CAVA Group, Inc. (NYSE: CAVA) — The Real Deal, Priced for Flawless Execution

An independent equity research note Target: CAVA Group, Inc. (NYSE: CAVA) | CIK: 0001639438 | Sector: Consumer Discretionary — Restaurants (Fast-Casual) Report date: 2026-06-27 | Price (2026-06-26): ~$83.40 | Market cap: ~$9.7B | EV: ~$9.3B (net cash) Fiscal year: 52/53-week, ending late December | Currency: USD

⚡ Claude’s Take

This block is the author’s own subjective opinion and general information only — it is not investment advice and is not a recommendation to buy or sell any security. The analysis that follows takes no position and carries no price target.

Verdict: HOLD / AVOID at ~$83 / accumulate-on-weakness — a genuinely excellent business at a price that already underwrites near-flawless execution. Estimated fair-value zone ~$55–75 (roughly 4.5–5.5x forward sales / ~35–42x forward EBITDA on a still-20%-unit-growth, net-cash compounder); I would accumulate meaningfully in the high-$40s to low-$60s — exactly where the stock traded as recently as November 2025 and February 2026 — and would not pay up here. Not a short: the growth is real, accelerating, and self-funded. Conviction: medium.

CAVA is the best fast-casual growth story in the U.S. public market not named Chipotle, and on the business, I am unreservedly positive. It is the category-defining Mediterranean concept with the second-best unit economics in all of restaurants — ~$2.9M average unit volumes, ~25% restaurant-level margins, ~50% first-year cash-on-cash returns, new boxes opening above the fleet average — a fortress net-cash balance sheet, no franchisees skimming the unit economics, and a runway from ~459 restaurants toward a stated 1,000+ by 2032. Crucially, and unlike Chipotle, its growth is not in an air-pocket: after a 2025 same-restaurant-sales (SRS) deceleration from +10.8% to +0.5% that crushed the stock, Q1-FY2026 (reported May 19, 2026) reaccelerated to +9.7% SRS on +6.8% positive traffic, and management raised full-year guidance. The single cleanest piece of evidence that the moat is real and operator-specific is the contrast with Sweetgreen — same “better-for-you fast-casual” playbook, collapsing comps (−8% in FY2025), 15% restaurant margins, cash-burning, and cutting openings to ~15. CAVA is doing everything right.

The problem is entirely the price. Even after halving from its ~$172 November-2024 peak, CAVA trades at ~7.2x trailing sales, ~55x trailing / ~50x forward EBITDA, and ~120–140x forward earnings — the richest multiple in restaurants by a wide margin and a price that already discounts the bull case (1,000+ units, sustained high-single-digit comps, margins holding) with very little margin of safety. The 2025 round-trip is the cautionary tale: this is a high-beta (~1.78), 55%-idiosyncratic-vol name whose multiple is the dominant driver of the stock — a single soft comp print took it down 75%. And the framing matters: FactorsToday shows momentum has already turned back up (six-month return ~+82% annualized, Sharpe 1.25, UBS upgrade to Buy at a $90 target on June 10), so this is not a falling knife I’m catching — it’s a re-rated growth darling that has rallied ~92% off its low and is once again priced for perfection. That asymmetry — pay near-full for the bull, eat −35–45% on any stumble — is why I want to own it, but not here. The honest summary: this is the CMG setup inverted. Chipotle is a great business finally on sale; CAVA is an even-faster-growing business that is still expensive. Wait for the air-pocket — they come — and back up the truck. Flips decisively bullish (at a lower price): two-plus more quarters of high-single-digit traffic-led comps plus visible margin expansion toward 26%+, bought in the $50s–low-$60s. Flips bearish: SRS rolls back toward flat while new-unit AUVs/returns fade — saturation, not a soft patch — at which point the multiple has ~40–50% of air beneath it. Tag: the real deal at an unreal price.

📈 Stock Price Action — Five-Year Event Map

Factual price history (CAVA has traded only since its June 2023 IPO, so this is a ~3-year map, not five). Price moves are FACT; attributed causes are INTERPRETATION. No price target, no recommendation.

