Aramark (NYSE: ARMK) — A Cost-Plus Caterer Re-Rated to a Record on Deleveraging and a Data-Center Dream

Independent fundamental research. Report date: 2026-07-17. The analysis is skeptical, evidence-driven, and competitive-advantage-centric.

The main body of this article carries no buy/sell recommendation and no price target. The single, deliberately-labeled exception is the Claude's Take block immediately below.

⚡ Claude’s Take

The author’s own independent opinion and general information — not investment advice. The main body of this article below carries no position.

Verdict: HOLD / AVOID-here. A genuinely better-run, de-risked business than it was three years ago — but now priced as if the improvement is a permanent compounding engine rather than a one-time repair. Great story, wrong price. Not a short. Accumulate-on-weakness in a ~$40–47 zone (≈11–13x forward EV/EBITDA on ~$1.5–1.6B FY26 EBITDA, ≈16–19x forward adjusted EPS), which is roughly 20–30% below the current $56.99.

Aramark is the #3 player in a business where scale economics reward only #1. Compass Group earns a ~7.2% operating margin; Aramark earns ~4.3%; that ~200–300bp gap is the financial signature of a company operating in the shadow of a larger, structurally-advantaged competitor. Return on invested capital is ~6.8% — at or below the cost of capital. By every durable-quality test I care about, this is a thin-margin, labor-arbitrage services roll-up with high customer retention but no pricing power (management says it outright: “we don’t price for profit,” and renewals reprice down). What has actually happened over the last two years is not a moat widening — it is a balance-sheet repair: net leverage fell from ~9x at the COVID trough to ~3.25x, the Vestis uniform business was spun off, and operating momentum (record 96%+ retention, ~5–6% net new business) has been genuinely strong. That is real, and I respect it. But it is a repair, and repairs re-rate once.

The problem is the multiple. At $56.99 the stock trades at its richest valuation in its entire public history — ~15x EV/EBITDA (vs. an 11–12x history), ~26x forward adjusted EPS, and on its own-history valuation range it sits at roughly the 98th percentile. It has closed — and in places exceeded — its historical discount to the higher-margin, higher-return market leader. The entire ~+60% move is H1 FY26, a momentum blow-off driven by a Q2 guide-raise and a genuinely intriguing but entirely unbooked “Aramark Nexus” AI-data-center services option that carries no revenue and NDA’d terms. The market is now paying for (a) deleveraging that has already happened, (b) a growth algorithm running at a cyclical/incentive-flattered high, and © free Nexus optionality. That is a lot to underwrite at a record multiple on a ~7%-ROIC business. Framing: a quality-momentum winner priced beyond its fundamentals — not a falling knife, not a compounder, a fully-valued repair story. Conviction: medium. What flips me bullish: Nexus converting into real, disclosed, above-corporate-margin revenue at scale (genuine mix-shift, not a headline). What flips me bearish: the first net-new or margin miss, which — given momentum crowding and a record multiple on ~7% ROIC — de-rates fast.

Tag: “The scale laggard, finally priced like the scale leader.”

📈 Stock Price Action — Five-Year Event Map

Factual price history, not a recommendation. Price moves are FACT; attributed drivers are INTERPRETATION.

Aramark is a complete round-trip and then some. From a COVID-era trough of ~$20.13 (June 2022), the stock spent roughly two years range-bound between the low-$20s and low-$40s, then went nearly vertical in the first half of fiscal 2026 to an all-time high of $58.15 (13 July 2026). It closed at $56.99 on 16 July 2026, inside a 52-week range of $35.73–$58.15 and just ~2% off its record high. Critically, essentially the entire period of outperformance is concentrated in the last five months (≈+60% since February 2026) — this is a recent, momentum-driven re-rating, not a slow compounding.

Cycle narrative. The 2022 low (#1–#2) is a leverage-and-cyclicality story: a ~9x-levered discretionary-exposed name in a rate shock, then a mechanical volume recovery. The Vestis spin (#3) is a bookkeeping step-down, not value destruction. Events #4–#5 are a slow, choppy grind as the market waited for the deleveraging to prove out. The decisive repricing is #6–#7: the Q2 FY26 report in May 2026 combined a beat, a full-year guide-raise, a leverage milestone, and the Nexus data-center reveal, igniting a momentum move that carried the stock ~+40% in nine weeks to an all-time high — closing the historical valuation discount to Compass Group. Today’s price sits at the very top of that move.

1. Executive Summary

Aramark is the world’s #3 contract food & facilities services provider (behind Compass Group and Sodexo), a ~$18.5B-revenue outsourcer of dining, catering, and facility management to Education, Healthcare, Business & Industry, Sports/Leisure/Corrections, and Facilities end-markets across 16 countries. Following the September 2023 spin-off of its Vestis uniform-rental business, it is a focused food-and-facilities pure-play.

The business is structurally mediocre and there is no durable moat. Gross margin is ~8.4%, operating margin ~4.3%, and net margin ~1.8% — this is a labor-intensive, cost-plus/P&L services book where the client holds pricing power at every rebid. Return on invested capital is ~6.8%, at or below the cost of capital, and has been for years. The company’s own filings admit that new business “carries lower profit margin” and renewals come “on terms less favorable or less profitable.” Customer retention is genuinely excellent (96.3% in FY25, a record; >98% in Q2 FY26) and there is a real multi-decade tailwind from first-time outsourcing of self-operated kitchens — but high retention coexists with zero pricing power. That is switching friction, not a moat. The decisive proof: scale leader Compass earns ~7.2% operating margin to Aramark’s ~4.3% — scale economics in this industry reward only the largest player, and Aramark is #3.

