How Company Analysis Works

Header & Pricing Signal

The header shows the company's name, ticker, sector and industry classification (from SEC filings), a recent price, and a colored valuation zone pill: Undervalued, Fair Value, or Overvalued.

How the valuation zone is determined

The zone is derived primarily from the company's Est. Fair Value — the median output of our 5-year discounted cash flow model — compared to the current market price:

• Undervalued — Est. Fair Value is more than 15% above the current price.
• Fair Value — current price is within ±15% of the Est. Fair Value.
• Overvalued — Est. Fair Value is more than 15% below the current price.

When DCF data is not available (e.g., for pre-profit companies or newly added issuers), the zone falls back to the composite score as a proxy.

Active Alerts

Alerts surface when the most recent snapshot of a company crosses a meaningful threshold relative to the discounted cash flow band, or when a multiple reaches an extreme versus its own recent history. Only alerts from the most recent analysis run are shown — old alerts are automatically suppressed.

What can trigger an alert

• Overvalued / Undervalued (DCF band). Market cap is more than 20% above the bull-case Est. Fair Value, or more than 20% below the bear-case Est. Fair Value, computed from the discounted cash flow model (see "Est. Fair Value Range" below).
• Multiple stretched / compressed. The company's EV/EBITDA is more than 1.5 standard deviations away from its own two-year average. "Stretched" means above; "compressed" means below.

What we deliberately suppress

When the model produces a result that's wildly out of band — for example, a young growth company whose DCF base value is 10× smaller than its market cap — we don't fire the alert at all. At those extremes, the model is outside its useful range, so the alert would be noise rather than signal. The DCF chart still shows whatever the model produced, but we won't shout about it in a callout.

Health Score & Pillars

The Health Score is a single 0–100 number summarizing how financially sound a company looks based on its most recent annual and trailing-twelve-month (TTM) financial statements. It's the rolled-up answer to "would a careful investor be comfortable owning this company?"

The six pillars

The overall score is a weighted average of six pillar scores, each on a 0–100 scale. Pillar weights vary slightly by company type (a high-growth software company is evaluated differently from a mature utility), but the default breakdown is:

• Growth Quality (15%). Multi-year revenue, profit, and cash flow growth rates (CAGR), with adjustments for whether growth is accelerating or decelerating. Consistent, profitable growth scores higher than erratic revenue spikes.
• Profitability Quality (20%). Margin levels and trends, return on invested capital (ROIC), and consistency over time. A company with durable, improving margins and high returns on capital scores well here.
• Balance Sheet Strength (20%). Leverage ratios, liquidity, interest coverage, and intangible/goodwill risk. Companies with low debt, ample cash, and comfortably covered interest payments score highest.
• Cash Flow Quality (20%). How well reported earnings convert to actual cash. Measures include the accruals ratio (the gap between accrual-basis earnings and cash earnings), dilution tracking, and free cash flow consistency.
• Efficiency & Capital Allocation (15%). Whether management is creating value with the capital it retains. Inputs include ROIC trends, asset turnover, working capital management, buyback effectiveness, and reinvestment productivity.
• Stability & Resilience (10%). How well the business holds up under stress. Evaluates revenue and margin volatility, debt maturity risk, impairment frequency, and cyclicality.

How a score becomes a label

The overall health score maps to a label so it's easier to skim:

• 75–100 → Strong
• 60–74 → Healthy
• 45–59 → Mixed
• 30–44 → Weak
• 0–29 → Stressed

Individual pillars use a separate grading scale:

• 90–100 → Elite
• 75–89 → Strong
• 60–74 → Acceptable
• 40–59 → Weakening
• 0–39 → Impaired

Top Positives & Top Concerns

Beneath the pillars we list the three most-impactful positive observations and the three most-impactful concerns the model identified during scoring. These are short plain-English statements like "Strong revenue growth — sales up more than 20% year over year" or "Low FCF yield — below 1%, price implies exceptional future growth." They're meant to give you a head-start on your own due diligence rather than replace it.

Health Score Trend

A small line chart showing how the overall and pillar scores have evolved over the most recent eight reporting periods. Useful for spotting whether a company is improving, deteriorating, or stable. A company whose Operating Trend pillar has been falling for four straight quarters tells you something a single point-in-time score cannot.

TTM Financials

"TTM" stands for trailing twelve months. Rather than showing a calendar year that may be six months stale by the time you read it, we sum the four most recent quarterly reports to give you the freshest possible annual view of the company.

