Operating Signals

Through co-brand agreements: a bank (Amex, or a Visa issuer) pays the airline for every mile purchased and for marketing access to its customers. Those remuneration payments are a high-margin annuity — the airline sells miles at a large markup over the cost of eventually providing the seats — and the stream is far more stable than flying passengers. At some carriers it can rival or exceed passenger operating profit. The framework tracks co-brand remuneration, card-spend growth, and the loyalty deferred-revenue liability across American, Delta, United, and Alaska. Regional and contract carriers with no consumer program are silent by construction — there is no annuity to measure.

Three disclosures moving together: co-brand remuneration rising year-over-year, cardholder spend growing at double-digit rates, and the loyalty deferred-revenue liability building — that liability is future travel already paid for by the bank, a balance-sheet record of the annuity compounding. New-account growth adds a fourth. When these run in the same direction concurrently, the airline's earnings mix is quietly shifting from cyclical, capital-intensive flying toward a stable, high-margin payments-adjacent stream. The investment relevance is the multiple: markets price airlines on cyclical earnings, and a growing annuity inside a cyclical wrapper is the kind of margin-quality change that gets re-rated late.

Weak (M1): the program is material — co-brand remuneration is disclosed or the loyalty liability is broken out — and there is a directional inflection in co-brand spend or remuneration. Medium (M2): the inflection plus at least two concurrent same-direction signals from the set of double-digit spend growth, remuneration growing year-over-year, new-account growth, and a rising loyalty liability. Strong (M3): the annuity at very large scale — remuneration on the order of twice operating income, the configuration where the card program dwarfs passenger profit — with double-digit spend growth continuing. The extraction is heuristic from the carriers' own remuneration and liability disclosures in filings.

Slowly — which is the pattern's premise. The disclosures accumulate quarter by quarter, but the market's habit of pricing airlines on fuel, capacity, and fare cycles means the annuity's stability tends to be recognized in repricing episodes rather than smoothly. The pattern is bullish-only: it flags the compounding of the high-quality stream, and deterioration simply stops the firing rather than inverting it. Educationally, it is a lesson in reading segment quality inside a consolidated income statement — two companies with identical operating income can differ enormously in how much of it is annuity versus cycle. The Live Tape shows which carriers are firing today.

Less than you would expect, and that resilience is the whole reason the annuity carries a different valuation than the flying business it sits inside. Cardholder spend is driven by everyday consumer spending across all merchant categories — groceries, gas, dining, general retail — not specifically by how much flying that customer does, so overall card spend and the remuneration payments tied to it tend to hold up far better in a travel downturn than passenger revenue does. The issuing bank also has its own separate, independent incentive to keep growing the co-brand book, since the interest and fee income on cardholder balances is its own profit center that has nothing to do with airline traffic. That structural insulation is exactly what makes the loyalty stream more recession-resistant than the cyclical, capital-intensive business wrapped around it, and it's the specific quality this pattern exists to surface inside a consolidated income statement that never separates the two on its own.

Net new assets (NNA) — or net deposits — measure the organic flow of client money onto a platform, stripped of market appreciation. They lead revenue mechanically: assets gathered today bill management or platform fees tomorrow, and next year, and the year after. A firm can report growing total assets purely because markets rose while quietly bleeding client money — the organic number is the honest one. The framework reads the organic-growth tables the companies themselves publish and applies different logic by lane: high-growth disruptor platforms like Robinhood and SoFi are judged on the level of organic growth, while traditional managers like BlackRock, Ameriprise, Franklin, Invesco, Raymond James, and T. Rowe are judged on deceleration.

Organic growth slipping below the 5% line and running below its own prior rate — especially when the higher-fee book is the part bleeding. That combination means fee revenue softness is already contracted into the asset base before it prints: the assets that left were billing more than the assets arriving. A sharp deceleration — the organic rate dropping five percentage points or more from the prior period — is the medium-grade bearish signal even before the absolute level looks alarming. On the other lane, a disruptor compounding organic growth above 20% is the bullish read, and above 20% with total platform assets up 35% year-over-year is the full-stack strong version.

Weak (M1): a disruptor's organic rate is disclosed but running at or below 20% (modest bullish), or a traditional manager's organic rate is under 5% and below its prior rate (single-signal bearish). Medium (M2): the disruptor compounding above 20% organic (bullish), or a traditional manager decelerating sharply — prior rate minus current rate of five points or more (bearish). Strong (M3): the disruptor above 20% organic with total platform assets up 35% or more year-over-year — growth confirmed across both the flow and the stock of assets. The disclosure form itself identifies which lane a company is in, and crypto-beta platforms are excluded because their asset swings track coin prices, not gathering.

Fee revenue follows the asset base with a short lag — typically the next one to two quarters reflect the new run-rate, and the effect compounds for as long as the flow trend persists, because each quarter's NNA permanently re-bases the billable assets. That persistence is why the framework treats flows as a leading indicator rather than a coincident one: a deceleration that starts this quarter is a multi-quarter revenue statement, not a one-time miss. Whether the stock has priced it depends on how visibly the company discloses the organic split. The Live Tape shows which names in the cohort are firing which direction today.

By comparing a firm's organic rate against its own recent history rather than reading a single snapshot. The medium-grade bearish trigger specifically requires the deceleration to be sharp — the current organic rate falling five percentage points or more below the PRIOR period's rate — not merely being below the 5% level in isolation. A firm that dips modestly for one quarter and bounces back the next would not clear that five-point drop in a way that persists into the following period, because the comparison resets each quarter against whatever the new prior rate becomes. That structure is a meaningfully harder signal to trigger from noise than a static level check would be: it's measuring the RATE of change in the firm's own trajectory, and a genuine multi-quarter deceleration shows up as repeated, compounding drops rather than one isolated soft print that quietly self-corrects.

A pre-commercial biotech has no product revenue — it burns cash toward a binary catalyst, usually a clinical readout or regulatory decision. When the disclosed cash runway ends before that catalyst arrives, the company faces a forced raise: it must sell equity (diluting holders) or take on financing on whatever terms the market offers, at exactly the moment it has the least leverage. That overhang is a survival-binary the market tends to under-price while attention stays on the science. The framework reads the "Liquidity and Capital Resources" section and the going-concern note of the 10-Q and 10-K — where companies are legally required to state exactly this problem in standardized language.

The company tells you, in regulated language. The first flag is an explicit shortfall statement — words to the effect that existing cash is "not sufficient to fund operations" through the anticipated milestones. The second, more severe flag is going-concern language: "substantial doubt" about the company's ability to continue, a phrase auditors and management do not use casually. Two structural filters matter: the pattern applies only to pre-commercial names — once product revenue exists, self-funding dissolves the binary — and it stays silent when the disclosed runway explicitly covers the next catalyst. A biotech that is merely burning cash is normal; one that discloses it cannot reach its own catalyst is the pattern.

The engine currently fires this pattern at medium and strong only. Medium (M2) requires a pre-commercial company with explicit runway-shortfall language in its filings — the standardized "not sufficient to fund operations" class of disclosure. Strong (M3) requires that shortfall plus concurrent going-concern or substantial-doubt language — the most severe liquidity disclosure U.S. reporting rules provide. A weak level based on softer runway-versus-catalyst-date arithmetic is deferred until the engine wires the filing-date math. The grading is deliberately conservative: it fires only on what the company itself has stated in regulated filings, never on estimated burn rates or third-party runway models.

No — it says nothing about the science. It is purely a balance-sheet statement: the money runs out before the next value-defining event, so a financing event sits between today and the catalyst. That matters even if the trial succeeds, because the raise typically happens first, on pre-readout terms, and dilution permanently reduces what each existing share captures of any later success. Some funding gaps resolve well — partnerships, non-dilutive deals, favorable raises. The framework flags the overhang so an investor evaluates the financing risk separately from the scientific thesis, instead of letting excitement about the catalyst obscure the dilution that likely precedes it.