The arc in plain numbers. CAVA priced its IPO at $22 on June 14, 2023, opened at ~$42 and closed its first day at $43.78 — an instant ~100% pop. It sagged to an all-time low of $29.05 by October 2023, then ran nearly six-fold to an all-time high of $172.43 on November 13, 2024, before a year-long growth de-rating took it down ~75% to $43.41 by November 20, 2025. A 2026 reacceleration has since lifted it ~92% back to $83.40 (close, 2026-06-26). The stock sits ~51.6% below its all-time high, inside a 52-week range of roughly $43.41–$98.79, and is one of the higher-beta (~1.78) names in consumer discretionary.

Cycle narrative. (1–2) The June-2023 IPO doubled on day one into a scarcity bid for a profitable-trajectory fast-casual debut, then gave most of it back as euphoria faded and early holders looked to distribute — the stock bottomed at $29.05 in October 2023. (3) 2024 was the melt-up: a genuine profitability inflection (restaurant margins to 25%, first full-year GAAP profit), a blistering +13.4% full-year SRS, addition to major indices, and the “next Chipotle” narrative drove a near-six-fold run to $172.43 — even as private-equity backers (Artal, Act III/Shaich) distributed ~$2.6B of stock into the strength. (4) Through 2025 the SRS line decelerated every quarter (+10.8% → +0.5%) as the younger, lower-income consumer pulled back; for a stock priced at ~20x sales, a decelerating comp is fatal, and the multiple unwound ~75% to $43.41 by November — essentially back to the IPO-pop level. (5) The Q4-2025 (Feb 24, 2026) and especially Q1-2026 (May 19, 2026) prints reaccelerated SRS to +9.7% on +6.8% positive traffic, management raised full-year guidance, and a UBS upgrade to Buy ($90 target, June 10) confirmed the re-rate; the stock has rallied ~92% off the low to $83.40. Each leg is traceable to a comp print and the multiple the market was willing to pay on it — the business has compounded throughout; the stock has been a referendum on the SRS line and the multiple.

1. Executive Summary

CAVA Group is the category-defining, company-owned, fast-casual Mediterranean restaurant chain — explicitly modeled on the Chipotle playbook and, on unit economics, the closest thing to Chipotle that exists. The investment question is not whether the business is good (it is excellent) but whether a price that already capitalizes near-flawless multi-year execution offers any margin of safety. We take no position and set no target; we lay out the embedded expectations, the scenario asymmetry, and the falsification tests.

The business. CAVA operates 459 company-owned restaurants (Q1-FY2026) across 28+ states and Washington, D.C., selling customizable Mediterranean bowls and pitas assembled to order — a deliberately narrow, throughput-optimized menu built on a “build-your-own” model (the 10-K touts “38 ingredients with over 17.4 billion combinations”). It owns essentially all its restaurants (no domestic franchising), so it captures the full retail sale and all of the unit-level economics. FY2025 revenue was $1,179.7M (+22.4%), restaurant-level profit margin was 24.4%, consolidated operating margin was 6.7%, and the company generated ~$153M of adjusted EBITDA and ~$26M of free cash flow — thin only because nearly all operating cash is reinvested into ~70–75 new restaurants a year. The balance sheet is a fortress: ~$403M cash and short-term investments and no funded debt (the ~$498M of balance-sheet “debt” is entirely capitalized operating leases).

The unit economics — the heart of the thesis. ~$2.9M average unit volume, ~25% restaurant-level margin, ~$1.2M new-unit build cost, ~50% first-year cash-on-cash returns, new restaurants opening at >100% productivity (above the fleet AUV), and ~38% digital mix. These are second only to Chipotle in all of restaurants and are the moat made visible: a no-moat entrant cannot earn ~50% cash-on-cash on its 459th box. The single most powerful piece of corroborating evidence is the contrast with Sweetgreen (SG) — the same better-for-you fast-casual idea, but with collapsing comps (FY2025 SRS −8%), 15.2% restaurant margins, negative adjusted EBITDA, and openings cut to ~15. CAVA’s advantage is operator-specific and real, not a category tailwind anyone can ride.

