document Type: Framework
Status: Active
Version: v1.0
Authority: Ads Brain Governed By MWMS HeadOffice
Applies To: Ads Brain Use Of Cost Caps, Target CPA, Target ROAS, Bid Caps, Manual Bidding, Maximum Conversions, Maximum Conversion Value, Value Rules, Delivery Constraints And Related Bidding Controls
Parent: Ads Brain Canon
Last Reviewed: 2026-07-19

Purpose

The Ads Brain Cost Control And Bidding Governance Framework defines the conditions under which MWMS may use bidding constraints and optimisation controls inside paid advertising platforms.

Cost controls are not treated as shortcuts to profitability.

They are treated as delivery constraints that may improve cost discipline only when the underlying campaign already has:

reliable tracking
known unit economics
sufficient conversion history
commercially viable creative
a functioning funnel
an appropriate offer
clear rollback rules

This framework exists to prevent cost controls from being used to hide structural weakness.

Its purpose is to:

govern when cost controls may be activated
define evidence requirements before activation
distinguish conversion volume from conversion value
align platform bidding with true business goals
prevent underdelivery from being misinterpreted
identify audience, placement and creative concentration
protect new creative from starvation
preserve testing validity
monitor commercial quality
enforce rollback discipline
align Ads Brain with Finance Brain authority
preserve durable governance while platform features change

Scope

This framework applies to:

lowest cost bidding
maximise conversions
maximise conversion value
target CPA
cost caps
target ROAS
bid caps
manual bidding where relevant
value rules
value adjustments
bidding control activation
bidding control monitoring
bidding control rollback
underdelivery diagnosis
delivery concentration diagnosis
audience concentration
placement concentration
creative starvation
conversion volume optimisation
conversion value optimisation
testing environments
scaling environments
mature campaign management
profitability protection
commercial quality monitoring
economic value monitoring

This framework does not govern:

final capital approval
final financial survivability decisions
offer approval
creative approval
campaign launch approval
experiment validity by itself
autonomous bidding changes
autonomous budget changes
autonomous scaling
autonomous value rule activation

Those remain governed by Finance Brain, Affiliate Brain, Experimentation Brain, Compliance Brain, Risk Brain, HeadOffice and related systems.

Definition / Rules

Core Principle

Bidding controls shape delivery.

They do not create demand.

They do not repair weak creative.

They do not repair weak funnels.

They do not repair weak offers.

They do not create trustworthy tracking.

They do not create commercial viability.

A bidding control may improve apparent efficiency while reducing:

delivery volume
learning
reach
customer quality
commercial contribution
creative exploration
audience discovery
placement diversity

Every control must therefore be judged by business outcomes, not platform efficiency alone.

Bidding Control Categories

Ads Brain recognises the following primary bidding control categories.

Open Delivery Controls

Examples:

lowest cost
maximise conversions
maximise conversion value

Purpose:

allow broader platform optimisation within budget.

Constraint Based Controls

Examples:

target CPA
cost cap
target ROAS
bid cap
manual bid limits

Purpose:

restrict delivery according to a cost, value or bid boundary.

Value Adjustment Controls

Examples:

value rules
audience value adjustments
geography value adjustments
device value adjustments
customer type adjustments

Purpose:

change the relative importance of certain conversions or users.

Open Delivery Control Governance

Lowest Cost

Purpose:

obtain the most available conversion volume within budget without a strict acquisition constraint.

May suit:

early data accumulation
creative validation
campaign learning
situations with limited conversion history
situations where unit economics remain comfortably inside tolerance

Risks:

cost volatility
easy conversion concentration
weak commercial quality
poor marginal economics
overdelivery into low quality inventory

Maximum Conversions

Purpose:

increase the number of declared conversion events.

May suit:

relatively similar conversion values
stable event quality
strong need for conversion volume
sufficient budget
reliable event tracking

Risks:

low quality conversion concentration
weak buyer quality
lower revenue or margin
easy event bias
weak repeat value

Maximum Conversion Value

Purpose:

increase reported conversion value.

May suit:

materially different transaction values
reliable value tracking
sufficient conversion history
higher value customers
mature campaign data

Risks:

lower conversion volume
reported value inflation
weak margin
poor cash timing
customer concentration
value signal instability

Open Delivery Rule

Open delivery controls are not ungoverned.

Ads Brain must still monitor:

acquisition cost
qualified conversion quality
customer quality
placement mix
audience mix
creative exposure
marginal economics
platform reporting variance
commercial contribution

Constraint Based Control Governance

Target CPA

Purpose:

guide the platform toward a desired acquisition cost.

