How the quality score works

What the score is

Each metric is scored from 0 to 10 against a benchmark, higher being better. Metrics roll up into five pillars, and the pillars combine into the overall score using the weights below. The result is a systematic read of reported fundamentals — the same arithmetic for every company, with no analyst judgement applied on top and no adjustment for what the market thinks.

A company is scored on up to 21 metrics. “Up to” matters: coverage depends on what a company’s filings support, and a metric that cannot be computed is left out rather than counted as a failure. The coverage section below spells out what happens then.

The five pillars and their weights

The weights are fixed, published, and shown in the app next to each pillar. They sum to 100, so a pillar’s weight is simply its share of the overall score.

Profitability carries the most weight because durable earning power is the strongest single signal of quality; Efficiency the least, because it largely describes how the other four are achieved.

What each pillar measures

Within a pillar, metrics carry one of three importance tiers — primary, secondary or supporting — and a metric’s tier decides how much it moves its pillar. The exact per-metric weights are calibration we keep tuning, so we publish the tier rather than a number that would be stale by the time you read it.

Profitability · weight 25

How much profit the company generates from its capital and sales.

• Return on Equity primary · vs sector — Profit earned on shareholders' capital.
• Return on Invested Capital primary · vs sector — Return on all capital the business employs, debt and equity.
• Net Margin secondary · vs sector — Share of revenue left as bottom-line profit.
• Free Cash Flow Margin secondary — Free cash flow as a share of revenue.
• Gross Margin supporting · vs sector — Revenue left after direct production costs.

Solvency · weight 20

The company's ability to service debt and meet its obligations.

• Interest Coverage primary — How many times operating profit covers interest expense.
• Net Debt / EBITDA primary — Net debt relative to annual cash earnings.
• Debt / Equity secondary — Borrowed capital relative to shareholders' equity.
• Debt / Free Cash Flow secondary — Total debt relative to yearly free cash flow.
• Current Ratio supporting — Short-term assets versus short-term liabilities.

Growth · weight 20

How fast revenue, earnings and cash flow are compounding.

• Revenue CAGR primary — Annualised revenue growth over recent years.
• EPS CAGR primary — Annualised growth in per-share earnings.
• FCF CAGR secondary — Annualised growth in free cash flow.
• Operating Income CAGR secondary — Annualised growth in core operating profit.

Efficiency · weight 15

How well assets and earnings convert into cash and output.

• Asset Turnover secondary · vs sector — Revenue generated per dollar of assets.
• FCF Conversion secondary — Share of net income that becomes free cash flow.
• Operating Margin Trend supporting — Direction of operating margin over recent years.

Predictability · weight 20

How stable and consistent results have been over time.

• Revenue Stability secondary — How steady year-over-year revenue growth has been.
• Earnings Stability secondary — How consistent earnings growth has been.
• Positive FCF Years secondary — Share of recent years with positive free cash flow.
• Operating Margin Stability supporting — How steady operating margins have been.

Why some metrics are judged against the sector

A 9% net margin is unremarkable for a software company and excellent for a grocer. Five metrics are therefore scored relative to a sector baseline rather than an absolute curve, so a value at the sector median scores as average by construction:

• Return on Equity
• Return on Invested Capital
• Net Margin
• Gross Margin
• Asset Turnover

The rest are judged on an absolute curve that is the same for every company — interest coverage of 1.5× is thin whatever the industry. Where a company’s profile carries no sector at all, the five fall back to a single global baseline instead of being dropped, so the score stays sector-aware rather than always being a comparison against a company’s own sector.

How to read a score

Individual metrics and pillars are banded the same way, and the same thresholds colour the score in the app:

• Strong — 7 and above.
• Average — 4 up to 7.
• Weak — below 4.

The overall score carries a verdict on a finer scale:

• Exceptional quality — 8.5 and above.
• High quality — 7 up to 8.5.
• Solid quality — 5.5 up to 7.
• Mixed quality — 4 up to 5.5.
• Low quality — below 4.

Read the pillars before the headline number. Two companies can both score 6.5 while being nothing alike — one steadily average everywhere, the other excellent and fragile at once — and which of the two you are looking at is the more useful fact.

When a score is partial — or absent

Coverage is decided per company, and the score is explicit about it rather than quietly filling the gaps:

• A metric that cannot be computed from the filings is absent, not scored as zero. A pillar score is the weighted average of the metrics that were computed, so a missing figure never drags a pillar down.
• A pillar is withheld entirely when less than half of its metric weight could be covered — too little evidence to state a score at all.
• The overall score is the weighted average of the pillars that survived, renormalised over their weights. When at least one pillar is missing, the score is marked partial.
• When not a single metric can be scored, the app shows “Not enough data” instead of a number. A score is not guaranteed for a given ticker.

Everything multi-year — the Growth pillar, Predictability and Operating Margin Trend — reads the last three to five annual periods, and is skipped entirely below three. The remaining metrics read the latest annual period or the current trailing statistics. So a recent IPO will typically score on profitability and solvency long before it scores on growth or consistency.

A few metrics are dropped rather than scored badly where the ratio would be meaningless: return on invested capital when invested capital is not positive, net debt to EBITDA without positive EBITDA, debt to equity with negative book equity, debt to free cash flow without positive free cash flow, and cash conversion without positive net income. This is deliberate — it stops a net-cash or loss-making company from landing on an accidental top score.

What the score does not say

The score is backward-looking and mechanical. It reads what a company has reported and says nothing about four things that matter just as much:

• Price. A high-quality business can be a poor investment at the wrong price. Valuation is a separate question — see our methodology for how we estimate what a company is worth.
• Management and capital allocation. No financial ratio captures who is running the business, or how well they reinvest.
• Competitive position. A moat shows up in the numbers years after it starts eroding.
• The future. Every input is history. A pending patent cliff, a lost contract or a new regulation is invisible here.

It can also be distorted by one-off items and unusual accounting — a large disposal flatters a year of margins and cash conversion without the business having changed. Treat the score as a fast, consistent first filter that tells you where to look harder, not as a recommendation.