Yes, and that is the intended behavior rather than a gap the pattern misses. The trigger requires the company's own disclosed shortfall or going-concern language — words to the effect that existing cash is "not sufficient to fund operations" through the anticipated milestones, or explicit substantial-doubt language. A company that proactively raises capital well ahead of its catalyst, refilling its runway before a filing ever has to state that shortfall, simply never generates the disclosure the pattern requires, so it stays silent by construction. The framework is reading a fact management is legally obligated to state once the gap becomes real, not independently modeling burn rates to forecast a shortfall before the company discloses one — so a biotech that raises early specifically to stay ahead of that threshold is demonstrating exactly the financial discipline the pattern exists to distinguish from the ones that wait too long.

A stated change in how much cash a company intends to return to shareholders, announced as a forward policy target on an investor deck or IR page — "total payout ratio of 87%," "return at least 40% of free cash flow," "dividend payout of 20–30%." Like multi-year growth targets, these statements live only on investor-relations surfaces, invisible to filings-based screens. The framework compares the stated target against the trailing realized shareholder yield — dividends plus buybacks as a share of net income, computed from the actual financials. A target materially above the realized base signals more cash coming back than the run-rate implies (bullish); materially below signals a pullback redirecting cash to capex, M&A, or debt paydown (bearish).

Reinvestment can be value-creating, but the pattern is about the removal of a trained expectation. Sell-side models and income-oriented holders price the company on its realized return behavior — the dividends and buybacks the financials actually show. A forward policy set materially below that base withdraws the support those holders were positioned around, and history says the repricing tends to precede any payoff from the redirected cash. The bullish mirror is symmetric: a target set well above the realized base is an un-priced strengthening of the buyback-and-income floor. In both directions, the signal is the break between stated forward policy and demonstrated trailing behavior.

The unit is percentage points of total payout ratio — target versus trailing realized. Weak (M1): a gap of 20–30 points, with the higher side at least 1.4 times the lower, so the firing requires a genuine pivot rather than a restatement of the run-rate. Bullish when the target exceeds realized; bearish when it falls short. Medium (M2): a gap of 30–45 points, same logic. Strong (M3): a gap above 45 points — a decisive forward-policy break from the realized base. The realized side is computed from quarterly net income, dividend payments, and the latest fiscal-year buyback aggregate — actual cash behavior, not management's characterization of it.

The announcement is public the day the deck posts, but the framework's working observation is that consensus anchors to realized financials and re-rates slowly against stated forward policy — the pivot typically works into estimates and positioning over the following quarters as the new payout behavior starts printing. That lag is longest when the policy statement sits only on an IR page and never passes through a filing or a headline. Educationally, the pattern is a reminder that capital-return policy is a management promise with a paper trail: comparing the promise to the demonstrated base rate is mechanical, and the market does it later than you would expect.

Yes, and that is a real, acknowledged limitation rather than a hidden flaw. The pattern depends entirely on a company having stated a specific forward payout policy target on an investor deck or IR page; without that disclosure, there is nothing to compare against the trailing realized shareholder yield, so the pattern simply never fires on that name regardless of how its actual capital-return behavior evolves over time. Not every company publishes this kind of forward-looking capital-allocation framework, and those that don't are effectively invisible to this specific signal even if their underlying payout behavior is shifting in exactly the way the pattern is built to catch on names that do disclose one. An investor following a company with no stated target has to fall back on tracking the trailing dividend and buyback figures directly, quarter over quarter, rather than relying on this pattern to flag the shift.

Hold is the percentage of wagered money the casino keeps — its win rate. In high-roller VIP baccarat, hold swings meaningfully quarter to quarter from pure chance, so a casino's reported earnings can look great or terrible for reasons that have nothing to do with how many customers showed up or how much they bet. Operators publish an expected hold band and, when luck runs outside it, self-quantify the distortion — statements like "EBITDA would have been $X million higher had hold met expectations." The framework separates luck from demand: when hold prints below the band while turnover (actual betting volume) is flat to up, the reported number understates true demand.

When the miss is luck, not demand. If actual VIP hold came in below the operator's stated expected band while betting turnover held up, the casino's true business performed better than the income statement shows — and hold mean-reverts, because luck does. The next few quarters tend to snap back toward the band, making the reported comparison easy. The mirror is the trap: a blowout quarter driven by above-band hold is over-earning that reverts bearish, and investors who extrapolate it are annualizing a lucky streak. The framework covers Las Vegas Sands and Wynn, the operators that disclose turnover, a stated band, and the luck adjustment together.

Weak (M1) fires when actual VIP or rolling-chip hold prints outside the operator's stated band by at least half a percentage point with turnover flat to up — establishing that the deviation is luck-driven, not demand-driven. Medium (M2) adds the operator's own quantification: the filing states the hold-adjusted EBITDA impact, and hold below the band reads bullish for the reversion (above-band reads bearish). Strong (M3) is defined — corroboration of the reversion direction from the free monthly regulator series in Macau and Nevada — but is deferred, so the engine caps this pattern at medium today. The grading leans entirely on operator-disclosed numbers rather than estimated hold.

Usually one to three quarters — luck has no memory, so hold tends back toward the stated band as soon as enough volume is wagered. The practical horizon is the next one or two earnings prints, where the year-over-year or sequential comparison flatters (after bad luck) or punishes (after good luck) the reported numbers. The discipline the pattern teaches generalizes beyond casinos: whenever a company's results contain a disclosed randomness component, strip it out before extrapolating. The operators do the arithmetic for you in the filing; the market frequently trades the unadjusted headline anyway. That gap between disclosed and priced is where the framework lives.

Mass-market slot and table hold is set by the games' own fixed house edge, and because it's generated from millions of small, independent bets, the law of large numbers keeps it tightly clustered around its theoretical rate quarter after quarter — there is very little luck-driven distortion left to normalize once volume is that large. VIP and rolling-chip baccarat behaves completely differently: the betting is concentrated in a comparatively small number of very large individual wagers, so a single high-roller's outsized win or loss can swing the realized hold percentage meaningfully away from its long-run expectation within one quarter. That statistical fragility — a small sample of huge bets rather than a large sample of small ones — is precisely the condition that creates the luck-versus-demand confusion this pattern exists to untangle, and it's why the operators themselves publish an expected band and a luck adjustment for the VIP segment specifically, not for the mass floor.

Sell-in is what a company ships into its distribution channel — that is what reported revenue measures. Sell-through is what end customers actually buy out of the channel. The two can diverge for quarters at a time: when distributors are working down inventory (destocking), sell-in runs below real end demand; when they rebuild (restocking), sell-in overstates it. The tell is in the volume/price decomposition companies disclose: if reported growth is carried entirely by price while volume is negative, the channel is destocking. The framework tracks this across chemicals, agriculture inputs, machinery, and industrial distributors — sectors where the volume/price bridge is disclosed in filings and the channel is thick enough to matter.

Read the volume and price components separately, not the headline. The bearish signature is revenue growth that is entirely price — volume negative — often paired with management language about customers "reducing inventories." The bullish signature is the mirror: the volume component turning positive after a destocking stretch, which historically leads the reported revenue turn by one to three quarters. Direction is set by the sign of the volume component, not the headline number: a company can print +4% revenue while shipping 3% fewer units, and that is a deteriorating business wearing a price increase. Companies that disclose only qualitative volume language are excluded — the framework requires the numeric bridge.

Weak (M1) fires when the volume component crosses zero in the signaled direction by a small margin — under 3% — and the move is volume-led rather than price-led. Medium (M2) requires the volume component at 3% or more with volume as the dominant driver, meaning the volume swing is bigger than the price swing; that test exists to filter out cases where price pass-through, not demand, explains the print. Strong (M3) adds a volume component of 5% or more together with explicit destocking or restocking language from management — a deep inflection with the company confirming the mechanism in its own words. Where the filing bridge is silent, the engine reads investor-relations disclosures as a fallback, capped at medium.

The volume decomposition typically leads the headline revenue turn by one to three quarters — that lead is the entire reason the pattern exists. A destock ends when channel inventories reach target levels; a restock runs until they are rebuilt. Each phase usually spans two to four quarters in industrial and chemical channels. For an investor, the framework's flag means the reported numbers over the next few quarters are likely to inflect in the direction volume is already pointing, before the sell-side models catch up. Whether to act on that is a judgment about what is priced in. The Live Tape shows which cohort names are firing which side today.