Prerequisites:

known allowable acquisition cost
reliable conversion event
sufficient conversion history
stable tracking
realistic target
acceptable underdelivery risk

Risks:

underdelivery
audience concentration
placement concentration
creative starvation
weak exploration
target chasing without customer quality

Cost Cap

Purpose:

constrain average acquisition cost while allowing some flexibility in individual auction outcomes.

Prerequisites:

known unit economics
sufficient campaign history
stable conversion quality
strong creative
clear rollback rule
acceptable delivery tradeoff

Risks:

reduced volume
slower learning
concentration
unstable delivery
apparent efficiency with weak contribution

Target ROAS

Purpose:

guide the platform toward a desired reported return on ad spend.

Prerequisites:

reliable conversion values
sufficient purchase history
known margin
known break even ROAS
known target ROAS
trusted attribution
Finance Brain review

Risks:

lower volume
overconcentration in high value segments
reported value inflation
weak margin
reduced discovery
poor customer diversification

Bid Cap

Purpose:

limit the maximum auction bid.

Prerequisites:

strong auction understanding
stable historical evidence
known delivery baseline
explicit use case
specialist review where required

Risks:

severe underdelivery
unstable reach
narrow auction access
weak learning
excessive operational complexity

Manual Bidding

Purpose:

apply direct bid logic where platform and campaign conditions justify it.

Prerequisites:

mature data
specialist competence
known auction dynamics
active monitoring
clear rollback
operational capacity

Risks:

overcontrol
poor bid calibration
delivery loss
complexity
false precision

Constraint Control Rule

Constraint based controls should normally be introduced after the campaign demonstrates structural viability.

They must not be used as a first response to:

weak creative
poor conversion rate
unclear offer
low buyer quality
broken tracking
weak landing pages
insufficient demand

Activation Preconditions

Before any material bidding control is activated, the following should be documented.

Commercial Preconditions

allowable acquisition cost
break even acquisition cost
target acquisition cost
break even ROAS
target ROAS
contribution margin
payout
refund exposure
cancellation exposure
fulfilment cost
customer quality
cash timing

Data Preconditions

tracking integrity
conversion event quality
value accuracy
attribution confidence
conversion lag
historical conversion volume
platform versus source of truth variance

Campaign Preconditions

campaign role
creative graduation state
creative depth
audience size
placement readiness
geography
funnel stability
offer stability
current delivery baseline

Risk Preconditions

underdelivery tolerance
allowable learning slowdown
audience concentration tolerance
placement concentration tolerance
creative starvation risk
rollback rule
review date
decision authority

Activation Gate Rule

No bidding control should be activated without enough evidence to explain:

why the control is needed
what success looks like
what failure looks like
what delivery distortion is acceptable
when the control must be removed

Campaign Role Logic

Creative Validation

Preferred approach:

avoid aggressive cost constraints that prevent fair learning.

Primary objective:

generate interpretable evidence.

Risk:

new concepts may be starved before receiving enough delivery.

Offer Validation

Preferred approach:

use controls only after conversion quality and offer economics are understood.

Primary objective:

confirm commercial viability.

Risk:

cheap conversion events may hide weak buyer quality.

Prospecting

Preferred approach:

balance volume, qualified acquisition and marginal economics.

Primary objective:

expand new customer or lead acquisition.

Risk:

constraint based controls may narrow discovery.

Scaling

Preferred approach:

use controls only where creative graduation, economics and delivery history support them.

Primary objective:

protect marginal economics during expansion.

Risk:

overconstraint may stop scale rather than improve it.

Retargeting

Preferred approach:

monitor frequency, audience size and incremental value.

Primary objective:

capture warm demand efficiently.

Risk:

platform reported ROAS may be inflated by demand that would have converted anyway.

Retention

Preferred approach:

optimise for repeat value and customer quality rather than raw conversion count.

Primary objective:

increase economically useful repeat behaviour.

Risk:

conversion value may not reflect true incremental value.

Catalogue Delivery

Preferred approach:

account for product margin, stock and product level economics.

Primary objective:

increase profitable product discovery.

Risk:

platform may favour high revenue but low margin products.

Reactivation

Preferred approach:

monitor list quality, recency and consent.

Primary objective:

recover inactive customer or lead value.

Risk:

cheap reactivation events may carry weak long term value.

Promotion Support

Preferred approach:

distinguish temporary promotional urgency from durable campaign performance.