What has changed is the balance sheet, not the moat. Net leverage has fallen from ~9.3x (FY21 COVID) to ~3.25x (covenant basis, FY25), with a target of <3x by FY26-end. Revenue has compounded from $12.1B (FY21) to $18.5B (FY25) on volume recovery, ~5–6% net new business, and pricing pass-through. Adjusted operating income has grown ~12% and adjusted EPS ~20% in FY25 (to $1.82), with FY26 guidance calling for organic revenue at the high end of 7–9%, AOI +12–17%, and adjusted EPS of $2.18–2.28 (+20–25%). Management has resumed buybacks (~$140M in FY25, first meaningful repurchase since pre-COVID) and raised the dividend 14%. Capital allocation is competent value-preservation — a well-executed repair — but the M&A roll-up has only ever bought scale that earns ~WACC, not excess returns, and tangible book equity is deeply negative (~−$3.7B).

The problem is entirely the price. At $56.99 Aramark trades at ~15x EV/EBITDA and ~26x forward adjusted EPS — its richest multiple ever (own-history valuation composite: 98th percentile), having closed its historical discount to the higher-quality Compass. The market is capitalizing (a) deleveraging that has already happened, (b) a growth algorithm at a cyclical/incentive-flattered high, and © free optionality on “Aramark Nexus,” a hyperscale AI-data-center services platform that management says could become its largest client but which carries no booked revenue and NDA’d terms. Consensus is bullish and the tape is a low-vol/dividend quality-momentum winner (m6 return +120% annualized, only −7.6% max drawdown). The asymmetry from here is unfavorable: the deleveraging equity-transfer is spent, the multiple is at a record on a ~7%-ROIC business, and the first growth or margin miss de-rates a crowded momentum name quickly. This article takes no position (see Claude’s Take for that); the body documents the embedded expectations.

2. Business Overview

Aramark provides outsourced food services (dining, catering, retail food, concessions, vending) and facilities services (custodial, plant operations & maintenance, grounds, energy management, capital project management) under multi-year contracts. It operates two reporting segments:

• FSS United States — FY25 revenue $13,211.9M (~71% of total), segment operating income $717.5M (5.43% margin).
• FSS International — FY25 revenue $5,294.4M (~29%), segment operating income $193.5M (3.66% margin), across ~16 countries; more heavily management-fee/labor-intensive (personnel cost ≈ 51% of international revenue).
• Corporate expense: $119.2M.

US revenue by end-market (FY25), with growth and margin texture:

How it makes money — two contract structures (the key margin/risk lever):

• Profit-and-loss (P&L) contracts ≈ two-thirds of revenue. Aramark books all revenue and bears all cost (sometimes paying the client a percentage of revenue or a minimum guarantee). Full operating risk, higher margin and upside, sensitive to volume and food-cost inflation.
• Client-interest / management-fee contracts ≈ one-third. The client reimburses cost and pays a fee. Lower risk and lower margin.

Revenue quality. The book is annuity-like in volume but not in price: FY25 client retention was 96.3% (a company record; Q2 FY26 >98%), and net new business was 5.6% of prior-year revenue ($1.6B gross wins, +12% YoY, including the largest single FSS US contract in company history — a major medical system commencing early 2026). Contracts in Education and Sports/Leisure run 5–15 years, often with up-front Aramark capital (leaseholds, equipment, client grants); on termination the undepreciated capital is reimbursed by the client, so exit is not financially punitive to either side. This structure produces stable volumes but caps pricing power — see .

Procurement. Aramark buys through its Avendra group-purchasing organization and distributes ~43% of US/Canada purchases through Sysco under a >40-year relationship — a real but non-proprietary scale lever shared, in kind, by all three global players.

Post-Vestis identity. The September 2023 spin-off of Vestis (NYSE: VSTS, uniform rental) removed the one route-dense, higher-return, capital-light-recurring model from the portfolio, leaving a more purely commoditized food-and-facilities company. Aramark received a ~$1.4B cash distribution from Vestis in connection with the separation, which mechanically cut its leverage.

Verdict: A large, diversified, recurring-volume outsourcer with a genuine first-time-outsourcing tailwind and record retention — but a low-margin, cost-plus economic model in which growth is margin-dilutive and price is set by the client.

3. Industry Dynamics

Market structure. The global contract food-services market is ~$320B and fragmented: the #1 player, Compass Group, holds <15% share. The industry is an oligopoly only at the national/multinational tier — Compass, Sodexo, Aramark — sitting above a long tail of regional operators and, most importantly, the ever-present self-operation alternative (clients running dining in-house). Operators frame roughly half of the addressable market as still self-operated, which is the secular growth engine: Compass reports that about half of its new wins come from clients outsourcing for the first time. (One third-party market report claims ~78% is already outsourced — a definitional mismatch we flag as an open question; the operator framing of “~50% still self-op” is the more conservative and more useful runway estimate.)

The secular tailwind is real. First-time outsourcing is a durable, multi-decade conversion story driven by clients’ desire to shed non-core labor, capex, and food-safety/regulatory burden. This gives the industry a structural volume growth rate above GDP and is the single best thing about owning any of the big three.