Growth — intact, not air-pocketed. Revenue compounded from ~$500M (2021) to $1,179.7M (2025). The 2025 story was an SRS deceleration — from +10.8% (Q1) to +0.5% (Q4), full-year +4.0% — that, against a ~20x-sales multiple, halved the stock. But Q1-FY2026 reaccelerated to +9.7% SRS on +6.8% positive traffic (the high-quality, traffic-led kind), unit count grew +20.2% YoY, and management raised FY2026 guidance to SRS +4.5%–6.5%, 75–77 net new restaurants, restaurant margin 23.7%–24.3%, and adjusted EBITDA $181M–$191M. The long-term target is 1,000+ U.S. restaurants by 2032 — credible but demanding (it requires accelerating to ~80–90 net new/year).

The catch — valuation and quality-of-earnings. At ~$83 the stock trades at ~7.2x trailing sales, ~55x trailing / ~50x forward EBITDA, and ~120–140x forward earnings — the richest multiple in restaurants. Two quality-of-earnings notes discipline the headline numbers: (1) FY2024 GAAP net income of $130.3M was inflated by an ~$83.7M deferred-tax valuation-allowance release — adjusted net income was only ~$50.2M — so the optical FY2024→FY2025 “earnings decline” to $63.7M is largely a tax artifact; operating income actually grew $61.0M→$79.3M. (2) ROIC is only ~6%, below the cost of capital — the company-owned growth model consumes capital today and earns its return through the unit base over time, but the return is real only if the 1,000-unit plan executes. Capital allocation is 100% reinvestment (no dividend, no buyback), and the dominant capital-markets fact is that founders and PE backers sold ~$3.4B of stock clustered at the 2024 peak, with near-zero conviction insider buying since.

The embedded expectation. At ~$83, the market is underwriting flawless delivery of the 1,000-unit plan plus sustained high-single-digit comps and stable-to-rising margins, with essentially no allowance for another air-pocket. The business may well deliver — but the price pays for the bull case and leaves ~35–45% of downside in any comp stumble, as 2025 demonstrated. This is a great business; the question the price forces is whether now is the time to own it.

2. Business Overview

What CAVA is. CAVA Group operates a company-owned (non-franchised) fast-casual Mediterranean restaurant chain. The format is the Chipotle assembly line transposed onto Mediterranean cuisine: guests build bowls or pitas from a base, proteins (grilled chicken, braised lamb, falafel, harissa honey chicken), grains, a signature set of dips and spreads (hummus, tzatziki, harissa, crazy feta — the brand’s culinary signature), toppings, and dressings. The menu is deliberately narrow and customizable — the source of throughput, supply-chain simplicity, and unit-economic consistency, exactly as at Chipotle. The first CAVA restaurant opened in 2011 in Bethesda, Maryland; as of the FY2025 10-K (fiscal year ended December 28, 2025) the company operated 439 CAVA restaurants in 28 states and Washington, D.C., reaching 459 by the close of Q1-FY2026 (April 2026).

The company-operated model — the key structural fact. Like Chipotle and unlike nearly every restaurant at scale (McDonald’s ~95% franchised, Yum ~98% franchised), CAVA owns and operates essentially all of its restaurants. The consequences run through everything:

• CAVA books the full retail sale as revenue ($1.18B), not a royalty slice, so its consolidated operating margin (~7%) is structurally far below an asset-light franchisor’s — it carries all store labor, occupancy, and food cost.
• It captures 100% of the unit-level economics and the upside — there is no franchisee taking the restaurant-level margin, which is why CAVA’s growth translates so directly into restaurant-level profit dollars (~$285M in FY2025).
• It controls brand, operation, food quality, and guest experience completely — no franchisor/franchisee value war.
• It bears full operating leverage in both directions — when traffic grows, incremental margins are strong; when traffic stalls (FY2025), deleverage bites (restaurant margin slipped 60bps to 24.4%).