What's shown

• Revenue — total sales
• Gross Profit — revenue minus cost of goods sold
• Operating Income — what's left after operating expenses (R&D, SG&A, etc.)
• EBITDA — operating income plus depreciation and amortization, a rough proxy for cash earnings
• Net Income — final reported earnings after interest and taxes
• Free Cash Flow (FCF) — operating cash flow minus capital expenditures, the cash actually available to investors
• Total Debt — short-term + long-term borrowings from the balance sheet
• Cash & Equivalents — cash on hand at the most recent reporting date

Where the data comes from

Every line is mapped from the company's XBRL filings with the SEC (10-K and 10-Q). We use a curated list of XBRL tags ranked by reliability — when the most authoritative tag for a metric is missing, we fall back to the next-best one. This is why two analysts reading the same 10-K can sometimes show slightly different values: there's typically more than one valid tag for a given concept and the choice matters.

Valuation Ratios

A grid of standard valuation multiples computed from the most recent market cap and TTM financials. These tell you how much the market is willing to pay for each dollar of revenue, earnings, or cash flow.

What's shown and what it means

• Market Cap = shares outstanding × current price. The total value of the company's equity.
• Enterprise Value (EV) = market cap + total debt − cash. The cost of buying the entire business and paying off its debt.
• EV / Revenue — useful for unprofitable or pre-profit companies where earnings ratios don't apply.
• EV / EBITDA — the most commonly used "true" valuation multiple because it isolates operating performance from capital structure and tax differences.
• P/E (TTM) — price ÷ trailing earnings per share. Easy to compute, blunt instrument; meaningless when earnings are negative.
• P/B (Price / Book) — price relative to balance-sheet equity. Most useful for asset-heavy businesses (banks, insurers, real estate).
• P/S (Price / Sales) — price relative to revenue. Useful early-stage proxy when there are no earnings yet.
• P/FCF — price relative to free cash flow. The cash-flow-purist version of P/E.
• FCF Yield — the inverse of P/FCF, expressed as a percentage. "How much free cash flow does this company generate as a percent of its market cap?"

Bands & percentiles

Where there's enough history, we also show how today's multiples compare to that company's own past: "EV/EBITDA is in the 78th percentile of its 5-year history" means the stock is trading at a richer multiple than 78% of the trading days in the last five years. Comparing a company to itself sidesteps the trap of comparing across industries that should never have the same multiple in the first place.

Forward P/E & PEG

Trailing P/E uses the last twelve months of earnings — historical data. Forward P/E uses next year's expected earnings instead, which is what investors actually care about: you're buying future profits, not past ones.

How we calculate it (hybrid approach)

1. Consensus path. When a recent analyst-consensus EPS estimate is available for the next fiscal year (we pull these from a free public data source updated regularly), we use that as the forward EPS and label the source as "consensus."
2. Extrapolation path. When no consensus is available, we project the next twelve months of EPS from the trend in the last four quarterly reports — essentially asking "if recent quarterly growth continued, what would NTM EPS be?" Growth assumptions are clamped to a sane range so a single outlier quarter can't produce nonsense. We label the source as "extrapolated" so you can see it's our projection rather than an analyst consensus.

The displayed Forward P/E is then simply price ÷ projected EPS.

PEG ratio

PEG = forward P/E ÷ expected earnings growth (in percent). The intuition is from Peter Lynch: a stock with a 30× P/E growing at 30% per year has a PEG of 1.0, which Lynch considered "fair value." A PEG below 1.0 suggests growth at a reasonable price; above 2.0 suggests you're paying a premium even after accounting for growth. PEG only makes sense when growth is positive — for declining or no-growth companies it returns no value.

Est. Fair Value Range (DCF)

A discounted cash flow (DCF) model estimates what a company should be worth today, based on the cash it's expected to generate in the future, discounted back to today's dollars. It's the most theoretically defensible way to value a business.

The intuition in one sentence

A dollar of cash flow next year is worth less than a dollar today (because you could invest today's dollar and earn a return), so we project cash flows several years out, divide each by a "discount factor" that reflects how risky the company is, and sum the results.

How our DCF works

We use a two-stage model — the standard approach taught in corporate finance:

1. Stage 1 (years 1–5): Free cash flow grows at the company's recent historical growth rate, with separate bear, base, and bull scenarios.
2. Stage 2 (years 6–10): Growth fades linearly from the stage-1 rate down to a long-run terminal rate (we use 3%, roughly nominal GDP growth). This prevents the model from assuming a young growth company will compound at 30% forever.
3. Terminal value: Beyond year 10 we apply Gordon Growth — the standard formula for capitalizing a perpetual cash flow stream growing at the terminal rate.