But the economics are capped by three structural features:

1. Labor intensity. This is fundamentally a labor-arbitrage and management business — International personnel cost is ~51% of revenue; Aramark has ~39,000 unionized US/Canada employees. Wage inflation is the dominant cost, and while contracts allow pass-through, it is frictional (client consent, lag) and it inflates revenue without adding margin.
2. Client pricing power at every rebid. Contracts are competitively re-bid; public-sector work is re-bid by law. The client, not the operator, sets terms at renewal — and Aramark’s own 10-K states renewals come “on terms less favorable or less profitable than the initial contract terms.”
3. A permanent low-cost substitute. Self-operation is always available. This is a ceiling on how much any operator can charge; if outsourcing gets too expensive, clients in-source.

Capital-cycle lens (Marathon). Capital is not the binding constraint here — the business is low-capital-intensity (capex ~2.6% of revenue, and much of that client-reimbursed). The binding constraint is contract pricing, and competition among three well-capitalized nationals plus self-op pins returns near the cost of capital. There is no supply-driven capital-cycle mispricing to exploit; the industry is in a steady, competitive equilibrium.

Verdict: A structurally mediocre-to-OK industry — a genuine multi-decade outsourcing tailwind bolted onto a low-margin, labor-intensive, client-priced service. Good enough to grow volumes reliably; not good enough to generate excess returns on capital for the #3 player.

4. Competitive Position

Name the moat: there isn’t a durable one for Aramark. In Greenwald’s taxonomy, the only genuine advantage available in this industry is economies of scale coupled with customer captivity — and it accrues to the leader, Compass, not to #3 Aramark.

• Scale/procurement advantage is real but goes to #1. Aramark’s purchasing scale (Avendra, Sysco) is genuine but shared in kind by all three nationals and is smaller than Compass’s (~$40B+ revenue) and Sodexo’s. The largest buyer buys cheapest; being sub-scale versus the leader means no proprietary cost advantage.
• The margin gap is the proof. Compass FY25 underlying operating margin ~7.2% (with organic growth +8.7% and net new +4.5%); Sodexo FY24 ~4.7%; Aramark FY25 GAAP operating margin 4.28%, and even its most favorable adjusted US-segment margin (~6.4%) trails Compass’s group figure. The scale leader out-earns Aramark by ~150–300bp. That is the textbook financial signature of scale economies that reward only the largest player.
• Switching costs are real but modest. 96.3%+ retention and 5–15-year contracts demonstrate stickiness — clients don’t switch operators casually because of embedded systems, on-site capital, and transition risk. But renewals reprice down, exit capital is reimbursed, and public contracts are re-bid by statute. High retention with no pricing power is switching friction, not a moat.
• No network effects, no consumer brand premium, no route density. The diner does not choose Aramark; the institution does, on price and service. There is no two-sided network, and the route-dense Vestis model was spun off.

The moat-to-financial-outcome test. A durable competitive advantage must show up as ROIC persistently above the cost of capital. Aramark’s ROIC is ~6.8% (FY25), up from 4.6% (FY23) and 5.2% (FY24) — improving, but at or below any reasonable WACC. If the “moat” (retention, scale, tech platform) were real and durable, it would have generated excess returns; it has not. The combination of a persistent ~200bp margin deficit to Compass and ROIC≈WACC is the financial fingerprint of a company without a durable advantage, operating in the shadow of one that marginally has it.

Verdict: No durable competitive advantage. Aramark is the #3 scale player in an industry where scale economics reward only #1. What it has — record retention and a genuine outsourcing tailwind — stabilizes volume but does not convert into pricing power or excess returns on capital. Any bull thesis must rest on margin self-help, deleveraging, and (speculatively) Nexus — not on a moat.

5. Growth History and Forward Opportunities

Historical growth has been strong but low-quality by composition. Revenue compounded from $12.1B (FY21) → $13.7B (FY22) → $16.1B (FY23) → $17.4B (FY24) → $18.5B (FY25), a ~11% CAGR. Decomposing it:

• Volume recovery off the COVID trough (2021–2023) was the largest early driver — a rebuild, not new value creation.
• Pricing/inflation pass-through inflated revenue in 2022–2024 without adding proportional margin (recall: no pricing power, just cost pass-through).
• Net new business of ~5–6% of prior-year revenue is the genuine engine — real market-share and outsourcing-conversion wins. But management concedes new business “carries substantial start-up costs and lower profit margin,” so growth dilutes margin in the near term.
• M&A was a minor contributor — acquisition cash was only $54–289M/year (tuck-ins like Next Level Hospitality, Union Supply, HHA Services), which is the right (disciplined) posture but means growth is overwhelmingly organic.

Margin is grinding up, slowly. Gross margin rose 7.83%→8.36% and EBITDA margin 5.94%→6.85% (FY22→FY25) — real operating leverage, but only ~50bp over three years, and FY25 adjusted operating margin expanded just ~25bp. This is a slow, hard-won grind, not a step-change.

