The Celerity Investment Framework

Abstract

Investment research often treats business quality, valuation,
financing and risk as adjacent topics rather than as one controlled
system. CIF v2.0 integrates them. It begins with external economic,
geopolitical, technological and industry state; identifies enterprise
capabilities that can repeatedly produce useful outcomes; classifies the
economic engine through which those capabilities must operate; tests
economic conversion; independently tests Financial Viability; closes
capital-allocation cohorts at realised owner return; measures
Shareholder Value Capture; values the resulting states; determines
portfolio role; and then connects, where relevant, to CPSR and
CMVTR.

The principal architectural advance since v1.2 is the Company Agent
layer. Identical observations can have opposite meanings when the
underlying economic mechanism differs. Negative working capital can be
structural float or creditor stress. Development expenditure can be
rational milestone capital or evidence that a mature enterprise cannot
self-fund. Deposits and insurance liabilities can be productive
balance-sheet inputs. A finite concession cannot be valued as an
unearned perpetuity. Conventional industrial free cash flow is therefore
not a universal owner-value measure.

CIF remains standardised where comparability is useful and adaptive
where causality demands it. Nine stable policy dimensions preserve
comparability. Ten Economic Engine Modules are frozen at top level.
Company-specific sub-architectures, overlays, capability states,
owner-cash definitions, viability gates and falsification tests reflect
the actual productive system. The persistent Company Agent State File
becomes the canonical company research record.

Executive Summary

Canonical sequence	External State -> Economic Engine Architecture -> Capability -> Execution -> Archetype-Specific Economic Conversion -> Financial Viability -> Capital Allocation Closure -> Shareholder Value Capture -> Valuation -> Portfolio Decision -> CPSR / CMVTR -> Monitoring -> New Evidence -> State Delta.

CIF is not a score or target-price generator. Its purpose is to make
explicit the causal chain that must hold for an enterprise to create
durable per-share value, determine what the market already pays for that
chain, and preserve the ability to change the decision when evidence
changes.

The nine-dimensional CIF scorecard remains a consistency and
comparability device. It cannot override a binding constraint, Financial
Viability failure, value-destructive capital cohort, broken
shareholder-capture mechanism, severe Decision Risk, inadequate evidence
or valuation that fails the required-return test.

THE CELERITY INVESTMENT FRAMEWORK

A controlled sequence with Decision Risk across the system and
monitoring as the feedback loop

EXTERNAL CONTEXT | Economic | Geopolitical & Sovereign | Technology & Infrastructure | Industry Structure

Capability What can the enterprise repeatedly do?	Company Agent Architecture What economic engine?	Economic Confirmation What does capability earn?	State-Based Valuation What state is it worth?	Portfolio Construction What exposure is rational?	Celerity Portfolio State & Resilience Research (CPSR)	Valuation, Volatility and the Price of Protection (CMVTR)

DECISION RISK OVERLAY | base rates | anchoring | narrative capture | financing & dilution | evidence transfer | double counting

CONTINUOUS EVIDENCE & MONITORING | baseline -> new evidence -> state delta -> re-underwrite -> new persistent baseline

Figure 1. The Celerity Investment Framework: a controlled
sequence with Decision Risk across the system and monitoring as the
feedback loop.

Part I – Purpose and first principles

1. What the
framework is designed to do

CIF is designed for long-horizon security selection and portfolio
management where the central problem is not merely whether an enterprise
is good, but whether it can form, operate, compound and renew
economically valuable capability through an economic architecture that
converts those capabilities into durable value for continuing
shareholders at a price offering an acceptable forward return.

It answers seven connected questions: what is changing outside the
company; what can the enterprise repeatedly do; what economic engine
converts that capability; can the enterprise finance the path to
closure; does reinvestment earn its cost; how much value reaches
continuing shareholders; and what portfolio action is justified given
uncertainty, path risk and alternative uses of capital?

2. The core investment
question

Core investment question	Does this security combine an understandable external environment, a durable capability system, the correct economic conversion architecture, Financial Viability, disciplined capital allocation, genuine Shareholder Value Capture, a valuation that compensates for uncertainty, an appropriate portfolio role and a decision process robust enough to change when the evidence changes?

This deliberately separates the enterprise from the security. An
exceptional enterprise can be a poor investment at an excessive price. A
capable enterprise can still be a poor security if repeated financing
transfers value to new capital providers. A highly capable cyclical
company can still suffer damaging market paths.

3. Seven operating
principles

Principle	Meaning
Capability precedes performance	Reported outcomes are lagging evidence.
Economic architecture precedes metric interpretation	A metric has no universal capability meaning until the causal engine is identified.
Economic transmission is mandatory	Capability must connect to productive outcome, owner economics and falsification.
Financial Viability is independent	Strategic capability does not guarantee ability to finance the path to closure.
Capital allocation closes at owner return	Completion, revenue and announced synergy are intermediate evidence.
Shareholder Value Capture is explicit	Enterprise improvement counts only if durable owner value per economic diluted share improves.
Valuation is conditional; research is persistent	Fair value belongs to a state; historical assumptions and decisions are versioned.

Part II – The architecture

4. External context

The framework begins outside the company. Economic conditions,
geopolitical and sovereign conditions, technology and infrastructure
cycles, and industry structure shape the opportunity set and constraints
under which capability must operate. These are translation layers into
company economics, not stand-alone macro forecasts.