They look identical in the volume/price bridge itself — negative volume with price carrying the headline — which is exactly why the two get confused. What separates them is duration and context. Destocking is a channel-inventory phenomenon: distributors are working down stock they already hold, it is typically flagged in management's own language about customers "reducing inventories," and it resolves on a bounded timeline of roughly one to three quarters once channel levels normalize. Share loss is structural: volume keeps declining well past that window, it tends to coincide with a named competitor reporting the volume growth the incumbent is missing, and there is no channel-inventory story that explains why it would reverse. The practical test is simply whether the volume decline actually turns around on the schedule destocking implies — if it doesn't, what looked like a channel adjustment was share loss wearing the same signature.

Deconversion and termination fees are exit payments a bank-technology processor collects when a client bank leaves — the contractual buyout of the remaining term. They land in revenue, so a quarter propped up by them can look like a beat. But each fee is doubly bad news: it is non-recurring by definition, and the payer is a lost future revenue stream walking out the door. The framework calls it income pollution — the sign-inverted mirror of one-time charges — because it masks franchise decay behind a healthy-looking print. A revenue beat built on exit fees is client attrition wearing growth's clothes.

The disclosures appear in successive 10-Qs and 10-Ks, usually as a named deconversion or termination-fee amount within the revenue discussion. The tell is the trend: a latest disclosed amount running well above the median of prior disclosures means client departures are accelerating, whatever the headline revenue says. The framework keeps the vocabulary deliberately tight — only deconversion and termination-fee revenue counts, so merger breakup fees and unrelated one-timers cannot false-positive the signal. The right mental model: subtract the fee from revenue and re-ask whether the quarter beat, then add the future revenue of the departed clients to the loss column.

Weak (M1): the latest disclosed deconversion or termination-fee amount exceeds 1.5 times the median of prior disclosing filings — elevated attrition, not yet a spike. Medium (M2): the latest amount above 2 times the trailing-filing median, with at least four historical disclosure points required so the median is meaningful. Medium is the ceiling at inception; a strong level — for example, a fee spike concurrent with visible organic-growth deceleration — is reserved until production observation supports a formal amendment. The conservatism is deliberate: the engine only grades what the company has itself disclosed across multiple filings, never inferred churn.

The fees are received immediately, but the lost recurring revenue disappears over the following quarters as departing clients complete their migrations — so the same event inflates today's print and deflates next year's comparisons. That timing inversion is the trap: the market sees the beat now and meets the shortfall later as if it were new information, when the spike had already announced it. The framework's flag is an early-warning reclassification — treat the fee revenue as a measure of attrition, not earnings power. Whether the attrition is idiosyncratic (a lost contract) or structural (a competitive share shift) is the follow-up question the firing should prompt.

Because a median computed from fewer than four data points is too easily distorted by a single unusual prior disclosure to serve as a trustworthy baseline. With only one or two historical figures, "twice the median" could mean almost anything depending on whether that one or two figures happened to be unusually high or unusually low in the first place — the framework has no way to tell a genuinely elevated baseline apart from a fluke on that thin a sample. Requiring at least four points before grading the medium level is a deliberate statistical floor: it ensures the comparison baseline is stable enough before the pattern claims a company's latest fee is meaningfully above normal for that specific name. A company with a thinner disclosure history can still be flagged at the weak level once it has enough history to compute any median at all; the higher-confidence medium grading is withheld until that baseline is solid.

The evidence says it depends on two things: size relative to the company, and why the program exists. A large-sample academic meta-analysis (Eshghi and Astvansh, 2024, covering 34,594 announcements) finds the market reaction scales with program size relative to the firm — absolute dollars are irrelevant — and that proactive programs draw neutral-to-positive reactions while reactive, demand-driven cuts are penalized. A "$300M program" headline means nothing until it is normalized: 5% of revenue at one company, 0.3% at another. The framework encodes exactly that: savings as a percentage of trailing revenue, gated on the proactive framing, extracted from the investor-deck disclosures where these programs are announced.

The framework's materiality floor is announced run-rate savings of at least 1% of trailing-twelve-month revenue. High-SG&A sectors — software, communications, healthcare — carry a raised bar of 1.6%, because their overhead runs roughly twice as heavy relative to revenue, so a given percentage cut is a smaller share of the addressable cost base. The economics of the upper levels are straightforward: at typical operating margins of 10–20%, a realized 5%-of-revenue savings run-rate translates to a double-digit percentage lift to operating profit — the level at which the program is not a footnote but the thesis. Programs are also handicapped for realization: announced savings rarely arrive in full.

Weak (M1): announced savings of at least 1.0% of trailing revenue (1.6% in high-SG&A sectors) — the canonical materiality floor. Medium (M2): at least 2.5% of revenue (4.0% high-SG&A) — sized to clear the floor even after applying roughly a one-half realization haircut to the announced number. Strong (M3): at least 5.0% of revenue (8.0% high-SG&A) — the level where the realized flow-through alone is a double-digit lift to operating profit at typical margins. The gate is always the ratio plus the proactive framing; a reactive cut announced into collapsing demand does not carry the same signal and is the configuration the academic record shows being penalized.

Programs are typically stated as run-rate targets one to three years out, and the flow-through builds gradually — severance and restructuring costs often land first, savings second. The framework's premise is that the sell-side is slow to model the realized flow-through of a proactive program, so forward operating margin gets lifted before consensus catches up. The investor's checkpoints are concrete: does the company report progress against the run-rate each quarter, and does the savings arrival survive contact with reinvestment (savings redirected into spend never reach the margin line)? The firing flags a program worth tracking; the tracking is where the judgment lives.

The dollar savings could genuinely be identical, but the market reaction the underlying research documents isn't really about the size of the savings — it's about what the announcement itself reveals about the company's situation. A proactive program reads as management choosing to improve efficiency from a position of relative strength, a deliberate forward-looking decision. A reactive program announced in response to collapsing demand reads as management being forced to cut because the business is already under acute stress, and that framing carries negative information about how bad the downturn is, entirely independent of the savings figure itself. Two programs of identical size can be priced very differently as a result, because the reactive one also functions as a confession that conditions are worse than previously disclosed, while the proactive one carries no such admission — which is why the pattern gates on the framing rather than on size alone.

The early bucket — accounts 30 or more days past due — is the first link in the consumer-credit loss chain. A borrower who misses one payment this quarter becomes a 60-day, then a 90-day account, and finally a charge-off over the following months, so the 30+ rate mechanically leads provisions and net charge-offs by one to two quarters. The information is in the rate of change, not the level: a low but rising early bucket says credit is deteriorating ahead of the reported loss numbers; a falling one says the loss line is about to improve. The framework tracks this for the U.S. card issuers that disclose it monthly or quarterly — Synchrony, Discover, Capital One, and American Express.

Reserves are a stock — the accumulated allowance sitting against the book, moved by management judgment and accounting rules as much as by fresh deterioration. The early-delinquency roll rate is a flow — the newest cohort of borrowers missing their first payments, before any management estimate intervenes. The framework maintains both as separate patterns precisely because they answer different questions: growing reserves tell you management is acknowledging risk; an inflecting early bucket tells you the raw material of future losses is changing right now. The early bucket typically moves first, which is why this pattern is classified as the leading read and why its trigger is the quarter-over-quarter inflection.

Weak (M1): the 30+ day rate inflects at least 15 basis points quarter-over-quarter off a multi-quarter base — a genuine rate-of-change move, not drift around a level. Medium (M2): an inflection of at least 20 basis points that is sustained, meaning the prior quarter's step moved the same direction — two consecutive same-signed moves rule out noise. Strong (M3): the sustained inflection plus the lagging net charge-off rate now also moving at least 20 basis points in the same direction in the same filing — the early bucket and the realized loss both visible at once, the full chain confirmed. Falling rates fire the bullish mirror at the same thresholds.