Forward opportunities (the bull’s list):

1. First-time outsourcing conversion — the durable, multi-decade volume tailwind.
2. Net new business momentum — Q2 FY26 retention >98%, $1B of gross wins YTD (ahead of pace), the largest-ever FSS US contract ramping in early 2026.
3. Margin self-help — supply-chain/GPO leverage, technology/digital ordering, labor productivity. Management targets 30–40bp/year of margin expansion (FY25 delivered ~25bp).
4. International momentum — a longer runway of underpenetrated outsourcing markets.
5. “Aramark Nexus” — a new hyperscale AI-data-center services platform (worker housing, dining, transportation, site services for data-center construction and operation). Management says it could become the largest single client in the portfolio, at >$100M/site annualized, at above-corporate margin, and capital-light/cost-reimbursable. Crucially, Nexus carries NO booked revenue, NDA’d terms, and is NOT in guidance — it is a genuine option, and the single largest swing factor in either direction (and, we suspect, a meaningful part of what the market is now paying for).

Verdict: Reliable, above-GDP volume growth of mixed quality. The organic net-new engine is genuinely good and the outsourcing tailwind is real; but a large share of historical “growth” was volume recovery and inflation pass-through, new business is margin-dilutive, and the most exciting forward lever (Nexus) is unproven. High growth, medium quality.

6. Financial Quality

Economics improve with scale — but only modestly, and off a very low base. The multi-year margin walk is the core evidence:

*FY23 GAAP EPS ($2.57) is flattered by Vestis spin-related gains; continuing-operations EPS was ~$1.72. Normalize this out.

Quality-of-earnings flags (three, all material):

1. The GAAP→AOI wedge is ~$189M (+24%). Management reports on Adjusted Operating Income (AOI) of $981.2M vs. GAAP operating income of $791.8M. The bridge: + acquisition-intangible amortization $124.6M (recurring, and rising as the roll-up compounds), + severance $36.4M, + gains/settlements $28.3M. Management is compensated on a number that adds back the amortized cost of its own acquisitions — a legitimate non-cash add-back, but one that flatters the return on an M&A-built asset base.
2. “Covenant Adjusted EBITDA” is more aggressive still — $1,464.8M vs. reported EBITDA $1,268M. It adds back SBC (~$58M) and a vague $125.6M “Other.” Management’s headline “~3.25x, near target” leverage uses this number; on clean reported EBITDA, net leverage is ~3.76x. Both are true; the reader should know which is quoted.
3. FCF is real but flattered by payables float. FY25 CFO of $921M included a +$196M accounts-payable build — and this is recurring (prior years +$120M / +$203M / +$433M / +$513M). Stripping the AP tailwind, “clean” CFO is ~$725M and FCF closer to ~$236M versus the ~$432–454M headline. Working-capital float is a finite tailwind that reverses if growth stalls.

ROIC reconciled. Against ~$8.2B of invested capital — of which $6.9B is goodwill and intangibles — Aramark earns ~6.8%. Return on tangible capital is high (the physical asset base is light), but there is no excess return on the capital actually deployed to build the company. This is the central financial fact: the business does not create value above its cost of capital.

Balance sheet. Total debt ~$5.72B; net debt ~$5.61B (Q2 FY26). The weighted-average interest rate steps up from ~4.03% (FY26) toward ~5.73% (FY29) as cheap tranches roll into a 2028–2030 maturity wall — a refinancing headwind that will partly offset the deleveraging benefit to EPS. EBITDA/interest coverage has improved to ~3.48x (from 1.79x in FY21). Tangible book equity is deeply negative (~−$3.7B; tangible book/share ~−$14), and reported ROE (~93%) is a meaningless artifact of a near-nil equity base — use ROIC/AOI, not ROE.

Verdict: Economics improve with scale, but slowly and from a low base, and the improvement is partly cosmetic (adjusted metrics, payables float). The business generates real cash but earns only ~WACC on deployed capital. Financial quality is adequate, not high.

7. Capital Allocation

The through-line is competent value-preservation, not value-compounding.

• Deleveraging is the whole capital-allocation story of the last four years. Net leverage fell from ~9.3x (FY21) to ~3.25x (covenant) / ~3.76x (clean) at FY25, with a stated target of <3x by FY26-end. Important nuance: the sharp FY23 drop was largely mechanical — the Vestis spin took debt off Aramark’s balance sheet and paid Aramark a ~$1.4B distribution. Only the FY24–FY25 improvement is truly organic deleveraging from FCF. Still, it has been executed competently and has materially de-risked the equity.
• M&A is a disciplined tuck-in roll-up that has only ever bought ~WACC returns. Acquisition spend of $54–289M/year (often earn-out structured) has cumulatively built the $6.9B goodwill-and-intangible pile that now earns ≈ its cost of capital. Management bought scale, not excess returns. Disciplined in price and size, but value-neutral in outcome.
• Shareholder returns are thin and recently resumed. The dividend is small (~$0.42/share, ~0.8% yield, ~34% payout) and was raised 14% in FY25. Buybacks resumed in FY25 ($500M authorization; ~4.0M shares / ~$140M repurchased — the first meaningful buyback since pre-COVID; ~$194M YTD in FY26). SBC is clean — ~0.3% of revenue, ~14% of FCF, non-dilutive on net (share count roughly flat).
• The pre-2020 LBO legacy was value-destructive. The private-equity-era leverage produced FY20–FY21 net losses at 9x+ leverage; the current management regime (CEO John Zillmer, CFO Jim Tarangelo) has spent four years repairing that damage.