Context layer	Core question	Typical transmission
Economic	What regime affects demand, financing, inflation and discount rates?	Revenue, pricing, wages, credit, working capital, multiple and required return.
Geopolitical / sovereign	Can state power, regulation, security or access change economics?	Market access, licences, taxes, supply chains, industrial support, sanctions, social licence.
Technology & infrastructure cycles	Where is the system in invention, build-out, adoption, consolidation or profit capture?	Demand, capex, obsolescence, productivity, bottlenecks and competitive redistribution.
Industry structure	Who has bargaining power and can returns persist?	Margins, entry, substitutes, capacity discipline and rent capture.

5.
Capability Science as the leading-indicator engine

Capability Science changes the unit of analysis. Instead of asking
only what assets, products or ratios the company possesses, research
asks what the organised enterprise can repeatedly produce under changing
conditions. Capability exists when people, knowledge, technology,
governance, relationships, processes and capital are integrated into
repeatable productive capacity.

Assessment field	Controlled values	Purpose
State	Forming / Operational / Compounding / Renewing / Decaying	Lifecycle state.
Trajectory	Improving / Stable / Deteriorating	Direction of travel.
Evidence confidence	Weak / Moderate / Strong	Strength of evidence, separate from capability quality.
Evidence class	Formation / Execution / Compounding / Decay	What kind of evidence has been observed.
Leverage	Low / Medium / High or qualitative	Whether change affects many other outcomes.
Constraint status	None / Possible / Confirmed	Whether it limits the wider enterprise system.

6.
Economic Engine Architecture and the Company Agent

The Company Agent Architecture supplies the causal layer between
organisational capability and investor economics. Before interpreting
financial statements, the analyst declares the economic engines present,
their relevance, shared capabilities, central costs, overlays, binding
constraint, owner-cash definition, independent viability gate, valuation
attribution and falsification register.

The Company Agent is a context-sensitive state architecture, not a
named-company template. Calibration contexts include a stressed loyalty
intermediary; a hit-driven project/IP studio; a federated diagnostic
network; a commercial radiopharma platform; a dense grocery retail
system; an insurer; a bank; a toll-road concession portfolio; a
pre-commercial medical-device developer; a technology distributor; a
diversified miner; a telecommunications network; and a recurring
healthcare-software business. Actual company names are deliberately
omitted from this Company Agent state discussion.

Configuration field	Required content
Primary engine(s)	One or more frozen modules.
Engine relevance	Revenue, gross profit, owner cash or strategic role; do not weight blindly by revenue.
Sub-architecture	Validated subtype where relevant.
Shared capabilities	Capabilities genuinely serving multiple engines.
Central/shared costs	Costs not already embedded in engine economics.
Overlays	Only those changing causal economics or evidence interpretation.
Binding constraint	Capability or external condition most likely to determine the next state.
Owner-cash definition	Archetype-specific formula stated before valuation.
Independent viability gate	Auditor, regulator, capital/liquidity, runway, covenant or equivalent evidence.
Valuation attribution	Which engine/capability creates each component of intrinsic value.
Falsification register	Observable state-change evidence ranked by valuation leverage.

7. The ten
frozen Economic Engine Modules

#	Economic Engine Module	Core conversion
1	Recurring Software / Data / Subscription	Recurring revenue -> gross profit (GP) -> contribution -> owner cash after maintenance development.
2	People-Led Services / Professional Capability	Expertise -> contribution -> GP/employee -> owner cash via utilisation and knowledge compression.
3	Merchant / Distributor / Intermediary	Volume -> retained spread/GP -> normalised working capital/finance -> owner cash.
4	Marketplace / Transaction Network	total transaction value / gross merchandise value (TTV/GMV) -> realised take rate -> contribution -> settlement/WC -> owner cash.
5	Replicated Unit / Site & Production Network	Network density -> throughput/productivity -> return on capital (ROC) / free cash flow to equity (FCFE) after replication and maintenance capital.
6	Asset-Heavy Cyclical / Resource	Price/volume/cost -> operating cash flow (OCF) -> owner cash after sustaining/replacement capability capital.
7	Contracted / Regulated Infrastructure	Contract/concession cash -> recurring Free Cash after lifecycle capital/financing and finite-right replacement.
8	Financial Intermediation / Balance-Sheet	Risk-adjusted spread/underwriting/investment return -> distributable capital after regulatory/economic capital.
9	Project / IP Portfolio / Hit-Driven	Project/title cohorts -> catalogue/franchise cash -> portfolio owner return after development capital.
10	Development / Pre-Cash-Flow Innovation	Milestone de-risking -> probability-weighted commercial value less runway/dilution to economic closure.

The top-level taxonomy is frozen. New sub-architectures may be added
where necessary, but a new primary module requires evidence that a
material economic engine cannot be represented causally by any existing
module without distorting its owner-cash definition.

8. The Celerity policy
scorecard

The capability map and economic-engine architecture are
company-specific, but CIF retains a stable nine-dimension policy
scorecard to preserve comparability. These are policy weights, not
empirically discovered coefficients, and change only through framework
version control.

CIF dimension	Weight
Economic	7.5%
Geopolitical / Sovereign	7.5%
Technology & Infrastructure Cycles	10.0%
Industry Structure	10.0%
Competitive Power	15.0%
Compounder Quality	20.0%
Management & Culture	10.0%
Innovation & Finance	7.5%
Valuation & Portfolio Construction	12.5%
Total	100.0%

The score is a consistency device, not a buy/sell rule and not a
substitute for causal company architecture.

