Analytical Framework
CBOM — Conviction-Based Opportunity Mapping. The analytical framework behind every PRZC Research report.
Every PRZC Research report is built around a single core question: what does the current price require the business to achieve, and how likely is that outcome?
CBOM is the framework for answering it. Rather than generating a price target from a forward model, we work in reverse — decomposing the market price into its implied expectations for growth, margins, and capital returns. We then assess those expectations against the evidence: historical base rates, competitive dynamics, industry constraints, and primary research.
The output is not a single forecast. It is a probability-weighted map of outcomes: what the business must achieve to justify the current price (the required performance), what we believe it is likely to achieve (the probable performance), and the gap between the two. Where the gap is wide and the evidence is strong, conviction is high. Where the gap is narrow or ambiguous, the opportunity is smaller — or the risk of disappointment is greater.
The framework applies identically across sectors. Whether we are analysing a datacenter REIT, a large-cap grocer, or an AI infrastructure play, the starting point is always the price and what it demands.
"The opportunity is not in predicting the future — it is in identifying when the price requires an outcome that the evidence does not support."
"We do not start with a view and work toward a price. We start with the price and interrogate what it believes."
Six integrated stages applied to every report, from initial price decomposition through to published thesis and ongoing scorecard tracking.
Price Decomposition
We begin by mapping what the current market price implies. Using reverse DCF and relative valuation decomposition, we establish the growth rate, margin trajectory, and return on invested capital that a stock must achieve to justify its valuation — not as a target, but as a requirement embedded in the price itself.
The gap between this required performance and what we assess as probable performance defines the opportunity. A wide gap in favour of the bull case is a potential buy signal; a wide gap in the bear direction signals elevated disappointment risk.
Valuation-First Analysis
The Required Business Performance® framework formalises Stage 01 into a structured assessment. We calculate the specific revenue growth rates, EBIT margin expansion, free cash flow conversion, and capital allocation efficiency that the current price demands over the next three to five years.
We then benchmark those requirements against historical achievement rates for comparable companies in comparable environments. This transforms an abstract valuation into a concrete question: has any business in this position historically achieved what this price requires?
No single-point forecasts are produced. Instead, the output is a probability distribution — a bear, base, and bull scenario weighted by our assessment of likelihood given the evidence.
Anti-Confirmation Bias
Before any thesis is published, it passes through an adversarial review stage. The analyst is required to write the strongest possible opposing case — even, and especially, when the primary thesis is bullish. Every published bear case must be treated with the same rigour as the bull case, and vice versa.
This stage is designed to surface evidence that confirmation bias would suppress. We actively seek data points, historical precedents, and structural arguments that challenge the thesis. If the disconfirmation case cannot be answered convincingly, the conviction rating is reduced or the thesis is withheld.
The result is that every published PRZC report contains an explicit bear case, regardless of its overall recommendation — because a thesis without a credible opposing argument is not a thesis; it is a hope.
Evidence Gathering
Quantitative modelling is necessary but not sufficient. Every report integrates primary research to ground the numbers in operational reality. This includes channel checks, industry interview synthesis, regulatory filing analysis, patent and product pipeline review, and alternative data sets where applicable.
We do not treat consensus estimates as a starting point — we treat them as a signal worth interrogating. Where consensus and primary evidence diverge, we investigate why, and we publish our reasoning. Where we rely on management guidance, we disclose that dependence explicitly and stress-test it against track record.
The objective is a research product that is not merely an aggregation of public information, but an independent analytical judgment on what that information implies.
Track Record
Every published thesis is tracked against outcome. We maintain explicit scorecards for all active and closed investment theses, updated at defined intervals — typically three and six months post-publication. The scorecard records the original thesis, the original probability distribution, and the actual outcome to date.
Failures are published, not hidden. Where a thesis was wrong — whether on direction, timing, or magnitude — we publish a post-mortem that identifies which assumption was incorrect and why. This is not a convention; it is a structural requirement of the CBOM framework.
Track record transparency serves two purposes: it holds the research to account, and it provides readers with a calibrated view of what the analytical process can and cannot do reliably.
Universal Application
The CBOM framework is applied identically regardless of sector, geography, or market capitalisation. Whether the subject is an AI infrastructure operator, a large-cap grocer, an energy transition business, or a mid-market private equity target, the analytical entry point is always the same: what does the price require, and how probable is that outcome?
Sector-specific knowledge informs the evidence assessment — we maintain active coverage across AI and infrastructure, datacenter real estate, retail and e-commerce, energy, open-source software, and macro markets. But the framework itself does not flex for sector. This discipline prevents the analytical drift that occurs when different valuation methodologies are selectively applied to flatter the conclusion.
Adversarial stress-testing is not optional. It is a mandatory stage in the publication process for every report, regardless of conviction level or direction.
The Problem
Analyst confirmation bias is the most persistent structural failure in institutional investment research. Once a thesis is formed — whether from a valuation screen, a management meeting, or an industry contact — the human tendency is to gather evidence that supports it and discount evidence that challenges it.
The result is research that reads like a verdict in search of a crime scene: the conclusion is fixed; only the supporting evidence is gathered. This is precisely the failure mode that produces consensus estimates that prove systematically optimistic at peaks and systematically pessimistic at troughs.
PRZC Research does not claim immunity from this tendency. It designs around it.
The Protocol
The Disconfirmation Protocol requires that, before any thesis proceeds to publication, the analyst must complete the following stages:
"A thesis that cannot survive its own bear case does not deserve to be published. The strongest argument against a position is not a risk factor — it is a window into where the analysis might be wrong."PRZC Research — Disconfirmation Protocol, Internal Standard
All published theses are scored. Failures are published with post-mortems. The scorecard is not a marketing exercise — it is an accountability mechanism built into the methodology.
Principle
When a report is published, its thesis is entered into the active scorecard with the original probability distribution, key assumptions, and disconfirmation variable. Outcomes are assessed at three and six months post-publication.
Principle
Where a thesis proved incorrect — on direction, magnitude, or timing — PRZC publishes a follow-up that identifies the failed assumption, explains why it was wrong, and updates the probability framework. This is non-negotiable.
Principle
Our AI Scorecard series exemplifies the approach: each report scores prior AI investment theses against actual outcomes, with explicit verdicts on what the framework got right, what it got wrong, and what has yet to resolve. No airbrushing.
Track record reports are published as part of our standard research series and are freely accessible. Browse the full research archive →
Every published report applies the full CBOM framework. Browse the research archive or commission a bespoke report on any topic or company.