Executive incentives (FY25 proxy) — above-average design. The annual incentive plan (AIP) pays on AOI, Free Cash Flow, Net New Sales, and ESG; the long-term PSU plan uses ROIC, FCF, Adjusted Revenue, Net New Sales, and Relative TSR — genuinely aligned with the deleveraging-and-returns thesis (the inclusion of ROIC and relative TSR is a positive). Caveats: FY25 AIP paid 133% of target despite AOI missing its goal (FCF and ESG carried the payout), and AOI adds back deal amortization. CEO total compensation was ~$14.5M (summary comp table) / ~$8.9M (compensation-actually-paid) — reasonable for an $18.5B-revenue enterprise.

Verdict: Competent, shareholder-conscious capital allocation in the current regime — but value-preserving, not value-creating. Management repaired a broken balance sheet well and is now returning modest cash; but the core capital-deployment engine (M&A + reinvestment) earns only its cost of capital. Give management credit for the repair; do not confuse the repair with a compounding machine.

8. Changes and Headwinds — Last Two Years

Strategic and structural changes:

• Vestis spin-off (Sept 30, 2023) — the defining structural change; created a focused food-and-facilities pure-play and mechanically cut leverage.
• Record operating momentum — FY25 retention 96.3% (record), net new 5.6%, $1.6B gross wins including the largest-ever FSS US contract; Q2 FY26 retention >98%, $1B wins YTD.
• Deleveraging milestone — net leverage below ~3.25x, “lowest since before the 2007 LBO,” per management; target <3x by FY26-end.
• Buyback resumption and 14% dividend increase (FY25) — the first capital-return normalization post-COVID.
• “Aramark Nexus” (unveiled Q2 FY26, May 2026) — the AI-data-center services platform (see ). Management frames it as potentially the largest client in the portfolio, capital-light and above-corporate-margin, but it is not in guidance and has no booked revenue. Treat as an option, not a forecast.
• FY26 guidance (raised at Q2): organic revenue growth at the high end of 7–9%, AOI +12–17%, adjusted EPS $2.18–2.28 (+20–25%), leverage <3x.

Headwinds and watch-items:

• No pricing power — management explicitly “does not price for profit”; inflation pass-through is frictional and margin-neutral.
• Sports/Leisure/Corrections margin compression — the largest US sector saw margins fall from high-single to mid-single-digit; open question whether cyclical or structural.
• New-business margin drag — the largest-ever FSS US contract and other 2026 ramps carry start-up costs that dilute near-term margin.
• Refinancing wall — the weighted-average debt rate steps up into the 2028–2030 maturity schedule, partly offsetting deleveraging’s EPS benefit.
• Discretionary/cyclical exposure — Sports catering, higher-ed enrollment demographics, B&I return-to-office trends, and corrections political risk are all sensitivities.
• Agency labor and wage inflation — the dominant cost line.
• Lost contracts happen — Aramark walked away from the University of Kentucky contract (deemed uneconomic), a reminder that retention statistics include disciplined exits but also competitive losses.

Verdict: The last two years strengthened the balance sheet and the operating narrative — genuinely — but did not change the underlying moat or margin structure. The changes justify a better business; they do not justify a different kind of business. The headwinds are ordinary for the model; the one genuine unknown is Nexus.

9. Risk Analysis

Key-person / governance: low-to-medium — CEO Zillmer engineered the turnaround; a transition would be a sentiment risk, though the bench and incentive design are sound. Insider signal is quiet-constructive (see / ): over 24 months, only one open-market purchase (CEO Zillmer, ~$250K at ~$39 in Aug 2025) and three sells by one officer (GC) — no broad or director selling, but no conviction buying at these prices either.

10. Valuation Discussion (Embedded Expectations)

Where the multiple sits. At $56.99, market cap is ~$15.0B and enterprise value ~$20.6B (equity + ~$5.6B net debt).

• EV/EBITDA ~15.3x TTM (~14.5x forward) versus an own-history range of 11–12x (FY23 11.1x, FY24 13.2x, FY25 12.0x) — the richest in company history (own-history valuation composite 98th percentile; P/E 42.6x = 95th, P/B 4.63x = 99.8th, P/S 0.78x = 99.8th).
• GAAP P/E ~42x; adjusted P/E ~31x trailing / ~26x forward (on FY26 adj EPS $2.18–2.28).
• The historical discount to Compass has closed. Aramark now trades at/above Compass’s ~13–15x EV/EBITDA despite ~150–200bp lower margin and ROIC ~6.8%≈WACC. Sodexo remains far cheaper (~7–8x EV/EBITDA). On relative quality, Aramark should trade at a discount to Compass, not parity.

What the current price embeds. To hold today’s ~$20.6B EV at a normalized 12x EV/EBITDA, EBITDA would need to reach ~$1.7B — roughly two more years of the guided growth algorithm executed cleanly. In other words, the current multiple is borrowing ~2 years of forward growth and assuming the multiple does not revert to its own history. Embedded in $57 is: (a) multi-year continuation of ~7–9% organic growth and 30–40bp/year margin expansion; (b) the deleveraging benefit (already largely realized — the 12x→15x re-rate is the market pricing the delevered, faster entity); and © meaningful free optionality on Nexus. The deleveraging equity-transfer is essentially spent as a future driver — it has already happened.