9.
Capability-to-economics transmission

Capability research becomes investment research only when the
transmission mechanism is explicit. The required chain is capability
change -> repeatable productive outcome -> archetype-specific
economic consequence -> investment consequence. Activity growth is
never a universal capability state.

CAPABILITY-TO-ECONOMICS TRANSMISSION

Investment & resource inputs	Capability formation	Execution	Economic outcome	Owner economics	Shareholder capture	Valuation

FINANCIAL VIABILITY & CAPITAL GATES | working capital | maintenance capital | regulatory capital | financing | dilution | cohort closure

A capability system is investable only when transmission is observable and falsifiable.

Figure 2. Capability-to-economics transmission: the Company Agent
adds causal economic architecture, Financial Viability and
shareholder-capture gates without replacing Capability Science.

Economic context	Primary conversion bridge
Merchant / distributor	Revenue -> gross profit -> working capital / finance -> owner free cash flow.
Marketplace	Transaction value -> realised take rate -> contribution -> settlement/working capital -> owner cash.
Project / IP	Project capital -> launch cash -> catalogue/franchise cash -> cohort return.
Development innovation	Cash burn -> milestone probability uplift -> commercial transition -> dilution-adjusted owner value.
Insurance	Premium/risk -> underwriting result + float return -> required capital -> distributable capital.
Banking	Deposits/assets -> net interest margin (NIM) / fees – credit cost – required Common Equity Tier 1 (CET1) -> distributable earnings.
Infrastructure	Traffic/contract cash -> lifecycle capital + financing -> recurring Free Cash over finite rights.
Resource	Realised price/volume/cost -> OCF -> sustaining/replacement/growth capital -> owner free cash flow.

For every material capability, the analyst states what productive
outcome must be produced, how that outcome will be observed, which
economic variable it should change, what capital is required and what
evidence would falsify the claimed transmission. A capability map that
cannot be translated into economics remains organisational description,
not investment analysis.

10. Financial
Viability

Financial Viability is an independent capability gate. It asks
whether the enterprise can reach its next economic closure point without
value-destructive financing. The relevant evidence differs by economic
engine. A mature commercial enterprise that repeatedly needs discounted
equity to fund ordinary operations is not equivalent to a
development-stage enterprise deliberately consuming funded runway to
reach a defined value-inflecting milestone.

Viability state	Interpretation
Self-funded	Ordinary operations and necessary capability renewal are funded internally.
Planned externally funded transition	External capital can be rational where it finances a defined milestone or project whose expected value exceeds financing and dilution cost.
Capital-dependent	Repeated external funding remains necessary despite commercial scale or stated maturity.
Stress-funded	Creditors, emergency debt or discounted equity finance ordinary survival; Financial Viability is downgraded.

Working capital is classified causally rather than by accounting
sign. Negative working capital can be Structural Float when customers
pay immediately and supplier terms are ordinary and durable; Growth
Funding when liabilities scale normally with activity; or Stress Funding
when creditors finance survival because liquidity or supplier quality
has deteriorated. Auditor going-concern language, covenant pressure,
regulatory capital, liquidity coverage, runway and supplier terms are
independent evidence gates where relevant.

11. Capital-allocation
closure

CIF v2.0 separates execution success from capital-allocation success.
An acquisition, project, automation programme, development title,
concession, clinical programme or transformation initiative remains an
open capital cohort until total economic capital, integration or
transition cost, working capital, financing and dilution are reconciled
against realised owner return.

Early evidence	Capital-allocation closure
Project completed / product launched	Cumulative owner cash and retained intrinsic value justify full economic capital.
Synergies announced or achieved	Post-tax incremental return on total capital closes above the required return.
Revenue grows after investment	Contribution and owner cash exceed incremental capital and funding burden.
Acquired capability survives	Shareholder value after purchase price, integration, financing and dilution is positive.

This rule prevents research from declaring victory because capability
exists inside the enterprise while the capital used to acquire or build
that capability has not produced an adequate return for owners.

12. Shareholder Value
Capture

Shareholder Value Capture is now an explicit CIF stage. Enterprise
value creation is insufficient. The framework asks how much value
accrues to the continuing shareholder after economically necessary
working capital, maintenance and growth capability capital, financing
claims, regulatory capital, options, convertibles, contingent
consideration and dilution.

Capture state	Interpretation
Positive capture	Enterprise value and owner cash rise faster than the economic diluted claim denominator.
Neutral capture	Enterprise growth is largely absorbed by new capital claims.
Negative capture	Economic diluted claims grow faster than enterprise value or owner cash.
Endogenous dilution	Future financing need is part of the valuation state, not an afterthought.

This stage closes a gap that can otherwise survive both capability
analysis and enterprise valuation: the organisation may improve while
the continuing shareholder receives little benefit because repeated
capital formation transfers value to new claimants.

13. Valuation is a
state, not a clock

Celerity valuation remains conditional on an economic state. Forecast
years organise a model, but time itself does not create value. The model
specifies the operating conditions, capability evidence, economic-engine
conversion, capital requirements, Financial Viability and economic claim
denominator that make forecast owner economics possible.

Valuation discipline	Required treatment
Current state	Use current economics and current evidence; do not smuggle future success into the base.
Central future state	State capability, engine economics, volume, price, margin, capital, financing and diluted claims required.
Downside state	Identify capability, viability or external failures causing permanent or prolonged impairment.
Upside state	Require stronger but observable evidence; do not capitalise narrative optionality by default.
Market-implied state	Reverse-engineer what current price assumes across engine and state transitions.
Entry-value discipline	Translate required returns back into the price compatible with central/downside economics.