Primary objective:

support time bound demand.

Risk:

cost control performance may collapse after promotion ends.

Conversion Volume Versus Conversion Value Governance

Ads Brain must distinguish between:

conversion count
qualified conversion count
revenue
net revenue
gross margin
contribution
customer quality
repeat value
refund adjusted value
cancellation adjusted value
cash timing

Conversion Volume Rule

More conversions are only useful where the event represents meaningful business value.

Conversion Value Rule

Higher reported conversion value is only useful where the value is accurate, economically meaningful and sufficiently durable.

Economic Value Rule

Finance Brain determines whether reported value represents true economic value.

Optimisation Objective Selection Record

Record:

campaign role
selected optimisation objective
reason
historical conversion volume
event quality
value quality
commercial goal
expected tradeoff
risk
review date

Value Rule Governance

Value rules or similar adjustments may be used to tell the platform that certain conversions or users have different value.

Possible adjustment dimensions include:

customer type
geography
device
audience
product
margin
repeat value
lead quality

Value Rule Preconditions

reliable source data
clear economic rationale
documented baseline
sufficient sample
known adjustment logic
Finance Brain review where material
rollback rule
monitoring plan

Value Rule Risks

artificial value inflation
audience concentration
delivery distortion
lower total contribution
weak learning
biased reporting
poor customer diversification

Value Rule Boundary

Value rules must represent real business value differences.

They must not be used to manufacture platform performance.

Underdelivery Diagnosis

Underdelivery may result from:

unrealistic cost target
unrealistic ROAS target
low bid cap
insufficient audience
weak creative
low conversion volume
poor event quality
restrictive targeting
low placement eligibility
seasonal demand change
platform instability
limited budget
conflicting controls

Underdelivery Diagnosis Sequence

Confirm tracking integrity.
Confirm offer and funnel stability.
Confirm creative quality.
Confirm audience size.
Confirm placement eligibility.
Confirm conversion volume.
Review active bidding controls.
Review recent campaign changes.
Review seasonality and auction pressure.
Determine whether rollback is required.

Underdelivery Rule

Underdelivery is not automatically proof that demand is weak.

It may be a control failure.

Delivery Concentration Diagnosis

A bidding control may concentrate delivery across:

audiences
placements
demographics
geographies
devices
creatives
days
hours
products
customer segments

Concentration may create apparent efficiency while reducing:

discovery
scale capacity
customer diversity
test validity
creative development
marginal durability

Concentration Review

Record:

spend distribution
impression distribution
reach distribution
conversion distribution
qualified conversion distribution
commercial contribution distribution
creative distribution
placement distribution
audience distribution
geography distribution

Concentration Rule

Concentration is acceptable only where it remains compatible with:

campaign role
testing purpose
customer quality
commercial contribution
risk tolerance
scaling durability

Creative Starvation Protection

Bidding controls and platform optimisation may starve:

new concepts
iteration challengers
placement adaptations
low history creative
new audience tests

Creative Starvation Indicators

negligible spend
negligible impressions
no conversion opportunity
known winner consumes most delivery
platform repeatedly avoids challengers
learning remains impossible

Creative Protection Rule

A creative test is invalid where the delivery system prevents the hypothesis from receiving enough opportunity to be assessed.

Ads Brain may need to separate:

scaled winners
new concept tests
iteration tests
placement tests

Platform Reporting Boundary

Platform reported results must be reconciled against business outcomes.

Compare:

platform conversions
source of truth conversions
qualified leads
sales
customers
gross revenue
net revenue
contribution
refunds
cancellations
repeat value
payout
margin
cash timing

Attribution Boundary

A bidding control may appear successful because of:

attribution window
view through conversion
modelled conversions
cross device attribution
duplicate claims
conversion lag

Platform reporting must not be accepted without reconciliation where the decision is commercially material.

Monitoring Requirements

Every active bidding control should be monitored for:

spend
delivery volume
underdelivery
conversion volume
qualified conversion volume
reported conversion value
source of truth value
CPA
ROAS
marginal CPA
marginal ROAS
marginal contribution
customer quality
audience concentration
placement concentration
creative exposure
frequency
fatigue
tracking confidence
attribution confidence

Monitoring Cadence

Cadence should reflect:

spend rate
conversion lag
campaign maturity
control strictness
capital exposure
operational capacity
risk

A control with high capital exposure requires more frequent review.

Rollback Governance

Every bidding control must have a rollback rule.