The discount rate (WACC) is size-adjusted

We don't use a single discount rate for every company. Mega-caps borrow more cheaply and have lower equity risk premia than small caps, so we apply a tiered weighted-average cost of capital (WACC):

• Market cap ≥ $200B → 7.5%
• Market cap ≥ $50B → 8.5%
• Market cap ≥ $10B → 9.5%
• Below $10B → 10%

What you see on the card

• Bear / Base / Bull (per share & total). Three scenarios reflecting conservative, central, and optimistic growth assumptions. The base case is the model's central estimate; bear and bull are roughly half and 1.5× the base growth rate, respectively.
• Model variant. Either "FCF two-stage" (the standard model above, used for profitable companies) or "Revenue DCF" (used for unprofitable but rapidly-growing companies — see below).
• WACC used. The actual discount rate applied for this company.

The Revenue DCF branch (for high-growth unprofitable companies)

A standard FCF-based DCF can't value a company with negative free cash flow — you'd be discounting losses. Famously articulated by Aswath Damodaran, the workaround for early-stage growth companies is to project revenue forward, assume the company eventually reaches a steady-state operating margin typical of mature peers, tax that, and treat the result as synthetic free cash flow. We do this automatically for companies with negative FCF, meaningful revenue (above $100M), and revenue growth above 15%.

Built-in sanity checks

DCFs can produce nonsense at the extremes — a high-debt cyclical at trough earnings might calculate a bear-case fair value of zero, implying bankruptcy. We refuse to display fair values that fall outside a sensible range relative to the actual market cap (specifically, between 10% and 1000% of it). When the model output is degenerate, the chart shows nothing for that point rather than a misleading value. Honest gaps are better than false certainty.

Market Cap vs DCF Fair Value Band

A line chart plotting actual market cap (the cyan line) against the bear-to-bull DCF range (the shaded purple band) over the last 24 months. This is the single most important valuation visual on the page: if the cyan line spends a long time above the band, the market has been pricing the company richly relative to its own fundamentals; if it spends time below, the market is pricing in pessimism.

How the historical points are computed

For each historical date, we look up that day's actual closing price and the most-recent financial statement that would have been publicly available on that date, then run the same DCF model described above. The result is the bear/base/bull range as of that historical date — what the model would have said at the time, given the data known then.

How to read it

• Market cap line consistently inside the band → company is being priced roughly in line with its fundamentals.
• Market cap line above the band for an extended period → either the market expects much faster growth than recent history, or the stock is overextended.
• Market cap line below the band → either the market expects deteriorating fundamentals, or there's an opportunity hiding.
• Gaps in the band → the model couldn't produce a sensible estimate for those dates (high-debt cyclicals at trough earnings, restated financials, etc.).

Quarterly Financials

A horizontally-scrollable table showing the last twelve quarterly reports for the major income-statement and cash-flow line items. Useful for spotting seasonality, identifying inflection points, and comparing a company's growth trajectory quarter-over-quarter.

Why some quarters are labeled "derived"

US public companies don't file a separate Q4 10-Q — the annual 10-K covers the whole fiscal year. That means Q4 doesn't appear directly in any single XBRL filing. We reconstruct it as Annual − (Q1 + Q2 + Q3) for every flow-based metric, which is the standard method analysts use. Balance-sheet items at year-end are taken straight from the 10-K. We mark these reconstructed quarters with a small "derived" label so you know they're computed rather than directly filed.

Revenue & Free Cash Flow

A horizontally-scrollable bar chart showing revenue (green) and free cash flow (cyan) side-by-side for the same 12 quarters as the table above. Revenue tells you about the top-line trajectory; FCF tells you whether that revenue is converting to cash. The two together usually reveal more than either alone:

• Both rising → healthy growth funded by the business itself.
• Revenue rising, FCF flat or falling → working capital, capex, or margin issues.
• Revenue flat, FCF rising → operational efficiency or capex discipline kicking in.
• Negative FCF bars (shown in red) → the company is consuming cash. Common for early-stage companies; concerning for mature ones.

TA Composite Score & Signals

The TA Composite Score is a more comprehensive 0–100 verdict that combines our health scoring, a multi-layered valuation assessment, and market-sentiment signals into a single reading. Where the Health Score asks "is this a sound business?", the Composite Score asks "given everything we can measure today, where does this company sit on the spectrum from attractive to unattractive?"