One to two quarters, mechanically — that is roughly how long it takes a 30-day account to roll through the buckets into a provision and a charge-off. The issuers publish the early data monthly in many cases, so the deterioration (or improvement) is on the public record well before the earnings release quantifies it in provisions. The pattern's educational point: consumer-credit earnings are among the most forecastable lines in the market if you read the leading bucket, yet the market routinely reacts to the provision print as if it were news. Both directions matter — an early-bucket improvement ahead of priced-in pessimism is as tradable an observation as deterioration.

The four issuers carry meaningfully different customer mixes, and that matters for the absolute LEVEL of delinquency each one runs at even in a healthy environment — American Express skews toward affluent, lower-loss-rate cardholders, while Synchrony, Discover, and Capital One each carry more mainstream and, in places, subprime exposure, so a given early-delinquency rate that would be alarming for Amex can be entirely unremarkable for a more subprime-weighted book. That's exactly why the trigger is built around a basis-point INFLECTION relative to each issuer's own multi-quarter base rather than a single absolute level shared across all four — the framework isn't comparing issuers against one common threshold, it's comparing each one against its own recent trend, which is what makes the signal genuinely comparable across credit books that differ substantially in composition.

By raising fees on a declining volume base. A marketplace's disclosed volume metric — GMV, GMS, GSV — measures the actual commerce flowing through the platform; revenue is that volume times the take rate. When volume declines year-over-year for most of a year while the take rate climbs 150 basis points or more, the company is monetizing a shrinking pie: reported revenue can look flat or even up while the underlying franchise contracts. The framework calls this the exact inverse of the payments take-rate pattern — there, volume grows while the fee erodes; here, fees rise while the volume dies. The second configuration is the dangerous one.

Because the fee squeeze itself accelerates the shrinkage. Sellers on a subscale marketplace face rising costs on falling sales, which pushes the marginal seller to leave, which thins selection, which reduces buyer traffic, which cuts the remaining sellers' volume further — inviting the next fee hike to fill the revenue hole. Each turn of that loop is individually rational for management and collectively corrosive for the franchise. The signature to watch: the volume metric down year-over-year in at least three of the last four disclosing filings while the take rate rises. Revenue is the last line to show the damage, which is exactly why the framework reads volume and take rate separately.

Weak (M1) fires on the volume flag alone: the disclosed volume metric declining year-over-year in at least three of the four most recent disclosing filings, without the take-rate confirmation. Medium (M2) adds the squeeze: the same volume decline with the take rate up 1.5 percentage points or more year-over-year — fee hikes visibly masking the shrinkage. Medium is the ceiling at inception: the strong level, which would key off an accelerating divergence between the two series, is reserved until production observation justifies a formal amendment. The take rate is computed directly as reported revenue divided by the disclosed volume, so the signal cannot be managed away by presentation.

Usually several quarters to a couple of years — take rates have a ceiling set by seller economics, and each hike buys less revenue as the base erodes. The pattern tends to end in one of two ways: a visible revenue decline once pricing runs out of room, or a strategic reset (fee cuts, repositioning) that makes the shrinkage plain retroactively. For an investor, the flag means the revenue line is currently the least informative number the company reports; the volume metric and the take-rate trend carry the real state of the franchise. The Live Tape shows any cohort names firing this pattern today.

The dynamic is corrosive, but not automatically terminal — there are two visible paths out, and both leave a trace in the same disclosures the pattern reads. The first is a genuine strategic reset: management proactively cuts fees or restructures the platform to rebuild seller economics, accepting near-term revenue pain specifically to stop sellers and buyers from continuing to leave. The second is repositioning — pivoting toward a different customer segment, category, or business model where the marketplace's remaining scale is actually competitive, rather than trying to defend a shrinking version of the original franchise. Either move shows up as the volume metric stabilizing or turning back up alongside a stated fee cut or strategic shift, which is the signature that the spiral has actually been broken rather than merely paused for a quarter. Absent one of those two moves, the mechanism the pattern describes — each fee hike inviting further seller attrition — tends to keep compounding on its own.

Management's forward view is the highest-information part of an earnings release, and it lives in the outlook section and prepared remarks — not in the standardized financial data. A guidance cut is management telling you, in its own words, that the trajectory is lower than previously stated: an explicit "lowering our full-year guidance," a next-period growth guide materially below the company's own trailing realized growth, or a margin outlook revised down. The framework treats this as a standalone bearish pattern regardless of valuation: a quality company that guides its outlook lower fires it just as a cheap one does, because the information content is in the reset itself.

Because consensus adjusts slower than the disclosure. The framework's documented observation is that sell-side estimate revisions and institutional positioning trail a visible guide-down by one to two quarters — the stock stays anchored to the prior trajectory while the new one works its way through models, price targets, and portfolio decisions. That lag means the first reaction frequently under-reacts, especially when the cut is quantified in a deceleration rather than a headline phrase. It is also why the pattern distinguishes a genuine reset from a generic cautious tone: vague conservatism does not fire; a quantified deceleration or explicit guide-down language does.

Weak (M1): an explicit guide-down phrase, or a quantified deceleration — forward revenue or bookings growth guided at least 10 percentage points below trailing year-over-year growth while trailing growth is still positive. A cautious release with neither does not fire. Medium (M2): the 10-point deceleration corroborated by explicit guide-down language or a concurrent margin guide-down — or both top line and profitability guided down qualitatively. Strong (M3): a severe reset — the forward guide at least 20 points below trailing, or guidance to an outright revenue decline — confirmed by a margin guide-down or explicit acknowledgment. The engine parses the outlook language directly from the earnings materials.

The framework flags the reset and its severity; the decision is yours. Some cuts are one-time and honest — a demand air pocket management sizes accurately — and the stock washes it out in a quarter. The pattern's warning is about the base rate: first cuts are often not last cuts, because the same forces that made the prior guidance wrong tend to persist, and the consensus lag means the full estimate reset takes quarters. The useful discipline is to compare the cut against the company's own trailing delivery rather than against the stock move: a 10-point deceleration is a fact regardless of whether the price fell 3% or 15% on the print.

It can be either, and the pattern deliberately doesn't try to attribute the cause — it fires purely on the fact and size of the disclosed reset, whether management's stated reason is company-specific (a botched launch, a lost customer, a pricing misstep) or macro (a sector-wide demand air pocket, a currency headwind, a tariff shift). Separating "our own execution problem" from "the whole macro backdrop moved against everyone" is a judgment call the framework leaves to the reader rather than building into the trigger, because two companies in the same sector citing the identical macro cause are informative about the sector as a whole in a way one company's execution miss is not. The practical distinction for an investor once the flag fires is durability: a cited macro cause the whole sector faces is likely to persist across peers, while a company-specific miss is more plausibly one-off — and telling those apart is exactly the reading the pattern hands off rather than resolves.

RPO is the signed-but-not-yet-recognized order book — revenue that is contracted but will only hit the income statement as the work is delivered, typically over the next one to two quarters and beyond. Because revenue lags bookings by that recognition cycle, RPO inflects before reported revenue does. The subtlety the framework enforces: for a healthy subscription or backlog business, RPO growth persistently above revenue growth is the normal, already-priced condition. The signal is the inflection — the gap widening as the order book accelerates away from the revenue print, or the order book shrinking while revenue still grows, which is the pre-rollover read before the P&L turns.

Compare two quarter-pairs, not one snapshot. A wide-but-decelerating gap — RPO growing 50 points faster than revenue but slower than it did last quarter — is momentum the market has already marked; the framework explicitly does not fire on it. What fires is acceleration: RPO growth exceeding revenue growth by at least 10 points and speeding up versus the prior quarter-pair. On the bearish side, watch for the order book decelerating hard while revenue still grows — revenue is coasting on past bookings. The data is a clean XBRL number, so the comparison is arithmetic, not narrative parsing. Software-sector bearish reads and defense backlogs are owned by other patterns to avoid double-counting.

Weak (M1): RPO growth leads revenue growth by at least 10 points and is accelerating — roughly the gap the trade literature has long treated as notable momentum. Bearish weak is the mirror: a 10-point lag, decelerating, while revenue still grows. Medium (M2): a lead of 20 points or more with acceleration of at least 5 points — or the cap level whenever total RPO exceeds 2.5 times annual revenue, because long-duration multi-year contracts inflate the order book without near-term revenue effect. Strong (M3): a lead of 30 points or more, accelerating by at least 10 points, with revenue itself still growing — or, bearish, RPO growth turning outright negative while revenue growth is still positive, the terminal rollover ahead of the print.