Method selection is company-specific and archetype-specific.
Discounted cash flow (DCF), reverse DCF, Owner Free Cash Flow (Owner
FCF), residual income, distributable capital, concession-duration cash
flow, asset value, dividend economics and risked program net present
value (NPV) are tools rather than doctrines. Conventional industrial
free cash flow is not imposed on banks or insurers; finite concession
rights are not treated as unearned perpetuities; development-stage value
incorporates runway and expected dilution; capital required to preserve
and renew productive capability is not casually added back.

14. Decision Risk
Overlay

Decision Risk applies across every layer. In addition to base-rate
neglect, anchoring, narrative capture, false precision, recency bias and
confirmation bias, v2.0 adds controls for activity/economics confusion,
management-representation reset, financing and dilution omission,
evidence transfer between economic engines and value double
counting.

Decision-risk question	Control
Are we using base rates?	Compare with relevant prior cycles, contexts and failure modes.
Are we anchored to market price?	Build valuation independently and use reverse valuation separately.
Are we seduced by management narrative?	Maintain a Management Representation Ledger with dated closure tests.
Are we mistaking activity for economics?	Require archetype-specific conversion.
Are we ignoring financing or dilution?	Run independent Financial Viability and claim-denominator tests.
Are we transferring evidence across engines?	Evidence upgrades only the capability/engine that causally produced it unless transfer is separately proven.
Are we double counting value?	Use a Valuation Attribution Ledger.
Are we rewriting history?	Persistent baseline and state-delta versioning.

Part III – Portfolio, resilience and path risk

15. Portfolio
construction

Portfolio construction asks a different question from security
selection: how much exposure is rational given the security, the
evidence and the rest of the portfolio? The same enterprise can justify
different portfolio weights at different prices or in different
portfolio states. Sizing considers classification, evidence quality,
valuation asymmetry, downside severity, Financial Viability, liquidity,
correlation, concentration, time horizon, Decision Risk and monitoring
clarity. Exact size bands remain mandate-specific implementation
parameters.

A high CIF score does not automatically justify a large portfolio
weight. High capability with excessive valuation may warrant patience.
Cheap valuation with weak capability or Financial Viability may remain a
special situation. A binding constraint can justify lower exposure even
when aggregate policy scores are high.

16.
Celerity Portfolio State & Resilience Research (CPSR)

Celerity Portfolio State & Resilience Research (CPSR) is a
separate controlled publication and methodology. CIF supplies the
company, valuation and portfolio state; CPSR analyses the risk of loss
through Threat and Vulnerability and diagnoses the nature of loss once
it occurs. The purpose is not to duplicate the Company Agent. Company
capability and Financial Viability enter CPSR as vulnerability evidence;
CPSR then asks how external shocks, financing conditions, liquidity,
duration and tail exposures can translate that vulnerability into
loss.

CPSR distinguishes market repricing from state-dependent economic
loss, structural loss and continuity loss. That distinction matters
because a temporary drawdown can leave the long-horizon ownership thesis
intact, while a superficially similar price decline can reflect
permanent deterioration in enterprise economics or financing
capacity.

17.
Valuation, Volatility and the Price of Protection (CMVTR)

Valuation, Volatility and the Price of Protection (CMVTR) remains
Celerity’s controlled methodology for cyclical-equity path risk and
dynamic portfolio protection. It sits downstream of fundamental
ownership, structural valuation and portfolio role. It separates
structural value, physical market risk, the market price of risk
embedded in options and the economics of risk transfer before
reconnecting them in a portfolio decision.

CMVTR is not an autonomous trading engine. Fundamental research
determines whether the business should be owned, state valuation
estimates what it is worth, portfolio construction determines how much
should be owned, and CMVTR assesses the damaging path between price and
value and the economics of transferring that risk. Research-stage or
shadow-stage components are not allowed to masquerade as validated
production signals.

18.
Controlled interfaces and decision rights

System	Primary question	Interface
CIF	Should the security be owned, at what value and in what portfolio role?	Master investment decision architecture.
Company Agent	What economic machine is the enterprise and what state is it in?	Persistent company state, viability, capture and valuation inputs.
CPSR	What is the risk of loss and what kind of loss would it be?	Consumes company vulnerability/capability and external Threat state.
CMVTR	What is the damaging price path and what is the price/economics of protection?	Consumes structural value, portfolio role and validated market-risk inputs.

Human decision rights are retained throughout. Quantitative models
and structured state files create a common language; they do not replace
ownership judgment. Material overrides are recorded so that discretion
itself can later be evaluated.

Part IV – Evidence and robustness

19.
Evidence hierarchy and analytical firewalls

The framework is commercially useful only if another analyst can
understand why a conclusion was reached and what would change it. Each
company research record separates source-derived facts, controlled
Celerity assumptions and Celerity analytical inferences.

Layer	Examples	Control
Source-derived facts	Results, filings, guidance, regulator findings, contracts, announcements.	Citation, publication date, availability date, source authority.
Controlled assumptions	Commodity paths, FX convention, payout, required return, terminal convention, model horizon.	Assumption register and version.
Analytical inferences	Engine classification, capability state, binding constraint, viability state, valuation and thesis consequence.	Explicit Celerity label, uncertainty and falsification condition.

Point-in-time discipline remains especially important in historical
evaluation. Later outcomes cannot leak backward into the assessment. The
frozen Capability Science historical-validation design retains dated
evidence packs, matched comparisons, contradiction protocols and
blinding controls. The live Company Agent architecture does not
retrospectively alter that experiment.