Scenario analysis (illustrative; FY28 adjusted EPS × exit multiple — NOT price targets):

Sell-side price targets currently cluster $60–70.5 (Citi $70.5, Oppenheimer $60–65), i.e., near/above the base case — the Street is underwriting close to the bull path. The bear case ($43–44) is simply own-history-normal multiple on modestly-normalized numbers — not a disaster scenario, which is what makes the risk/reward at $57 asymmetric to the downside.

Verdict: The price embeds continued flawless execution and a permanently-elevated multiple and Nexus optionality, on a ~7%-ROIC business. The deleveraging story that justified the re-rating is now in the price. Embedded expectations are demanding. (No price target; see Claude’s Take for a subjective zone.)

11. Variant Perception

Consensus / prevailing bull view: Aramark is a de-risked compounder — record net-new business, >98% retention, a durable first-time-outsourcing runway, GPO/technology-driven margin expansion, deleveraging below 3x, and free AI-data-center (Nexus) optionality. On this view the record multiple is justified by a structurally better, faster-growing, less-levered company, and the Street’s $60–70 targets follow.

The strongest bear case (the variant view): This is a thin-margin (~5% AOI), commoditized labor-arbitrage services business with ROIC ≈ WACC and no pricing power, trading at a record multiple on cyclically- and incentive-flattered peak numbers. The margin deficit to Compass is structural (scale rewards only #1); inflation pass-through is not pricing power; FCF is flattered by payables float; adjusted metrics add back real recurring costs; and the deleveraging that drove the re-rating is a one-time repair now fully reflected. Nexus is a pre-paid, NDA’d option the market is capitalizing before a dollar of revenue. The stock is a crowded low-vol/dividend quality-momentum winner, and the asymmetry is a de-rate on the first growth or margin miss.

The 3–5 assumptions that actually matter:

1. Net new stays ~4–5% and retention >96% — the volume engine. Falsified by: a couple of quarters of net-new deceleration or a retention slip below 96%.
2. Durable 30–40bp/year margin expansion — FY25 delivered only ~25bp. Falsified by: flat/compressing adjusted margin as new-business start-up costs and wage inflation bite.
3. Nexus becomes real, disclosed, above-corporate-margin revenue — the optionality. Falsified/confirmed by: FY27 disclosure of actual Nexus revenue and margin (or continued vagueness).
4. The multiple holds ~15x — the dominant risk. Falsified by: any reversion toward the 11–12x own-history norm (≈20–25% of the equity value).
5. FCF conversion improves organically (not via payables float) — falsified by: CFO growth that lags AOI once the AP tailwind fades.

Factor-positioning read (evidence for consensus being offsides). The tape is a low-vol/dividend quality-momentum winner, not a falling knife: m6 return +120% annualized (~+22% raw), y1 +33.5%, max drawdown only −7.6%, beta ~0.83. The residual Value (+0.21) and DividendYield (+0.24) factor loadings are a fading artifact of the still-modest absolute multiple, not a sign of cheapness — the re-rating has largely converted a value/income name into a momentum name. Consensus is offsides to the upside: the crowded, positively-skewed positioning means the pain trade is a momentum unwind on the first disappointment, not further upside surprise.

Verdict: Consensus is pricing a repaired cyclical as a permanent compounder. The variant view is that the repair is real but finite, the moat is absent, and the multiple has run ahead of the fundamentals — leaving unfavorable asymmetry at $57.

12. Fact vs. Interpretation

| # | Statement | Type | Basis | | : | :------------------------------------------------------------------------------------------------- | :------------- | :-------------------------------------------------------------- | | 1 | FY25 revenue $18.51B; gross 8.36%, operating 4.28%, EBITDA 6.85%, net 1.76% | Fact | FY25 10-K / Aggregated fundamental data (reconciled to filings) | | 2 | ROIC ~6.8% (FY25), at/below WACC; ROE (~93%) is a garbage artifact of ~nil equity | Fact + Interp | Aggregated fundamental data (reconciled to filings); negative tangible book | | 3 | Compass ~7.2% op margin vs Aramark 4.28% → scale rewards only #1; Aramark has no durable moat | Interpretation | Compass FY25 results; margin comparison | | 4 | Client retention 96.3% (FY25 record); >98% (Q2 FY26); net new 5.6% | Fact | FY25 10-K; Q2 FY26 call | | 5 | Renewals reprice “less favorable/less profitable”; “we don’t price for profit” → no pricing power | Fact (quote) | 10-K risk factors; earnings call | | 6 | Net leverage ~3.25x covenant / ~3.76x clean (FY25), from ~9.3x (FY21); target <3x FY26 | Fact | FY25 10-K / credit ratios | | 7 | GAAP→AOI wedge ~$189M; “Covenant EBITDA” $1,465M vs reported $1,268M | Fact | FY25 10-K reconciliations | | 8 | FCF flattered ~$196M by payables float; “clean” FCF ~$236M vs ~$432–454M headline | Fact + Interp | FY25 cash-flow statement | | 9 | Trades at richest-ever multiple (~15x EV/EBITDA; 98th-pctile own history); discount to Compass closed | Fact + Interp | Own-history valuation percentiles (public price + fundamentals); ROIC multiples | | 10| Nexus could be largest client / >$100M/site, above-corporate margin | Interpretation | Management commentary (Q2 FY26); no booked revenue | | 11| Insiders: 1 open-market buy (CEO ~$250K, Aug 2025), 3 sells (one officer) over 24 months | Fact | Form 4 corpus | | 12| Deleveraging is a one-time repair now largely in the price | Interpretation | Valuation embedded-expectations analysis |