One to two quarters is the normal recognition lag, which is what makes the divergence a leading indicator rather than a coincident one. The framework's broader observation is that the lag in market pricing is longer than the accounting lag: sell-side estimates and institutional positioning tend to trail a visible bookings inflection by one to two quarters even though the disclosure is public the day the filing lands. That gap between publicly visible information and consensus adjustment is precisely what the pattern exploits — educationally, it is a demonstration that "the market knows everything instantly" is a slogan, not a description of how backlog disclosures get priced.

Yes — it is the clearest current expression of the bullish leg. The canonical live cases in the framework's own material are industrial, not software: Caterpillar's power-generation order book and Quanta's grid and data-center backlog, where contracted orders have been compounding away from already-growing revenue prints. That is the M3 shape — order book accelerating on top of growing revenue. Contrast that with the mega-cap software names, where enormous RPO-over-revenue gaps exist but are decelerating and long priced. The pattern's discipline is exactly that distinction: it fires on the industrial names where the gap is widening, and stays silent on the famous names where it is merely large.

It means holding or raising price while volume falls, instead of chasing freight with discounts. In less-than-truckload (LTL) trucking, the two numbers to watch are yield excluding fuel (revenue per unit of freight, stripped of fuel surcharges) and tonnage per day. A carrier that keeps yield rising while tonnage troughs is repricing its book — and that discipline historically leads the operating-ratio improvement and earnings recovery by two to four quarters, arriving before volume does. The inverse is the bearish read: tonnage rising while yield falls is a carrier buying market share with cheap freight, which shows up later as margin damage. The framework covers Old Dominion, Saia, and XPO.

Put yield-ex-fuel and tonnage side by side each quarter — both are in the operating-statistics table. The disciplined signature is yield up while tonnage is down: management is accepting less freight at better prices and defending the network's economics through the trough. The volume-chase signature is the mirror — tonnage up, yield down — which feels like recovery on the revenue line but is share bought with price, and it structurally damages the industry's pricing when several carriers do it at once. The strongest confirmation is the operating ratio improving year-over-year despite falling volume: cost and price discipline showing up in the profit metric while the freight recession is still on.

Weak (M1): yield-ex-fuel growing but under 3% year-over-year while tonnage per day is negative — nascent discipline at the volume trough. Medium (M2): yield-ex-fuel up 3% or more with tonnage still negative — the historically disciplined-carrier band, price firmly leading volume. Strong (M3): yield up 5% or more, tonnage still negative, and the operating ratio improving year-over-year despite the decline — discipline already converting into profitability before the freight cycle turns. The bearish volume-chase reads fire on the inverted configuration. All inputs come from the carriers' own disclosed operating statistics, so the signal is arithmetic, not sentiment.

Freight downturns typically run several quarters to a couple of years, and the payoff sequence is well ordered: pricing discipline shows up first in yield, then in the operating ratio within two to four quarters, then in earnings as tonnage eventually returns to a repriced network. The carrier that held price through the trough recovers with structurally better margins; the one that chased volume recovers with a degraded book it must spend years repricing. For an investor, the pattern's value is timing-independent of the macro call — you do not need to predict when freight demand returns, only observe which carriers will be worth more when it does.

Because LTL is genuinely a management-controlled pricing decision in a way the other freight modes aren't. LTL carriers set rates through published tariffs and long-term contracts across a dense hub-and-spoke network, so holding or raising price through a downturn reflects a deliberate choice about how to defend the network's economics. Full-truckload pricing, by contrast, is set largely in a fragmented spot market where price mostly reflects the immediate supply of available trucks and freight loads — "discipline" there is much harder to separate from a carrier simply riding whatever the spot market happens to offer, since no single carrier controls that outcome the way an LTL carrier controls its own tariff. Rail and air cargo run on their own distinct capacity constraints and contract structures entirely, which would need their own separate instrument rather than a re-application of this one. The framework scopes to LTL because that is where price genuinely is the management decision the pattern is trying to read.

Capital-instrument placements are the razor; the high-margin recurring consumables and service revenue they generate are the blade. When a diagnostics or medtech company places instruments or grows procedure volume, that installed base pulls through consumable revenue for years — and the placement number leads the recurring-revenue number by several quarters. The framework reads the leading flow (placements, procedure growth, installed-base growth) as the early signal: expanding placements today are next year's consumable revenue, visible before it prints. The cohort is medtech and diagnostics names with a durable installed base — companies like IDEXX, Intuitive Surgical, Align, Edwards, and Dexcom. Pure-consumable businesses with no placed instrument are excluded by construction.

Watch the leading KPI, not the revenue line. Bullish: installed-base or procedure growth accelerating to 8% or better — that expansion mechanically pulls through recurring revenue over the following quarters. Bearish: the same KPI decelerating into single digits with management itself flagging the slowdown — the framework requires the management acknowledgment, because a single soft quarter of placements can be timing noise. The recurring-revenue mix matters too: the pattern only carries weight where the blade is genuinely high-margin and recurring, which is why the disclosure of recurring mix is part of the weak-level gate. Revenue can look fine for two or three quarters after placements stall; that lag is the trap.

Weak (M1) establishes the setup: an instrument base exists, the leading flow KPI is low-positive — roughly 3–6% installed-base or procedure growth — and the recurring mix is disclosed. Medium (M2) is the inflection: the leading KPI accelerating to 8% or more (bullish), or decelerating to single digits with a management deceleration flag (bearish). Strong (M3) is a decisive move: the leading KPI at 15% or more with a second KPI confirming (bullish), or procedure growth at 2% or below alongside the management flag (bearish). The extraction is heuristic from quarterly operating tables — no narrative guessing — and the direction is set entirely by the polarity of the leading flow.

Several quarters — that is the mechanical lead the pattern is built on. An instrument placed this quarter generates consumable and service revenue starting next quarter and compounding as utilization ramps, so an acceleration or deceleration in placements is a forecast of recurring revenue two to four quarters out that the company has effectively already published. The framework flags the inflection and its severity; it does not tell you to buy or sell. The educational point is about where to look: for these business models, the operating table in the quarterly filing carries more forward information than the income statement above it, and the market frequently prices the income statement first.

It reads whatever figure the company itself discloses as its installed-base or procedure-volume metric, and for the covered cohort that disclosure is typically already a net number — companies in this space generally report active systems or cumulative placements net of retirements, not raw unit shipments. The framework does not separately model churn on top of the disclosed KPI; it trusts the company's own reported figure rather than reconstructing net growth from shipment counts, which is precisely why the weak-level gate requires that a leading KPI actually be disclosed in the first place. A company placing many new units while quietly losing older systems to replacement or churn would show the true, netted trend in the metric the pattern reads — a gross-shipments number that ignored retirements would overstate the real installed-base tailwind, which is the failure mode the disclosure requirement is built to avoid.

Because it carries dramatically higher margins than the underlying commerce. When a marketplace sells ad placements to its own sellers, the incremental revenue arrives with almost no incremental cost — so as advertising grows as a share of the volume transacted on the platform, operating margin expands mechanically, typically showing up two to four quarters after the mix shift begins. The framework tracks ad revenue as a percentage of gross merchandise volume (GMV) and, critically, the growth premium — how much faster ad revenue grows than GMV itself. The cohort is marketplaces that separately disclose both numbers, names like eBay, Amazon, MercadoLibre, Sea, and Alibaba; platforms that bundle advertising into a generic services line stay silent by construction.

Two numbers moving together: ad penetration of GMV rising, and ad revenue growing meaningfully faster than GMV. The growth premium is the sharper tell — a platform whose advertising grows 10 or 15 points faster than its commerce volume is monetizing existing traffic at higher margin, not just riding volume. Penetration gives the runway context: below 2% is nascent with years of headroom; at 3–5% the shift is established and compounding. This is a bullish-only pattern by design — saturation of the ad load suppresses the firing rather than inverting it into a bearish signal, because a mature ad mix is a level, not a deterioration.