20. Historical
paired-company research

Celerity continues to use historical paired-company assessments as a
research laboratory for the capability layer. The purpose is not to
claim that a small pilot proves predictive power. It is to stress the
method against contemporaneously credible alternatives and identify what
the framework discriminates.

The principal pilot lesson remains that strong enterprises can score
similarly on static capability domains. More discriminating information
appears in trajectory, transferability, adaptability, reinforcement,
capital-allocation direction, whether scale improves the customer
proposition and whether one capability makes the next capability cheaper
or easier to form. This finding remains consistent with the richer
Company Agent architecture.

21. Contemporary
cross-context calibration

Since v1.2, the live framework has been stress-tested across
materially different economic contexts rather than only
telecommunications and resources. The calibration set spans
intermediary/loyalty economics, project/IP, diagnostics, radiopharma,
grocery retail, insurance, banking, toll-road concessions,
pre-commercial medical devices, technology distribution, diversified
resources, telecommunications and recurring healthcare software.

The calibration programme is not presented as statistical proof. Its
purpose was contradiction and boundary testing. The result was that no
additional top-level economic engine was required. Sector-specific
causality was captured through sub-architectures and overlays, while the
ten-module top-level taxonomy remained stable.

22.
Stable policy weights, frozen engines and adaptive company
states

CIF now uses three distinct forms of structure. Stable policy weights
preserve cross-company comparability. Frozen Economic Engine Modules
preserve causal consistency at the business-model level. Adaptive
Company Agent states preserve enterprise specificity. This avoids both
extremes: changing the whole framework for every industry, or forcing
every company into the same capability checklist and owner-cash
definition.

23.
What is evidence, what is design, what remains hypothesis

Category	Examples	Publication treatment
Externally established / broadly supported	Industry structure, competitive advantage, return on invested capital (ROIC) / reinvestment logic, behavioural decision risk, realised-volatility and variance-risk-premium mechanisms.	Foundation with cited literature; company applicability still requires evidence.
Celerity design choice	Capability lifecycle states, stable CIF policy weights, ten-engine taxonomy, Company Agent State File, state valuation, evidence ledger and monitoring cadence.	Operating design evaluated over time.
Open empirical hypothesis	Capability reinforcement predicts exceptional compounding; specific Company Agent signals improve selection; CPSR/CMVTR variables improve risk decisions.	Label as hypothesis until historical and live evidence is sufficient.

Part V – Implementation in practice

24. Coverage
initiation

• Define the investment question and
classification.

• Assemble the source hierarchy and dated evidence
pack, including historical company evidence where it changes present
interpretation.

• Declare Economic Engine Architecture,
sub-architectures and overlays.

• Map the company capability system and identify
state, trajectory, evidence confidence, high-leverage capabilities and
binding constraints.

• Assess external context and the nine CIF policy
dimensions.

• Build the archetype-specific
capability-to-economics bridge.

• Run the independent Financial Viability gate.

• Open material capital cohorts and define closure
tests.

• Build the economic claim/dilution genealogy and
Shareholder Value Capture test.

• Construct downside, central, upside and
market-implied valuation states.

• Apply Decision Risk, determine portfolio role and
define upgrade, downgrade and thesis-break evidence.

• Determine whether CPSR and/or CMVTR are relevant
and what model status is validated for use.

• Freeze the initial Capability & Valuation
Review and Company Agent State File; begin the evidence ledger.

25. The
canonical Company Agent State File

Controlled component	Purpose
Evidence archive / source register	Preserve source provenance and dated evidence.
Economic Engine Architecture	Define primary engines, sub-architectures, shared capabilities, central costs and overlays.
Capability State Matrix	Maintain state, trajectory, confidence, leverage and constraints.
Management Representation Ledger	Preserve promises and dated closure tests rather than allowing narrative reset.
Capital cohort genealogy	Track acquisitions, projects, development and transformations from formation to owner-return closure.
Working-capital / funding classification	Separate Structural Float, Growth Funding and Stress Funding.
Independent Financial Viability gate	Auditor, regulator, capital/liquidity, runway, covenant or equivalent evidence.
Owner-cash definition and economic model	Translate capability and external states through the correct archetype.
Economic claim / dilution genealogy	Measure the continuing shareholder denominator.
Valuation states and Attribution Ledger	Maintain conditional values and prevent DCF/NPV/optionality/multiple overlap.
Decision Risk / falsification register	Document uncertainty, contradictions and valuation-leverage evidence.
Version log	Preserve what was known, assumed and decided at each material state.

26. Monitoring and
the evidence ledger

Monitoring is not a lighter version of research. It is the mechanism
that makes the original research falsifiable. The evidence ledger
records only information that changes capability, economic-engine state,
Financial Viability, capital cohorts, Shareholder Value Capture,
valuation, Decision Risk or portfolio action.

Evidence field	Required entry
Date	When the evidence became available.
Engine / capability / overlay	What part of the causal architecture is affected.
Evidence	Observable fact or disclosure.
Type	Formation / Execution / Compounding / Decay / External-state / Valuation / Viability / Capital.
Direction	Positive / Negative / Mixed / Neutral.
Confidence	Weak / Moderate / Strong.
Economic transmission	Which owner-economic or risk variable should change.
Capital / viability effect	Whether a cohort, runway, funding or claim denominator changes.
Thesis effect	Strengthens / No change / Weakens / Requires model revision.
Action	Monitor / update state / update model / revalue / deep review / portfolio review.