13. Open Questions

1. Nexus economics — actual contract count, revenue run-rate, margin, capital commitment, and duration. Entirely unverified; validate at FY27 disclosure. The single largest swing factor.
2. Self-op vs. outsourced TAM penetration — is the runway ~50% (operator framing) or ~78% already outsourced (one market report)? Materially changes the secular growth ceiling.
3. Sports/Leisure/Corrections margin compression — cyclical (event/catering mix) or structural (competitive/contract terms)?
4. Start-up drag from the largest-ever FSS US contract on FY26 segment margins — magnitude and duration.
5. Refinancing glidepath — how much of the deleveraging EPS benefit is offset by the 2028–2030 rate step-up?
6. FY26 adjusted-EPS growth composition — how much is 53rd-week lap, interest tailwind, and buyback vs. genuine operating gains? Validate the run-rate.
7. Buyback capacity — exact remaining authorization and management’s pace intent at record multiples.

14. What Must Be True

For the bull case to be right (the stock compounds from $57):

• Net new business stays ~4–5%+ and retention holds >96% for multiple years (durable share gain).
• Adjusted margin expands a durable 30–40bp/year (not the ~25bp of FY25) despite no pricing power and new-business drag.
• Nexus converts into real, disclosed, above-corporate-margin revenue at scale — a genuine mix-shift, not a headline.
• The multiple holds ~15x (or higher) — i.e., the market permanently re-rates a ~7%-ROIC business.
• Falsification test: two consecutive quarters of net-new deceleration below ~4%, or flat/compressing adjusted operating margin, or FY27 disclosure that Nexus is immaterial — any one breaks the compounder narrative and the multiple.

For the bear case to be right (the stock de-rates toward $40–47):

• Growth normalizes to ~4–5%, margin gains stall, and Nexus fails to become material.
• The multiple reverts toward its 11–12x EV/EBITDA / 16–18x adjusted-P/E own-history norm.
• Payables-float tailwind fades and FCF conversion disappoints; refinancing at higher rates eats the deleveraging EPS benefit.
• Falsification test: if Aramark sustains >6% organic growth with durable margin expansion and books material Nexus revenue at above-corporate margin, the bear (multiple-reversion) thesis is wrong and the premium is earned. Absent that trifecta, the record multiple is the vulnerability.

Synthesis: The bull needs a trifecta (durable growth + durable margin + Nexus) and a permanently-elevated multiple, all on a business that earns its cost of capital. The bear needs only mean-reversion of the multiple on any single stumble. That asymmetry — not a view that the business is bad — is the crux at $57.

15. Source Appendix

See Appendix B below for the full, dated source list. Primary sources: Aramark FY2021–FY2025 10-K filings (SEC EDGAR, CIK 0001584509); FY2025 10-K filed 2025-11-25 (period ended 2025-10-03); Q1/Q2 FY26 earnings 8-Ks and call transcripts; DEF 14A proxies (2024, 2025); Form 4 insider filings (trailing 24 months). Quantitative cross-checks used aggregated fundamental data reconciled to the filings, public price history, and a public factor/risk model. Peer data: Compass Group and Sodexo public results. All non-obvious facts are dated and attributed in the appendix.

APPENDIX A — Standard Diligence Questionnaire

Aramark (NYSE: ARMK) — as of 2026-07-17. Supplemental to the article; answers use Fact/Interpretation/Assumption labels where material.

General

What thoughtful questions have other investors asked about this company?

• Is Aramark a structurally-improved compounder or a repaired cyclical near its multiple ceiling? (Interpretation: the latter — the improvement is a finite balance-sheet repair, not a moat change.)
• Can margins durably expand without pricing power? (Open question — FY25 delivered only ~25bp vs a 30–40bp target.)
• What is “Aramark Nexus” really worth, and why is it not in guidance? (Open question — no booked revenue, NDA’d terms.)
• How much of FY26 EPS growth is “real” vs. 53rd-week lap, interest tailwind, and buyback?
• Why does it now trade at parity to higher-quality Compass Group?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: a cyclical/operational high — post-COVID volume recovery is complete, margins are at a multi-year high (still low in absolute terms), and FY26 adjusted-EPS growth (+20–25%) is flattered by a 53rd-week lap, a falling-interest tailwind, and buybacks. Not a trough.

Driven by external environment or internal actions? Both: external (volume normalization, inflation pass-through) drove 2021–2024; internal (net-new wins, deleveraging, cost programs) drives the incremental margin. Pricing is externally set at rebid.

How stable are revenues? Fact: highly stable in volume — 96.3% retention (FY25 record), multi-year contracts, ~two-thirds P&L / one-third management-fee. Stable volume, weak price.

Outlook for products/services / market size? Fact/Interpretation: a ~$320B global market, fragmented (#1 <15% share), growing above GDP via first-time outsourcing (~50% of TAM still self-operated per operator framing). Durable volume tailwind; commoditized economics.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Roughly stable — a three-national oligopoly at the top with a long regional tail and permanent self-op substitute. Competition is on price at rebid.

How profitable is the business (ROIC, ROE)? Fact: ROIC ~6.8% (FY25) — at/below WACC. ROE (~93%) is meaningless (near-nil / negative tangible equity). Use ROIC/AOI. This is the central quality tell: no value creation above cost of capital.