Weak (M1): ad penetration of GMV is disclosed and ad revenue is growing faster than GMV — a positive premium — but penetration is still nascent, under 2%; an Amazon-style multi-year disaggregation of advertising services trending up also qualifies. Medium (M2): penetration at 2% or more with the ad-versus-GMV growth premium at 10 points or better — the mix shift is established and material. Strong (M3): penetration at 3% or more (5% for a platform of Amazon's scale) with the premium at 15 points or better — a decisive, margin-accretive mix shift that historically leads operating-margin expansion by two to four quarters. All of it is extracted from the company's own disclosed figures.

The mix shift leads reported operating-margin expansion by roughly two to four quarters — advertising booked today flows through to the margin line as it scales against a fixed cost base. Whether it is priced is the investor's question, but the framework's premise is that markets systematically under-model mix: they extrapolate the blended margin rather than decomposing it, so a business whose highest-margin segment is compounding fastest tends to beat the extrapolation for several quarters running. The pattern flags exactly that condition. The ETF X-Ray will also decompose this at the holdings level if you own the cohort through a fund rather than directly.

That is a real, well-documented risk in marketplace economics, and it is a genuine limitation of this pattern taken on its own: the ad-mix signal by itself has no countervailing decay built in for a degraded shopping experience, because it only measures whether advertising is growing faster than GMV and expanding as a share of it, not whether buyers are quietly disengaging as a result. The risk this pattern can't see directly shows up somewhere else in the same company's disclosures — decelerating GMV growth, softening buyer-engagement metrics, rising customer complaints about search relevance. Which is exactly why this signal is meant to be read alongside those other numbers rather than in isolation: a marketplace posting an accelerating ad mix against decelerating GMV is showing the tension the ad-mix pattern by itself cannot capture, and that combination is a genuinely different and more cautionary read than the ad-mix figure alone.

Two documented mechanisms. First, debt overhang: in a leveraged company, part of every dollar of operating improvement accrues to creditors as reduced default risk rather than to shareholders — the academic treatments (Cai and Zhang, 2011; DeAngelo, Gonçalves, and Stulz, 2018) show that a credible path to lower net leverage transfers that value back to equity and tightens credit spreads, with the payoff largest for high-leverage, credit-stressed names. Second, credibility: proactive deleveraging is the empirical norm — roughly 93.7% of stated deleveraging is executed through debt repayment and earnings retention — so a stated target is usually a plan, not a wish. The market, anchored to the current credit profile, prices the de-risking late.

The specific shape of the statement. The strongest form is a quantified "from current to target" — "from 4.5x to below 3.0x net debt to EBITDA" — because it states both the starting point and the size of the intended move in turns of leverage. The weaker but still meaningful form is explicit deleveraging language paired with a target that is conservative for the sector. The framework reads only these disclosed forms and never estimates the current leverage itself — a discipline choice, since EBITDA definitions vary and an inferred starting point would let the signal fire on arithmetic the company never endorsed. The statements live on investor decks and IR pages, outside the filings feed.

Weak (M1): a stated from-current-to-target delta of 0.5–1.0 turns of net debt to EBITDA, or deleveraging language whose target still sits high for the sector — near the falsifiability boundary. Medium (M2): a stated delta of 1.0–2.0 turns, or deleveraging language with the target landing at or below the sector's conservative leverage ceiling; the language-only path caps here without a quantified delta. Strong (M3): a stated delta of 2.0 turns or more of intended deleveraging — a decisive credit-improvement signal that requires the company to have quantified its starting point. The pattern is bullish-only; leverage increases are covered by other parts of the catalog.

The balance-sheet work itself typically runs one to three years — debt matures and gets repaid, retained earnings accumulate. But the equity benefit often front-runs the completion: credit spreads tighten as the plan demonstrates progress, the equity's risk discount narrows, and the debt-overhang transfer accrues quarter by quarter rather than at the finish line. The framework's premise is that the sell-side anchors to the current credit profile and models the forward de-risking slowly, so the stated target is un-priced information for a while after it posts. Free registration shows which tickers are currently firing this and every other pattern in the catalog.

The pattern fires on the stated target at the moment it's disclosed and doesn't itself track ongoing follow-through, which is a real limitation worth naming plainly. What corroborates against abandonment being the likely outcome is the base rate the pattern's own design leans on: roughly 93.7% of stated deleveraging targets are executed through debt repayment and earnings retention, a high enough completion rate that abandonment is genuinely the exception rather than the rule. Still, an investor holding a position on the strength of this signal should check the company's own subsequent quarters for whether reported net leverage is actually moving toward the stated target — a company that keeps restating the same ambition while net debt to EBITDA never budges is showing the minority case the base rate warns is possible, and confirming or contradicting that requires tracking the reported leverage ratio itself over time, not the initial firing alone.

Spread-based life and annuity carriers earn the gap between what their investment portfolio yields and the rate they credit policyholders. That net investment spread is the earnings engine for the annuity book. Its forward driver is the reinvestment gap: the difference between the yield available on new money today and the yield of the bonds rolling off the portfolio. When new-money yields sit well above the portfolio yield, every maturing bond gets reinvested at a better rate, and the spread widens mechanically for years as the portfolio turns over. When new-money yields sit below, the same machine runs in reverse. The framework covers spread-material carriers — MetLife, Prudential, Principal, Equitable — and excludes protection-driven names like Aflac and Globe Life where the spread book is immaterial.

The MD&A spread and yield disclosures: the stated portfolio yield, the crediting rate, and — most forward-looking — any disclosure of new-money yields versus the roll-off yield. The direction of the reinvestment gap is the signal: a favorable gap of half a percentage point or more means the earnings tailwind is already locked into the portfolio math, quarter after quarter, as assets mature and reprice. The market's habit is to value these carriers on book-value multiples that under-model the multi-year earnings drift the gap implies — in either direction. A narrowing or negative gap is the bearish mirror: spread compression grinding through results long after the rate environment that caused it.

Weak (M1): the spread state is disclosed and there is a single-quarter inflection — a reinvestment gap (new-money yield minus portfolio yield) of roughly 50 basis points in either direction, or a disclosed spread-attribution pointing one way. Medium (M2): the gap at roughly 75 basis points with the direction sustained rather than one print. Strong (M3): a gap of roughly 100 basis points or more — a deep, multi-quarter spread tailwind (or headwind) that will roll through earnings as the portfolio turns over. The magnitudes are approximate by design, since disclosure formats vary by carrier, and the direction of the gap sets the bullish or bearish read.

Years — and that duration is the point. An insurance portfolio turns over slowly, so a 100-basis-point reinvestment gap does not arrive as one good quarter; it compounds as each maturing tranche reprices, producing a multi-year earnings drift that is largely predetermined by today's yield math. That predictability is what the market's book-value-multiple habit under-weights: two carriers at the same book multiple can face opposite multi-year spread trajectories that are already visible in their disclosures. The framework flags the direction and depth of the gap; judging what the current price assumes about it remains the investor's work.

The direction tends to be similar across the sector, but the magnitude and timing depend heavily on each insurer's own portfolio shape, specifically how much of it is set to mature and roll off in the near term versus locked into longer-duration holdings. A carrier whose book is concentrated in bonds maturing soon will feel a falling-rate environment's negative reinvestment gap quickly and acutely, because a large share of its portfolio reprices at once; one with a longer average duration spreads the same effect out over more years, making it less acute in any single quarter even though the direction is the same. So while the broad rate environment sets the general tone for the whole sector, the actual gap this pattern reads is company-specific — driven by each carrier's own disclosed portfolio yield and maturity schedule — which is exactly why the framework reads individual filings rather than inferring the effect from the macro rate decision alone.

It is distributable cash flow divided by the distributions actually declared — how many times over the company's cash generation covers what it pays out. A ratio of 1.4x means the company generates $1.40 of distributable cash for every $1.00 distributed, leaving retained cash to self-fund growth without leaning on capital markets. The framework treats this as the vertical-specific leading number for midstream MLPs and C-corps — names like Enterprise Products, Energy Transfer, Williams, and Enbridge — because the market tends to screen these stocks on commodity beta and yield while under-weighting the coverage math that actually determines whether the payout survives.