27. Baseline
-> New Evidence -> State Delta

Monitoring does not restart the company at each result. The canonical
operating sequence is Baseline -> Monitor -> New Evidence ->
State Delta -> Re-underwrite the affected causal chain ->
Valuation Delta -> New Persistent Baseline. Unaffected capabilities,
capital cohorts and management representations remain in state. This
preserves organisational memory and prevents management narrative from
resetting the research record.

A reporting event therefore begins with delta detection, not with a
blank-page company review. The analyst asks what actually changed, which
engine or capability it affects, whether the evidence is formation,
execution, compounding or decay, how it transmits into owner economics,
whether Financial Viability or the claim denominator changes, and
whether valuation or portfolio action must change.

28. Review cadence

Review	Default cadence	Purpose / output
Evidence capture	Continuous / event-driven	Record material company, industry, regulatory, management, capital and financing developments.
Company Agent monitoring	Quarterly	Update engine state, capability state, trajectory, confidence, constraints, viability and falsification queue.
Results reconciliation	Half-year/full-year or each reporting event	Reconcile new evidence with operating and owner-economic outcomes.
Valuation refresh	At results and after material state/price change	Update conditional values, required-return analysis and market-implied assumptions.
Full company re-underwrite	At least annually	Rebuild the thesis against the persistent record rather than erasing it.
Portfolio review	Quarterly and before major allocation changes	Check role, concentration, correlation, liquidity and opportunity cost.
CPSR review	As Threat/Vulnerability state changes	Reassess risk of loss and nature of loss.
CMVTR review	Frequency appropriate to validated data/materiality	Update path risk, model status and risk-transfer economics.

29.
Company-specific capability and valuation publications

Publication / record	Purpose
Initiation Capability & Valuation Review	Full enterprise architecture, CIF assessment, model, valuation states, Decision Risk and monitoring plan.
Company Agent State File	Persistent canonical company state; the primary controlled record.
Results & Capability Update	Reconcile new results as a state delta rather than rewrite the whole company.
Valuation State Update	Rebuild the economic state and fair-value range after material evidence or price/value displacement.
Event-Driven Research Note	Assess one thesis-changing event and its causal transmission.
Annual Company Review	Complete re-underwrite and controlled new publication edition.

Capability and valuation can be published as one integrated paper or
coordinated papers. Separate documents must not become separate theses.
They use the same evidence date, Company Agent state, assumptions,
owner-cash definition and upgrade/downgrade conditions.

THE COMPANY RESEARCH OPERATING CYCLE

Publications are versioned outputs from one persistent Company
Agent State File

1 Initiation / full review Freeze capability, valuation and monitoring	2 Evidence ledger Capture material evidence prospectively	3 Results & event reconciliation Reconcile new evidence with causal thesis

ONE CONTROLLED COMPANY RESEARCH RECORD | persistent Company Agent state, evidence, economics and valuation history

6 Annual re-underwrite Rebuild against the persistent record	5 Upgrade / downgrade conditions Make falsification and thesis change explicit	4 Valuation & portfolio decision Refresh conditional value and rational exposure

Figure 3. The company research operating cycle. Publications are
versioned outputs from one persistent Company Agent State File.

30. Event-driven
review

Event-driven review follows a fixed causal sequence: identify the
changed fact; map it to the relevant engine, capability, overlay or
external state; classify the evidence; identify productive and financial
transmission; update Financial Viability, capital cohort or shareholder
claim where relevant; then decide whether valuation or portfolio action
changes. Celerity publishes when evidence changes the decision system,
not merely because the news cycle changes.

31. Portfolio, CPSR and
CMVTR review

Published company research and mandate-specific implementation remain
governed separately. Public research states capability, valuation and
risk conditions. Portfolio construction applies mandate-specific
concentration, liquidity, tax, correlation, cash and risk limits without
altering the underlying thesis. CPSR and CMVTR retain explicit model
status and do not override fundamental ownership judgment.

32. Version control and
governance

Control	Framework rule
Framework version	Architecture, policy weights and core definitions change only through documented CIF versioning.
Company Agent Architecture	Top-level engine taxonomy frozen at v1.0; sub-architectures and overlays evolve under controlled versioning.
Company state version	Material state changes create a new persistent baseline/update; history is not overwritten.
Assumption register	External assumptions are dated and versioned.
Model status	Validated, provisional, shadow or degraded status is visible where relevant.
Evidence date	Facts enter when publicly available, not at the period end to which they relate.
Override record	Material human departures from model or scoring output are recorded with rationale.
Publication independence	Public research conclusions remain independent of mandate-specific portfolio implementation.

Part VI – A scalable research system

33. Standard outputs

Output	Minimum content
Company master record	Identity, classification, engine architecture, evidence sources and version history.
Company Agent State File	Capability states, ledgers, viability, owner cash, claims, valuation attribution and falsification.
CIF scorecard	Nine stable policy dimensions plus Decision Risk.
Financial / economic model	Causal bridge from engine states to owner economics.
Valuation states	Downside, central, upside, reverse valuation and required-return discipline.
Monitoring dashboard	Company-specific leading capability, viability and economic indicators.
Evidence ledger	Dated material evidence and state effect.
Upgrade / downgrade matrix	Pre-specified evidence that strengthens, weakens or breaks the thesis.
Publication record	Latest canonical paper plus dated updates.
Portfolio / CPSR / CMVTR record	Mandate-specific allocation, resilience and path-risk decisions under separate governance.

