Methodology

01 Universe Construction

The coverage universe is built from first principles rather than index membership. Candidate companies pass through a qualitative pre-filter before any quantitative work begins:

• Market cap tier: primarily micro-cap (<$300M) and small-cap ($300M–$2B), where analyst coverage is sparse and mispricings tend to persist longer.
• Business quality floor: the company must demonstrate a plausible path to — or current evidence of — positive free cash flow generation. Pure concept or pre-revenue businesses are generally excluded.
• Valuation gap trigger: an initial calculation must show potential upside of at least 50% to an independently derived fair value estimate before a company enters the research pipeline.
• Public filings availability: the company must have at least two years of audited financial statements accessible via SEC/EDGAR, a recognised stock exchange, or an equivalent regulated filing system.

02 Data Sourcing and Validation

All financial data follows a zero-trust validation pipeline:

• SEC 10-K, 10-Q, 20-F, 6-K filings (or equivalent international regulatory filings)
• Company-issued earnings releases, investor presentations, and annual general meeting materials
• Exchange-filed prospectuses, material change notices, and insider transaction reports

Every key figure (revenue, EBITDA, net debt, shares outstanding, capex) must be reconciled between at least two independent sources before it appears in a model. Discrepancies trigger a forensic review of the raw filing rather than acceptance of either figure. Items that cannot be independently confirmed are displayed as [awaiting: description] rather than estimated.

03 Financial Modelling and Valuation

Valuation is approached through multiple independent lenses and cross-checked for internal consistency. No single method is treated as authoritative.

• Discounted cash flow (DCF): a three-scenario model (bear / base / bull) with explicit assumptions on revenue growth, operating margins, capex intensity, and terminal growth rate. WACC inputs are derived from publicly observable market data at the time of modelling.
• Earnings power value (EPV): a no-growth steady-state calculation based on normalised earnings, providing a conservative floor value independent of growth assumptions.
• Asset-based / book value: used primarily for capital-intensive or asset-heavy businesses; tangible book adjusted for known off-balance-sheet items and asset quality.
• Relative valuation: EV/EBITDA, EV/Revenue, P/E, and P/FCF multiples benchmarked against a curated peer group. Outlier peers are flagged and excluded from the median.

A position in the radar requires a margin of safety — the percentage gap between current price and independently derived fair value — commensurate with the risk profile. Higher uncertainty (fewer audited periods, more cyclical revenue, thinner liquidity) demands a larger margin before a company is rated High or Highest conviction.

04 Adversarial Bull vs. Bear Review

Each company thesis is deliberately stress-tested through a structured adversarial process before it is published. The goal is to surface the strongest possible case against the investment — not to confirm the thesis.

A company remains in the radar only when the bull case is materially stronger than the bear case after honest scrutiny. When the bear case cannot be adequately rebutted, the company is removed or downgraded regardless of the initial valuation appeal.

05 Conviction Scores — What They Mean

The conviction score (1–5) reflects analytical confidence in the research thesis — the degree to which the available evidence supports the hypothesis that the company is materially undervalued with an identifiable path to realisation. It is not a buy/sell signal, a return forecast, or a risk rating.

06 Research Phase

Each company is tagged with a research phase (1–3) reflecting progress through the analytical pipeline, distinct from the conviction score:

• Phase 1 — Screening: The company has passed the initial quantitative screen and qualitative pre-filter. Financial data is being gathered and the preliminary thesis is being formed.
• Phase 2 — Deep research: Primary filing cross-validation is underway or complete. The valuation model is built. Catalyst, risk, and competitive landscape analysis is in progress.
• Phase 3 — Thesis complete: The full research cycle is complete — all sections are documented, adversarial review is finished, and the thesis summary is in final form. The company is ready for active monitoring.

07 Data Freshness and Ongoing Maintenance

Each company page shows a "last updated" date that reflects when the underlying research data was last reviewed or refreshed. This is not a real-time price feed — financial statement data typically lags the most recent quarterly report by weeks, and market prices are not updated continuously.

Major corporate events (earnings releases, acquisitions, director changes, profit warnings) that materially affect the thesis trigger a priority review. Absence of a recent update date does not mean the thesis is invalid — it may reflect a business whose fundamentals change slowly. However, older data should be weighted with appropriate caution.

The  stale indicator appears when a company's last-updated date exceeds a threshold relative to its reporting cycle, prompting a refresh review.

08 What This Screen Is Not

• Not a portfolio: Presence on this radar does not imply any position is held or recommended. See the Disclaimer for full conflict-of-interest disclosure.
• Not exhaustive: This is a curated screen, not a comprehensive market survey. Many undervalued companies are not covered because they have not yet entered the research pipeline.
• Not a ranking: The order in which companies appear on the dashboard reflects research state and recency of update, not a rank-ordering of investment merit.
• Not a model portfolio: Position sizing, portfolio construction, correlation, and diversification considerations are entirely outside the scope of this research screen.
• Not forward-looking guidance: Any projections or scenario analysis shown are illustrative modelling assumptions, not earnings forecasts or price targets in a regulated sense.