Coverage sliding toward 1.0x, which historically leads a cut by one to two quarters. The comfortable self-funding band is roughly 1.3–1.5x; a ratio drifting into the 1.10–1.20x range and flat-to-compressing is the pre-cut watch zone; below 1.10x the company is under the minimum the MLP market has traditionally demanded; and below 1.0x sustained, the company is paying out more than it generates — arithmetic that only leverage or asset sales can bridge. Add a stretched balance sheet (debt above 4.5 times EBITDA) and the cut risk becomes imminent rather than theoretical. The bullish read is the mirror: coverage stable at 1.3x or better signals payout durability the yield screen misses.

Weak (M1) fires two ways: bearish when coverage sits in the 1.10–1.20x band and is flat-to-compressing (the pre-cut watch), or bullish when coverage is at 1.30x or better and stable (the self-funding signal). Medium (M2) is bearish-only: coverage below 1.10x, under the traditional MLP minimum — the leading signal a cut decision is approaching. Strong (M3): coverage below 1.0x sustained with a leverage amplifier — debt above 4.5 times EBITDA — the imminent-cut configuration where the payout is being funded by the balance sheet. The engine extracts distributable cash flow and declared distributions from the earnings exhibits directly.

The typical lead is one to two quarters once coverage is decisively below 1.10x — boards defend distributions as long as they credibly can, then cut once the math forces the decision. That lag is the opportunity and the warning: the coverage deterioration is published in the earnings materials well before the cut announcement, but yield-focused holders often see only the still-attractive headline yield, which is highest precisely when the cut is closest. Educationally, the pattern is a case study in reading the funding of a payout rather than its size. The framework flags both directions, since durable 1.3x+ coverage on a name priced for a cut is information too.

Because sub-1.0x coverage on its own still leaves an escape valve: a company with a genuinely underleveraged balance sheet could bridge a temporary shortfall by drawing on unused credit capacity or issuing new debt without meaningfully straining its credit profile, buying time without real distress. What removes that escape valve — and what the leverage amplifier is specifically testing for — is a company already carrying a stretched balance sheet (debt above 4.5 times EBITDA) at the same time it's paying out more cash than it generates. A company in that combined position can't self-fund the gap out of cash flow and can't cheaply borrow more to bridge it either, which is the configuration where a distribution cut moves from merely likely to close to unavoidable. Requiring both conjuncts together is what separates a genuinely concerning coverage ratio from one that has already run out of ways to buy time.

Companies state multi-year financial ambitions on investor-day decks and IR pages — "5–7% EPS CAGR through 2030," "double revenue by 2030," "roughly 20% operating margin by 2030." These targets live only on investor-relations surfaces; they never appear in a quarterly filing or the standardized data feeds, so most systematic screens never see them. The framework extracts them and measures the gap between the targeted growth rate and the company's trailing realized growth rate. A target far above the demonstrated run-rate is one of two things: an over-promise that consensus will fade as reality lags (bearish), or a credible raised bar the market has not yet modeled (bullish).

The single-snapshot discriminator is the most recent trajectory: is the latest year-over-year growth already tracking the target pace? A company targeting 15% growth off a 6% trailing base, whose most recent year printed 14%, is showing the acceleration that makes the target credible — the bullish read. The same target with the latest year still at 6% is ambition unsupported by delivery — execution and credibility risk, the bearish read. The framework's threshold for "already tracking" is the latest year-over-year growth at or above roughly 85% of the target pace. Sell-side models anchor to trailing growth either way, which is why the gap itself is under-priced in both directions.

Weak (M1): the target CAGR exceeds trailing realized CAGR by 3–6 percentage points and is at least 1.7 times the trailing rate — a genuine stretch, not rounding. Direction is set by the recent-trajectory test: bearish when the latest year-over-year growth is below about 85% of the target pace, bullish when at or above it. Medium (M2): a gap of 6–10 percentage points, same direction logic. Strong (M3): a gap above 10 points — a decisive divergence between stated ambition and demonstrated trajectory. EPS-based targets use net income as the per-share proxy, a conservative choice that ignores buyback-driven share-count help toward the target.

Targets fade slowly — that is the mechanism. A company missing the pace toward a 2030 ambition does not announce the miss; it simply keeps printing quarters below the required rate while the deck stays on the IR page. Consensus estimates drift down over several quarters as the arithmetic becomes undeniable, and the credibility discount arrives with them. The bullish leg resolves faster: an accelerating run-rate that validates a raised bar tends to force estimate revisions within a few quarters. Either way, the pattern's edge is simply reading a public document the standardized data pipeline ignores. The Live Tape shows current firings and their direction.

This is exactly why EPS-based targets are measured against net income rather than reported EPS itself. A company can shrink its share count through buybacks and mechanically lift reported EPS growth even while the underlying business — net income — isn't accelerating at all, which would let a stagnant operating business look like it's tracking an ambitious EPS target purely through capital allocation rather than actual performance. By comparing the target and the recent-trajectory test against net income instead of the per-share figure, the framework deliberately strips out that buyback tailwind and asks whether the un-levered business itself is accelerating toward the stated ambition — a more conservative test that a company can't pass through financial engineering alone, however useful buybacks might be for other reasons.

Because an online travel agency recognizes revenue when the stay happens, not when the booking is made — so the booked-volume number (room-nights or nights booked, reported net of cancellations) runs one to three quarters ahead of the revenue the income statement shows. A quarter of decelerating bookings is a forecast of soft recognized revenue two or three quarters out, published in the same release as a perfectly healthy-looking revenue line. The framework tracks this lead across Airbnb, Booking Holdings, and Expedia. TripAdvisor is excluded (its model is reviews and experiences, not booked stays) and Trip.com is excluded because its foreign-filer disclosures do not carry an extractable unit series.

Three tells, in escalating order. First, booked-volume growth decelerating while management itself acknowledges softening or moderating demand — the framework requires that acknowledgment, and checks that the booking window is not lengthening, because a lengthening window can make bookings look soft without demand actually weakening. Second, gross-bookings dollar growth falling below booked-volume growth (pricing is weakening under the units) or the booking window actively shortening (travelers committing later — classic caution). Third, and most structural: a rising cancellation rate, which means the gap between what was booked and what will actually be stayed and recognized is widening. Each added tell moves the firing up a level.

Weak (M1): booked-volume is disclosed and decelerating, management concedes softening demand, and the booking window is not lengthening — the clean early tell without mechanical excuses. Medium (M2): the weak conditions plus corroboration from the dollars — gross-bookings growth running below booked-volume growth, or the booking window shortening. Strong (M3): all of that plus a rising cancellation rate, meaning the booked-versus-stayed gap is turning structural rather than cyclical. The pattern is bearish-only: it exists to surface demand softening ahead of the P&L, and the extraction is heuristic from the companies' own disclosed operating metrics, not from web-scraped or third-party travel data.

One to three quarters — the average gap between booking and stay. That makes this one of the more mechanically reliable leads in the framework: the revenue weakness is not a possibility to be handicapped, it is largely already contracted (or, with rising cancellations, un-contracted) and waiting to be recognized. The educational takeaway is to read the operating metrics table before the income statement in these releases: the market's first reaction often anchors on recognized revenue and the current-quarter beat, while the bookings deceleration sitting two lines lower is the actual news. The Live Tape shows when any of the covered names is firing.

The mechanical lead exists in both directions, but the informational asymmetry mostly runs one way. A travel company with accelerating bookings and a lengthening booking window typically volunteers that story proactively — management highlights it on the call, guidance reflects it, and analysts update models on the spot, so the good news tends to already be reflected quickly in both the disclosure and the price. Softening demand is the case companies are more likely to frame cautiously rather than loudly flag, and it's also where the mechanical accounting lag — revenue recognized on the stay, not the booking — creates the widest gap between what's already knowable from the operating table and what's priced into the stock. The framework concentrates its coverage where that gap between disclosed fact and market reaction tends to be largest, and for this particular lead-lag relationship that is consistently the deceleration side.