34.
Commercial robustness without product dependence

The framework remains independent of any future distribution or
investment product. A research method is more durable when evidence
standards, company records, valuation discipline and monitoring do not
need to be redesigned for each delivery format.

Four characteristics make the framework scalable. Comparability:
stable policy dimensions and evidence language across companies.
Specificity: each company retains its own causal engine and capability
architecture. Longitudinal memory: the Company Agent State File,
evidence ledger and version history create a time series of judgments
rather than disconnected reports. Modularity: initiation papers,
periodic reviews, valuation updates, dashboards, CPSR and portfolio
implementations can be generated from the same controlled state.

The architecture creates a research asset that compounds. Every
additional covered company increases the cross-context evidence base.
Every review adds longitudinal evidence on capability formation,
execution, compounding, decay, Financial Viability and shareholder
capture. Every explicit upgrade or downgrade creates a testable
decision.

35. Conclusion

CIF begins from a simple idea: investors should analyse the
productive system that creates financial outcomes, not only the outcomes
themselves. Capability Science supplies the leading-indicator
organisational view. The Company Agent identifies the economic machine
through which capability operates. Archetype-specific conversion and
Financial Viability determine whether that machine can produce
sustainable owner economics. Capital-allocation closure tests whether
reinvestment actually earns its cost. Shareholder Value Capture
determines who receives the economics. State valuation determines what
that condition is worth. Portfolio construction, CPSR and CMVTR
determine how the security should be owned under uncertainty. Monitoring
closes the loop by forcing new evidence back into the persistent
state.

Canonical sequence	External State -> Economic Engine Architecture -> Capability -> Execution -> Archetype-Specific Economic Conversion -> Financial Viability -> Capital Allocation Closure -> Shareholder Value Capture -> Valuation -> Portfolio Decision -> CPSR / CMVTR -> Monitoring -> New Evidence -> State Delta.

The framework is neither a static checklist nor a one-time valuation
model. It is a controlled research system designed to preserve
causality, uncertainty, evidence lineage, shareholder economics and the
ability to change one’s mind. This is the current Celerity Investment
Framework.

Appendix
A – Celerity policy weights and interpretation

Dimension	Weight	Interpretive emphasis
Economic	7.5%	Regime, nominal growth, rates, credit, inflation, pricing and macro transmission.
Geopolitical / Sovereign	7.5%	State power, market access, strategic importance, regulation, licence and supply security.
Technology & Infrastructure Cycles	10.0%	Technology reality, build-out phase, bottlenecks, adoption, value capture and obsolescence.
Industry Structure	10.0%	Rivalry, entrants, supplier/buyer power, substitutes, capacity and regulation.
Competitive Power	15.0%	Durable barriers, process power, scale, switching, brand, networks, resources and customer value.
Compounder Quality	20.0%	ROIC, incremental returns, reinvestment runway, cash conversion, resilience and compounding duration.
Management & Culture	10.0%	Capital allocation, decision quality, coherence, learning, incentives and culture.
Innovation & Finance	7.5%	Productive innovation, optionality, funding, dilution, adjacent capability and capital discipline.
Valuation & Portfolio Construction	12.5%	Market-implied expectations, fair-value states, expected return, downside, portfolio role and size.

Interpretation rule: the aggregate score supports comparison and
research prioritisation. It does not override causal architecture, a
confirmed binding constraint, Financial Viability failure, broken
Shareholder Value Capture, severe Decision Risk, inadequate evidence or
valuation that fails the required-return test.

Appendix
B – Capability and Company Agent dictionary

Term	Celerity definition
Capability	Repeatable organisational capacity to produce a required outcome under changing conditions.
Operational	Capability repeatedly produces the required outcome.
Compounding	Accumulated capability makes later execution, formation or renewal faster, cheaper, better or more reliable.
Renewing	Capability is being redesigned or refreshed to preserve productive capacity.
Decaying	Stored productive capacity is weakening before or alongside reported performance.
Economic Engine	The causal business architecture through which capability converts into owner economics.
Financial Viability	Ability to reach the next economic closure point without value-destructive financing.
Capital cohort	A bounded acquisition, project, development or transformation investment tracked to owner-return closure.
Shareholder Value Capture	Value accruing to continuing owners after all economic capital, financing and dilution claims.
Structural Float	Durable negative working capital arising from ordinary customer/supplier economics rather than distress.
Stress Funding	Creditors or emergency capital financing ordinary survival.
State Delta	The controlled change to the persistent Company Agent caused by new evidence.

Appendix C –
Default review cadence

Cadence	Review object	Core question
Continuous	Evidence ledger	Did a material fact change engine, capability, viability, valuation or risk state?
Quarterly	Company Agent	What improved, weakened, became a constraint or changed confidence?
Reporting event	Economic conversion	Did operating/financial outcomes confirm or contradict the causal thesis?
Reporting event / material change	Valuation	Did fair value, required return or market-implied state change?
Quarterly	Portfolio	Does exposure remain rational relative to value, evidence, correlation and alternatives?
At least annually	Full re-underwrite	Would the security qualify today, and how does the answer differ from the prior state?
As material	CPSR / CMVTR	Has vulnerability, Threat, path risk or protection economics changed enough to warrant action?

Appendix D –
Minimum company research pack

• Investment question and classification.

• Source and evidence register.

• Economic Engine Architecture, sub-architectures
and overlays.

• Capability State Matrix.

• Management Representation Ledger.

• Capital cohort genealogy and closure tests.

• Working-capital/funding classification and
independent Financial Viability gate.