Possible rollback triggers include:

severe underdelivery
unacceptable volume loss
customer quality decline
commercial contribution decline
audience concentration
placement concentration
creative starvation
tracking failure
attribution confidence decline
marginal economics breach
operational instability
control no longer matches campaign role

Rollback Options

remove the control
loosen the target
return to open delivery
pause the campaign
separate creative testing
restore prior budget
restore prior optimisation objective
investigate tracking
investigate funnel
investigate offer
investigate auction conditions

Rollback Record

Record:

control
activation date
baseline
trigger
observed effect
rollback action
result
next review
decision authority

Bidding Control Decision Record

Each control decision should record:

platform
campaign
campaign role
control type
optimisation objective
target
current budget
historical spend
historical conversions
qualified conversions
reported value
source of truth value
allowable acquisition cost
break even acquisition cost
target acquisition cost
break even ROAS
target ROAS
contribution margin
customer quality
refund risk
cancellation risk
cash timing
creative graduation state
creative depth
audience size
placement readiness
tracking confidence
attribution confidence
underdelivery tolerance
concentration risk
creative starvation risk
expected benefit
expected tradeoff
rollback rule
review date
decision
decision authority
Brain routing

Bidding Governance Scorecard

Score each category from 1 to 5:

campaign role clarity
optimisation objective clarity
tracking integrity
conversion event quality
value data quality
historical conversion sufficiency
unit economics clarity
customer quality
creative graduation readiness
creative depth
audience sufficiency
placement readiness
underdelivery tolerance
concentration risk
creative starvation risk
marginal economics
cash safety
operational capacity
monitoring readiness
rollback readiness
governance clarity
overall activation readiness

The scorecard supports judgement.

It does not replace judgement.

Relationship To Ads Brain Platform Intelligence

Platform Intelligence explains:

how platforms optimise
how delivery concentrates
how learning systems behave
how bidding changes delivery
how platform reporting differs from business truth

This framework governs:

whether a bidding control may be activated
what evidence is required
how it must be monitored
when it must be rolled back

Relationship To Ads Brain Scaling Intelligence

Scaling Intelligence determines whether a campaign is ready for controlled expansion.

This framework governs whether bidding constraints are suitable during that expansion.

Cost controls do not create scaling readiness.

Relationship To Ads Brain Creative Testing Structure Framework

Creative tests require sufficient and fair delivery.

Bidding controls must not invalidate creative testing through starvation or concentration.

Relationship To Ads Brain Creative Iteration Engine

Iteration relies on interpretable source versus challenger evidence.

Bidding controls must not prevent challengers from receiving enough delivery.

Relationship To Ads Brain Creative Signal Interpretation Framework

Signal Interpretation determines whether apparent efficiency reflects:

persuasion strength
delivery bias
audience concentration
placement concentration
attribution distortion
true commercial value

Relationship To Finance Brain

Finance Brain governs:

allowable acquisition cost
break even acquisition cost
target acquisition cost
break even ROAS
target ROAS
contribution margin
capital exposure
cash safety
economic value

Ads Brain must not override Finance Brain economic authority.

Relationship To Experimentation Brain

Experimentation Brain governs:

test validity
evidence quality
comparison discipline
low volume interpretation
control versus challenger fairness

Relationship To Affiliate Brain

Affiliate Brain governs:

offer eligibility
traffic permission
payout quality
program durability
commission structure
compliance with affiliate terms

Relationship To Data Brain

Data Brain supports:

tracking integrity
value accuracy
source of truth comparison
attribution reconciliation
conversion quality analysis
reporting consistency

Relationship To Compliance Brain

Compliance Brain governs:

platform policy
claims
restricted categories
audience use
consent
data use

Relationship To Risk Brain

Risk Brain governs:

account exposure
platform concentration
capital risk
operational risk
brand risk

Relationship To HeadOffice

HeadOffice resolves:

cross Brain conflict
high consequence exceptions
strategic control decisions
governance escalation

Failure Modes Prevented

This framework prevents:

using cost controls to rescue weak creative
using cost controls to rescue weak funnels
using cost controls to rescue weak offers
using cost controls without known unit economics
using target ROAS with unreliable value data
using target CPA with weak event quality
treating underdelivery as proof of weak demand
starving new creative tests
confusing apparent efficiency with commercial value
ignoring audience concentration
ignoring placement concentration
ignoring creative concentration
using value rules without economic support
scaling controls without rollback rules
ignoring customer quality
ignoring refunds and cancellations
relying on platform reporting alone
treating lower CPA as automatic improvement
overconstraining campaigns until learning stops
allowing bidding complexity to exceed operational capacity