The take rate is what a payments company keeps of each dollar it processes — transaction revenue divided by total payment volume. Headline volume (TPV, GPV) can grow briskly while the blended take rate erodes underneath, because mix shifts toward lower-yield products, larger merchants negotiate better pricing, or incentives to win volume climb. The arithmetic is unforgiving: 10% volume growth against a take rate compressing a few percent nets to much less revenue growth than the headline suggests, and the erosion compounds. The framework computes this from the companies' own disclosed volume and revenue grids — PayPal's five-quarter tables, or client incentives outpacing revenue at Visa and Mastercard.

Divide transaction revenue by volume yourself, every quarter, and watch the trend in basis points — companies rarely headline it. The early tell is volume growing while the blended take rate compresses even a few basis points, or contra-revenue lines (client incentives at the networks) growing faster than net revenue. The pattern is bearish-only by design: a rising take rate on rising volume is simply a healthy business and produces silence, not a bullish firing. The distinction from a shrinking marketplace matters too — this pattern is volume up, yield down; the inverse (volume down, fees up) is a different and nastier pattern in the catalog.

Weak (M1): volume still growing while the blended net take rate compresses at least 3 basis points in a single quarter, or client incentives outpacing net revenue — the first visible leak. Medium (M2): volume growth of 7% or more with the take rate compressing at least 8 basis points year-over-year — the celebrated headline now demonstrably masking erosion. Strong (M3): compression of 12 basis points or more year-over-year — deep, multi-quarter, mix-driven erosion that compounds into forward transaction-margin pressure. The covered cohort is PayPal, Mastercard, Visa, Affirm, and Block, the names whose disclosures let the arithmetic be done cleanly from filings.

The framework flags the condition; it does not issue instructions. Some erosion is strategic and survivable — deliberately taking lower-yield enterprise volume for scale, or funding incentives to defend share — and some is competitive pricing pressure that never stops once it starts. What the pattern insists on is that you not credit the volume headline without netting the yield against it, because the market habitually over-weights the big round volume number and under-weights the basis-point leak that determines forward margin. A firing is an invitation to ask which kind of erosion this is. Free registration shows current firings and their magnitude.

They're mirror images of each other, and the sibling pattern names the relationship explicitly from its own side. There, volume is falling while the take rate is rising — a company propping up its revenue line on a shrinking pie. Here, volume is growing while the take rate is compressing — a company's headline metric looking healthy while the yield underneath quietly erodes. Both patterns exist for the same underlying reason: the direction of volume by itself tells you almost nothing about whether unit economics are improving or worsening, because volume and yield can move in opposite directions and either one alone can mask what the other is doing. Reading only the volume headline — up is good, down is bad — is exactly how an investor misses both of these configurations; the framework requires the pair, computed together, before it will fire either one.

Because rock is too heavy to ship far. Aggregates — crushed stone, sand, gravel — cost more to transport than to produce beyond a short radius, so each quarry operates as a regional toll-road with little practical competition inside its zone. That local-monopoly structure lets a disciplined producer raise average selling price per ton even while shipment volumes decline in a construction downturn — and make the increase stick. The framework reads the two numbers producers publish in every earnings release, price per ton and shipment tonnage, and looks for exactly that split: price up 5% or more year-over-year while volume falls, held for two consecutive quarters.

Persistence into falling volume. Anyone can raise price when demand is booming; the tell of structural power is a price increase that holds while shipments decline — customers pay more for less rock because there is no realistic alternative supplier within trucking distance. The framework requires both anchors: price per ton up at least 5% year-over-year and tonnage down year-over-year, at the latest and the prior disclosing quarter. One strong quarter can be mix or a contract reset; two consecutive quarters of price-into-falling-volume is a producer demonstrating that its pricing is a decision, not a cyclical gift.

Weak (M1): average selling price per ton up 5% or more year-over-year with shipments down year-over-year, both conditions met at the latest and prior disclosing quarter — the baseline discipline signature. Medium (M2): the same two-quarter configuration with price per ton up 8% or more at both anchors — decisive pricing power through the downturn. Medium is the ceiling at inception; a strong level — for instance, confirmation through gross-profit-per-ton expansion — is reserved for a future amendment after production observation. The inputs are the producers' own published operating statistics, so the read is arithmetic rather than narrative.

Operating leverage on a repriced base. A producer that pushed price through the trough enters the recovery with a permanently higher revenue-per-ton foundation, so returning volume multiplies against better unit economics — the earnings recovery is steeper than the volume recovery. That is the bullish logic of firing during the downturn rather than after it: the pricing evidence is available precisely when the market is discounting the sector on cyclical volume fears. The pattern is bullish-only; producers that discount into weakness simply do not fire. As always, the framework flags the mechanism — what the current price already assumes is the investor's half of the work.

Not permanent, but unusually durable, because what protects it is physical geography and permitting rather than anything a competitor can simply choose to do differently. A rival could in principle undercut local pricing by opening a competing quarry close enough to serve the same customers within an economical trucking radius, but new aggregates permits are notoriously hard to obtain — environmental review, local zoning opposition, and the plain scarcity of geologically suitable rock deposits near the population centers that actually need the material mean new supply rarely enters an established quarry's zone. The more realistic threat isn't a price war from a new entrant; it's a shift in underlying demand severe enough that even a local monopoly can't hold price through it, or gradual substitution toward recycled or alternative materials in some applications. Absent one of those, the geographic moat the pattern reads tends to persist for a very long time.

US single-family housing starts — a Census series, readable via FRED — measure ground actually broken on new homes. For companies whose revenue is new construction, the government tape moves before the 10-Q does: starts today become builder revenue recognized over the following one to two quarters as the homes complete and close. The framework applies the read-through only to new-construction-pure names — the homebuilders themselves, whose volume effectively is starts, plus Builders FirstSource at roughly 70% new-residential revenue. Building-products companies levered to repair-and-remodel are excluded by construction, because their demand does not track the starts series.

Smoothing plus persistence. The engine takes the 3-month average of single-family starts and its year-over-year growth rate; the signal fires when that growth crosses plus-or-minus 5% and holds at both the latest and the prior monthly print. Two consecutive prints beyond the threshold filter out the single-month noise the series is famous for — weather, permitting timing, revisions. Direction follows the tape: sustained acceleration reads bullish for the cohort, sustained deceleration reads bearish. The macro turn becomes decisive at plus-or-minus 10%. Everything comes from a free public series published monthly, weeks to months ahead of any company report.

Weak (M1): the 3-month-average starts growth at or beyond plus-or-minus 5% year-over-year at both the latest and prior monthly print — a directional read-through, bullish on acceleration, bearish on deceleration. Medium (M2): the same persistence with the latest move at or beyond plus-or-minus 10% — a decisive macro turn. Medium is a hard ceiling by design: this is a lead indicator about the cohort's future revenue environment, not a confirmed company-specific event, and the framework caps confidence accordingly. A strong firing requires something true about the company itself; a macro series, however clean, cannot supply that on its own.

One to two quarters, set by the construction and closing cycle — a home started today is revenue when it closes. That makes the read-through one of the more mechanical macro-to-micro links available: the demand change is not being forecast, it is being counted by the Census while the builders' own reports still reflect the prior quarter's tape. The discipline the pattern encodes is cohort hygiene — the read applies only where starts genuinely are the business. Applying the same series to a kitchen-remodel-levered products company produces noise, which is why the registry is deliberately narrow. Free registration shows the current cohort firings.

Because single-family and multi-family construction are driven by genuinely different forces, and the covered cohort — the pure homebuilders plus Builders FirstSource — is overwhelmingly exposed to single-family, owner-occupied construction rather than apartment or rental development. Multi-family starts are shaped more by institutional and rental-market economics — cap rates, rental-vacancy trends, construction financing for large multi-unit projects — than by the consumer demand and mortgage-rate forces that drive single-family home purchases. Blending the two series together would muddy the read-through for a cohort whose business is specifically single-family, which is exactly the same cohort-hygiene discipline the pattern already applies by excluding remodel-levered products companies. A pattern built around multi-family starts would need its own distinct cohort of apartment developers and residential REITs; it isn't a natural extension of this one.