• Archetype-specific owner-cash model.

• Economic claim/dilution genealogy and Shareholder
Value Capture.

• CIF nine-dimension scorecard.

• Downside, central, upside, market-implied and
required-return valuation states.

• Valuation Attribution Ledger.

• Decision Risk assessment and pre-mortem.

• Upgrade, downgrade and thesis-break
conditions.

• Monitoring dashboard and evidence ledger.

• Portfolio role and CPSR/CMVTR model-status record
where applicable.

• Publication, model and assumption version log.

Appendix E –
Controlled interface map

Publication / methodology	Status in CIF v2.0
Celerity Investment Framework (CIF)	Current master investment decision architecture.
Celerity Company Agent Architecture v1.0	Controlled company-level causal architecture referenced by CIF.
Celerity Portfolio State & Resilience Research (CPSR)	Controlled resilience/loss methodology referenced downstream by CIF.
Valuation, Volatility and the Price of Protection (CMVTR)	Controlled path-risk/risk-transfer methodology referenced downstream by CIF.
Capability Science Foundations / Applied Theory	Upstream organisational theory and capability definitions.
WP-001 and TS-001 to TS-003	Frozen historical-validation methodology; not retrospectively rewritten by CIF v2.0.

Appendix F –
Acronyms and publication rule

Every acronym is written in full at first use, with the acronym
immediately afterward in brackets. Acronym-only use is permitted only
after that definition.

Acronym	Meaning
CIF	Celerity Investment Framework
CPSR	Celerity Portfolio State & Resilience Research
CMVTR	Valuation, Volatility and the Price of Protection
DCF	discounted cash flow
Owner FCF	Owner Free Cash Flow
FCFE	free cash flow to equity
OCF	operating cash flow
ROIC	return on invested capital
NPV	net present value
NIM	net interest margin
CET1	Common Equity Tier 1
TTV	total transaction value
GMV	gross merchandise value

References
and controlled Celerity sources

[1] Borness, S. (2026). Capability Science: Foundations. Celerity
Research, Publication Edition v1.0.

[2] Borness, S. (2026). Capability Science: Applied Theory. Celerity
Research, corrected publication edition.

[3] Celerity Research (2026). The Celerity Investment Framework:
Capability, Value and Portfolio Decision Under Uncertainty, Publication
Edition v1.2. Superseded by this v2.0 edition.

[4] Celerity Research (2026). Celerity Company Agent Architecture
Taxonomy, v1.0, Frozen Top-Level Architecture.

[5] Celerity Research (2026). Celerity Portfolio State &
Resilience Research (CPSR), canonical methodology and controlled
extensions.

[6] Celerity Research (2026). Capability Science for Investment
Research: A Historical Validation Framework for Identifying Future
Exceptional Long-Term Compounders, WP-001 Frozen v1.0; Technical
Standards TS-001 to TS-003.

[7] Celerity Research (2026). Historical Company Assessment pilot
series, HCA-001 to HCA-010. Pilot findings are not presented as
completed statistical validation.

[8] Celerity Research (2026). Valuation, Volatility and the Price of
Protection: The CMVTR Framework for Cyclical Equity Risk and Dynamic
Portfolio Protection. Methodology Edition v1.0, Canonical Final.

[9] Celerity Research (2026). CMVTR Empirical Calibration Register,
controlled working versions.

[10] Barney, J. (1991). Firm Resources and Sustained Competitive
Advantage. Journal of Management, 17(1), 99-120.

[11] Teece, D.J., Pisano, G. and Shuen, A. (1997). Dynamic
Capabilities and Strategic Management. Strategic Management Journal,
18(7), 509-533.

[12] Porter, M.E. (1980). Competitive Strategy; Porter, M.E. (1985).
Competitive Advantage. Free Press.

[13] Kahneman, D., Sibony, O. and Sunstein, C.R. (2021). Noise: A
Flaw in Human Judgment. Little, Brown Spark.

[14] Tetlock, P.E. and Gardner, D. (2015). Superforecasting: The Art
and Science of Prediction. Crown.

[15] Andersen, T.G., Bollerslev, T., Diebold, F.X. and Labys, P.
(2003). Modeling and Forecasting Realized Volatility. Econometrica,
71(2), 579-625.

[16] Corsi, F. (2009). A Simple Approximate Long-Memory Model of
Realized Volatility. Journal of Financial Econometrics, 7(2),
174-196.

[17] Barndorff-Nielsen, O.E., Kinnebrock, S. and Shephard, N. (2010).
Measuring Downside Risk: Realised Semivariance. In Volatility and Time
Series Econometrics. Oxford University Press.

[18] Bollerslev, T., Tauchen, G. and Zhou, H. (2009). Expected Stock
Returns and Variance Risk Premia. Review of Financial Studies, 22(11),
4463-4492.

Research disclaimer

This publication contains general research and analytical methodology
only. It is not personal financial advice, investment advice, a
recommendation to buy or sell any security or financial product, or a
representation that any historical or modelled outcome will recur.
Company-specific forecasts and valuations published by Celerity are
analytical estimates and may differ materially from realised outcomes.
The framework is designed to make uncertainty, assumptions and decision
conditions explicit; it cannot eliminate model risk, market risk or
human judgment error.

Publication Edition v2.0 | 1 September 2026 | Celerity
Investments Pty Ltd | Celerity Research

General information

Celerity publishes general research and commentary only. Nothing in this publication constitutes financial advice, investment advice, personal advice, an offer, solicitation or recommendation to buy or sell any financial product or security.