The three pillars of the composite

• TA Health Score (0–100). The six-pillar health assessment described above — growth quality, profitability, balance sheet strength, cash flow quality, capital allocation efficiency, and stability. This captures the fundamental quality of the business.
• Valuation Score (0–100). A three-layer valuation assessment (see below) that blends absolute multiples, growth-adjusted ratios, and DCF-based expectations analysis. This captures whether the market price is reasonable for what the business delivers.
• Sentiment Score. A weighted blend of market-based signals — see below.

How the Valuation Score works (three layers)

Rather than relying on any single valuation method, the Valuation Score blends three complementary approaches, each capturing a different angle on whether the price makes sense:

• Layer 1 — Absolute Multiples (40% weight). Where the company's EV/EBIT, EV/FCF, P/E, and FCF yield sit relative to sector peers. EV/EBIT carries a 50% weight premium over other multiples, informed by Joel Greenblatt's research showing it is the strongest single predictor of future stock returns across decades of backtesting. Percentile normalization ensures cross-sector comparisons are meaningful.
• Layer 2 — Growth-Adjusted Ratios (30% weight). PEG-style ratios that ask whether the valuation is reasonable given the company's growth rate. For high-growth technology companies, this layer also evaluates Rule of 40 (revenue growth plus margin).
• Layer 3 — DCF & Implied Expectations (30% weight). Combines the Monte Carlo probability of undervaluation with a reverse DCF analysis. The reverse DCF solves for what growth rate the market is implicitly pricing in, then compares that to the company's actual historical growth. A wide gap between implied and actual growth suggests the market may be over- or under-estimating the business.

Value-trap guard

A stock can screen as "cheap" and still be a poor investment if the business is deteriorating. The Composite Score includes a value-trap guard that flags companies where low valuation multiples coexist with declining health scores, weakening cash flow, or rising leverage. When the guard triggers, the composite score is adjusted downward and conviction is reduced, rather than allowing a superficially attractive valuation to mask fundamental deterioration.

The signals that feed into Sentiment

• Earnings Quality. How well reported earnings are converting to actual cash. Built from accruals ratio (the gap between accrual-basis earnings and cash earnings) and cash-conversion ratio (operating cash flow ÷ net income).
• Insider Activity. Officers and directors filing Form 4 disclosures with the SEC when they buy or sell company stock. Open-market purchases by senior insiders are weighted more heavily than sales (which can have many non-signal motives). Rule 10b5-1 planned sales are excluded.
• Short Interest. The fraction of a company's float being sold short, sourced from FINRA's free public dataset. High short interest combined with high days-to-cover signals heavy bearish positioning.
• Institutional Flow. Quarter-over-quarter changes in the number of institutional holders increasing versus decreasing their position, sourced from 13F filings.

Composite weights are dynamic

Different signals matter more at different stages of a company's life. For a financially distressed company, the health pillar weighs heavier; for a stable mature business, valuation matters more. The composite shifts weight toward the most diagnostic pillar based on the underlying health score. A signal that's unavailable for a company (e.g., no insider data because none has been filed recently) is excluded from the average and the remaining weights are renormalized.

How the Composite Score maps to a signal label

The composite signal labels describe overall model assessment — they are research signals, not recommendations to act:

• 80–100 → Strong Buy (model sees compelling combination of quality, value, and sentiment)
• 65–79 → Buy
• 50–64 → Hold
• 35–49 → Cautious
• 20–34 → Sell
• 0–19 → Strong Sell

Note: the valuation zone (Undervalued / Fair Value / Overvalued) shown in the header is separate from the composite signal label. The zone is driven by the DCF-based Est. Fair Value vs. price; the signal label blends health, valuation, and sentiment together.

Conviction

Alongside the score, we report a Conviction level (High, Medium, Low) based on how much agreement there is between the three pillars. When health, valuation, and sentiment all point in the same direction, conviction is High; when they disagree, conviction is Low and the user is reminded to do their own work.

Cross-Check: Greenblatt Magic Formula

Joel Greenblatt's Magic Formula is one of the most empirically validated stock-picking methodologies in finance. First described in The Little Book That Beats the Market (2005), it has been backtested across US, European, Nordic, and Asian markets over multiple decades with consistently strong results.

How it works

The Magic Formula ranks every company in a universe by two factors, then combines the ranks:

1. Earnings Yield (EBIT ÷ Enterprise Value). How much pre-tax, pre-interest profit the company generates per dollar of total business cost. Higher = cheaper. This is the same signal captured by EV/EBIT in our valuation multiples, just expressed as a yield.
2. Return on Capital (EBIT ÷ Tangible Capital). How efficiently the company's operating assets generate earnings, excluding goodwill and intangibles from the capital base. This is a stricter variant of ROIC — by removing goodwill, it reveals the true efficiency of the company's operations independent of acquisition history.