Drift Protection

The system must prevent:

cost controls becoming default campaign settings
arbitrary targets being entered without evidence
platform terminology becoming permanent Canon architecture
Finance Brain economic authority being bypassed
creative testing being invalidated by starvation
underdelivery being accepted without diagnosis
reported ROAS being treated as profit
conversion count being treated as customer quality
value rules being used to manipulate reporting
rollback rules being omitted
platform settings being copied blindly across campaigns
one platform method being generalised across all platforms
bidding controls remaining active after conditions change

Governance Boundaries

This framework does not authorise:

autonomous target CPA activation
autonomous cost cap activation
autonomous target ROAS activation
autonomous bid cap activation
autonomous manual bidding
autonomous value rule activation
autonomous budget changes
autonomous campaign duplication
autonomous scaling
autonomous placement exclusion
autonomous audience exclusion
automatic MCR updates
technical development

Every material action remains subject to human review and the authority of the relevant Brain.

Dynamic Platform Boundary

Platform bidding features, names, defaults and mechanics may change.

Operational playbooks may record current implementation.

This framework preserves durable governance logic.

Tool Specific Boundary

Named platform features may be referenced operationally.

They do not become permanent architectural dependencies.

Architectural Intent

The Ads Brain Cost Control And Bidding Governance Framework exists to ensure MWMS uses platform bidding controls as disciplined commercial tools rather than as substitutes for strategy.

Its role is to:

align bidding with campaign purpose
protect test validity
protect commercial quality
protect capital
prevent delivery distortion
preserve creative learning
enforce monitoring
enforce rollback
maintain cross Brain governance

Cost controls can improve efficiency.

Used badly, they can suppress learning, hide weakness and reduce total business value.

MWMS therefore treats every bidding control as a governed intervention.

Version History

Version: v1.0

Date: 2026-07-19

Author: MWMS HeadOffice

Change:

Created Ads Brain Cost Control And Bidding Governance Framework using the strongest non duplicative bidding and cost control intelligence absorbed from Sam Piliero The Facebook Ads Blueprint.

Added:

Core Principle
Bidding Control Categories
Open Delivery Controls
Constraint Based Controls
Value Adjustment Controls
Lowest Cost governance
Maximum Conversions governance
Maximum Conversion Value governance
Target CPA governance
Cost Cap governance
Target ROAS governance
Bid Cap governance
Manual Bidding governance
Activation Preconditions
Activation Gate Rule
Campaign Role Logic
Conversion Volume Versus Conversion Value Governance
Optimisation Objective Selection Record
Value Rule Governance
Value Rule Preconditions
Value Rule Risks
Value Rule Boundary
Underdelivery Diagnosis
Underdelivery Diagnosis Sequence
Delivery Concentration Diagnosis
Creative Starvation Protection
Platform Reporting Boundary
Attribution Boundary
Monitoring Requirements
Monitoring Cadence
Rollback Governance
Rollback Options
Rollback Record
Bidding Control Decision Record
Bidding Governance Scorecard
Cross Brain Relationships
Failure Modes Prevented
Drift Protection
Governance Boundaries
Dynamic Platform Boundary
Tool Specific Boundary

Clarified:

bidding controls shape delivery but do not create demand
cost controls cannot repair weak creative, funnels, offers or tracking
lower CPA does not automatically mean stronger commercial performance
conversion volume and conversion value require different evidence
platform value is not automatically economic value
underdelivery requires diagnosis
audience, placement and creative concentration must be monitored
new concepts must be protected from delivery starvation
every material control requires monitoring and rollback
Finance Brain retains economic authority
platform features remain dynamic operational intelligence

Pages Created:

Ads Brain Cost Control And Bidding Governance Framework

Pages Updated:

None

Pages Deprecated:

None

Registry Requiring Update:

Ads Brain Page Registry

Required Registry Change:

Add Ads Brain Cost Control And Bidding Governance Framework as an Active Ads Brain Framework at version v1.0 under Ads Brain Canon.

System Map Update Required:

Yes

Required System Map Change:

Add the framework between Ads Brain Platform Intelligence, Ads Brain Scaling Intelligence, Ads Brain Creative Testing Structure Framework, Ads Brain Creative Signal Interpretation Framework, Finance Brain, Experimentation Brain, Affiliate Brain, Data Brain, Compliance Brain and Risk Brain.

Canon Version Update Required:

Change Log Entry Required:

Yes

END OF FULL FILE OUTPUT