How profitable is the industry / barriers to entry? Low-margin (Compass ~7.2% op, Sodexo ~4.7%, Aramark ~4.3%). Barriers to entry at the national tier are real (scale procurement, reference base, capital) but do not confer excess returns on #3.

Can the business be easily understood? Yes — outsourced institutional dining and facilities under multi-year contracts.

Undermined by foreign low-cost labor? No — service is delivered on-site, locally. The labor cost is the risk, not offshoring.

Do brands matter? Minimal — the institution (not the diner) chooses on price/service; no consumer brand premium.

Nature of competition? Price and service quality at competitive rebid; incumbency/switching friction provides retention but not pricing power.

Customers’ switching costs? Real but modest — embedded systems, on-site capital, transition risk create friction (hence 96%+ retention), but renewals reprice down and exit capital is client-reimbursed. Friction, not a moat.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The contract book / retention franchise is an unrecognized intangible; conversely, $6.9B of recognized goodwill+intangibles earns only ~WACC.

Off-balance-sheet liabilities? Operating leases (capitalized), multiemployer pension exposure (union labor), and client-contract capital commitments. Nothing unusual flagged.

How conservative is the accounting? Interpretation: management-favorable in presentation — AOI adds back recurring acquisition amortization; “Covenant Adjusted EBITDA” ($1,465M) is more aggressive than reported EBITDA ($1,268M) and is the number quoted for leverage; headline FCF is flattered by recurring payables float (~$196M FY25). All GAAP-compliant but requires normalization.

How CapEx-hungry? Fact: capex ~2.6% of revenue (~$489M FY25), much of it client-contract investment that is reimbursed on termination. Capital-light relative to the asset-heavy comparison, but working-capital and client-capital intensive.

Capital Allocation & Management

How much FCF, and how is it used? Fact: headline FCF ~$432–454M FY25 (clean ~$236M ex-float). Used primarily for deleveraging (9.3x→~3.25x), plus a small dividend (~$111M) and resumed buyback (~$140M FY25).

Significant acquisitions recently? Only tuck-ins ($54–289M/year) — disciplined in size/price but cumulatively built a ~WACC-earning goodwill pile. No transformative M&A.

Buying back shares? Yes, resumed FY25 ($500M authorization; ~$140M FY25, ~$194M YTD FY26) — first meaningful buyback since pre-COVID.

Issuing large amounts of stock to insiders? No — SBC is clean (~0.3% of revenue, ~14% of FCF); share count roughly flat.

Compensation policy / incentive alignment? Above-average design: AIP on AOI/FCF/Net New Sales/ESG; PSUs on ROIC/FCF/Adjusted Revenue/Net New Sales/Relative TSR. Caveat: FY25 AIP paid 133% despite an AOI miss (FCF/ESG carried it). CEO pay ~$14.5M SCT — reasonable for scale.

Motivations of management? Interpretation: current regime (Zillmer/Tarangelo) is credibly focused on deleveraging, returns, and disciplined growth — a repair-and-normalize mandate, executed well. Insider buying is minimal (one CEO purchase at ~$39; nothing near $57).

Valuation & Market Data

ADR / MLP / K-1? No — ordinary US common stock (Delaware C-corp), NYSE-listed.

Dividend policy? Small and growing — ~$0.42/share, ~0.8% yield, ~34% payout; raised 14% in FY25. An income residual, not a thesis.

How profitable is the business? Low margin (net ~1.8%), ~WACC ROIC — see above.

Is net income diverging from cash from operations? CFO ($921M) exceeds net income ($326M) largely on D&A ($476M) and payables float ($196M). Cash generation is real but the quality of the CFO/NI gap includes finite working-capital tailwinds.

Risks & Downside

What would cause the stock to decline? A multiple de-rate from the 98th-percentile own-history level on any net-new/margin miss; Nexus disappointing; a discretionary-demand downturn; refinancing drag. (See risk matrix.)

Risk of catastrophic loss? Low — diversified, cash-generative, going concern; leverage (~3.5–3.8x clean, negative tangible equity) is the only structural fragility and is being reduced.

Chance of total loss? Very low, absent a severe demand shock colliding with the debt load.

Recent News & Events

Has the business environment changed recently? Yes, favorably on operations (record retention/net-new, deleveraging <3.25x, buyback resumption) and speculatively via Nexus (AI-data-center services, unveiled Q2 FY26). No adverse regulatory/accounting change identified. News flow is light and skews positive (sell-side PT raises to $60–70.5).

Significant acquisitions? Only tuck-ins.

Change in accounting policies? None material identified; note the reliance on adjusted/covenant metrics.

Recent changes — markets, facilities, management? Vestis spin (2023); largest-ever FSS US contract ramping early 2026; Nexus platform launch (2026); CEO/CFO stable.

APPENDIX B — Source Appendix

Aramark (NYSE: ARMK) — research as of 2026-07-17. Primary sources first. All URLs accessed July 2026. Company CIK 0001584509.

Primary — SEC filings (EDGAR)

Primary — Earnings call transcripts

Quantitative cross-checks (third-party; reconciled to filings)

Peer / industry

Note on adjusted metrics: Aramark reports Adjusted Operating Income (AOI) and Covenant Adjusted EBITDA that differ materially from GAAP operating income and reported EBITDA. This analysis quotes GAAP figures as primary and flags the reconciliations. All non-GAAP figures are as disclosed by the company.