Each company is ranked on both factors, and the two ranks are added together. The lowest combined rank = best (cheapest high-quality company). We convert this to a percentile so 100th = top-ranked.

What we show

• Earnings Yield — the raw yield and the company's rank within the universe.
• Return on Capital — the raw return and rank.
• Combined Rank / Percentile — where the company sits in the Magic Formula ranking.
• Agreement — whether the Magic Formula's assessment aligns with our Composite Score:
  • Aligned — both models see the company similarly (within 20 points).
  • Mild Divergence — moderate disagreement (20–35 points apart).
  • Divergent — strong disagreement (35+ points apart). When this happens, we provide an explanation of what's likely causing it.

Why we show it as a cross-check, not an input

Both ROIC and EV/EBIT are already embedded in our health and valuation pillars. If we added the Magic Formula rank into the Composite Score, we'd be double-counting those signals. Instead, we display it alongside our score so you can see when two independent methodologies agree or disagree. Disagreement is often more informative than agreement — it tells you which factors the models weight differently and prompts further investigation.

Tangible ROIC vs Standard ROIC

Our profitability pillar now blends two variants of ROIC: standard (60% weight) and Greenblatt's tangible variant (40% weight). The tangible version excludes goodwill and intangible assets from the capital base. For companies that have grown heavily through acquisitions, standard ROIC can significantly understate operating efficiency because goodwill inflates the denominator. When the two diverge by more than 2×, we flag it as a "high goodwill divergence" — a signal that the company's operations are more efficient than the headline ROIC suggests.

Which companies are excluded

Per Greenblatt's methodology, financials (banks, insurers, asset managers) and utilities are excluded from the Magic Formula ranking. Their capital structures make EBIT/EV and tangible ROC comparisons unreliable — a bank's "capital" is mostly deposits and an insurer's is mostly reserves, neither of which is comparable to a manufacturer's invested capital. For these companies, the Magic Formula section will not appear.

Monte Carlo DCF

The standard DCF gives you three discrete scenarios: bear, base, bull. A Monte Carlo DCF gives you a distribution by running the DCF thousands of times, each time drawing the key inputs (growth rate, margin, discount rate) randomly from sensible ranges around their central estimates. The output is the probability distribution of fair values — and from that, the probability that the current price is below the model's idea of fair value.

What we report

• Probability undervalued. The fraction of simulations in which the model's fair value came out higher than the current stock price.
• Probability undervalued by 20%+. A stricter version: the fraction of simulations where fair value exceeded price by more than 20%, often called "margin of safety."
• 5th / 25th / 50th / 75th / 95th percentile values. The distribution itself. The 50th percentile is the median fair value; the 5th and 95th give you a sense of how wide the cone of uncertainty really is.
• Mean & median. Two summary statistics. A wide gap between mean and median signals a skewed distribution (often the case for high-uncertainty growth names).

Health-adjusted discount rate

The Monte Carlo DCF nudges the cost of capital based on the company's TA Health Score: weaker companies get a higher discount rate (because they're riskier), stronger companies get a slight discount. This is roughly how analysts adjust for company-specific risk in practice, just made systematic.

Crowd Sentiment

An aggregate view of how TradeApes users have rated this company across the My Thesis form (see below). Shown as percent bullish / neutral / bearish along with the total number of ratings and a "crowd conviction" label that reflects how strongly users agree with each other.

It's useful as a reality check on your own thesis ("am I seeing something everyone else missed, or something everyone else has already concluded?") but it's a popularity contest, not an oracle. Markets routinely price in things consensus believes — sometimes correctly, sometimes not.

My Thesis

A structured form for recording your own conviction about a company across several dimensions: management quality, competitive moat, capital allocation, pricing power, valuation sentiment, and overall conviction. You can also write a free-form bull case and bear case.

Why we ask for structured ratings

Writing down your thesis before you buy and revisiting it later is one of the few habits that consistently distinguishes successful investors from unsuccessful ones. Forcing the rating into a 1–5 scale across consistent dimensions makes it possible to:

• Spot when your view has shifted (you rated something a 4, but six months later you're acting like a 2).
• Compare your conviction across the names in your portfolio.
• Aggregate ratings into the Crowd Sentiment view above so other users can see how the community feels.