Revenue model and segments. Revenue is ~99% restaurant food-and-beverage sales. FY2025 consolidated revenue was $1,179.7M, of which the CAVA segment was ~$1,169M; the small residual is the CPG “dips & spreads” business (CAVA-branded hummus, tzatziki, dressings sold in grocery — notably across Whole Foods, in roughly 650 stores) plus the now-complete wind-down of the legacy Zoe’s Kitchen segment. The CPG line is a brand halo and supply-chain asset, not a material P&L contributor (<1% of revenue); CAVA centrally produces dip/spread bases for its own restaurants and is building a ~$30M Virginia production facility (announced March 2026) to scale that capability. CAVA is, for practical purposes, a pure-play single-concept restaurant operator — the story is clean: revenue = units × AUV × comp, and value creation = unit growth at high incremental returns, plus comp, minus margin and capital intensity.

History and key actors. CAVA was founded in 2006 by three childhood friends — Ike Grigoropoulos, Ted Xenohristos, and Dimitri Moshovitis — as a full-service Mediterranean restaurant, pivoting to the fast-casual format in 2011. The pivotal corporate event was the November 2018 acquisition of Zoe’s Kitchen for ~$300M ($12.75/share, all-cash), financed in significant part by Ron Shaich’s Act III Holdings — Shaich being the founder/longtime CEO of Panera Bread, now CAVA’s Chairman of the Board. The Zoe’s deal gave CAVA an “extensive portfolio of real estate” (the 10-K’s phrase) that it used to expand rapidly by converting Zoe’s locations into CAVA boxes; by March 2023 all Zoe’s units had been closed or converted, and the combined entity was public-company ready. CAVA IPO’d in June 2023 at $22/share. Today CEO Brett Schulman (in the seat since 2010) and co-founder/Chief Concept Officer Ted Xenohristos run the operation under Shaich’s chairmanship. Digital infrastructure (the “connected kitchen”), a loyalty program launched in 2023, and a pickup/drive-thru format (“CAVA pickup”) round out the operating system.

Verdict. CAVA is a focused, premium, company-operated fast-casual operator with best-in-class-but-one unit economics and a clean, single-concept growth model. The company-operated structure means it captures all unit-level upside and bears all the operating leverage — a higher-quality economic model than a franchisor while growing into a long runway, but a more volatile one when traffic turns. The business quality is not in question; the model is excellent.

3. Industry Dynamics

A large, mature, brutally competitive industry — with one genuinely advantaged neighborhood. U.S. restaurants is a ~$1.1T, low-growth, structurally unattractive industry: low barriers to entry (anyone can open a restaurant), high barriers to scale (building thousands of consistent, profitable units is extraordinarily hard), razor-thin median margins, high failure rates, commodity and labor cost exposure, no pricing power for the median operator, and chronic oversupply. Marathon-style capital-cycle logic applies with force: high returns at the winners attract a flood of capital and copycats, and most of that capital is destroyed.

Fast-casual is the structurally best format, and “better-for-you” is its share-gaining frontier. Within restaurants, fast-casual (fresh-ish food, QSR speed, $11–15 check) has taken share from both casual dining (too slow/expensive) and legacy QSR (lower quality) for two decades. CAVA rides the specific thesis that Mediterranean is the next mainstream better-for-you cuisine — the same category-creation arc Chipotle ran with Mexican: an ethnic, health-coded cuisine with broad demographic appeal (CAVA skews Millennial/Gen Z) that can scale nationally. The category’s economics at the winners are genuinely good — high AUVs, strong unit returns, real brand pull — which is precisely why it is now attracting capital.

The Marathon capital-cycle read is, unusually, favorable for the incumbent. Capital is chasing better-for-you fast-casual, but the supply response is concentrated and largely failing. Sweetgreen — the most-capitalized direct analog — is shrinking FY2026 openings to ~15 and burning cash; a long tail of regional “better-bowl” concepts lacks the unit economics to scale. CAVA is currently the rational actor taking share while competitors retrench — the favorable phase of the capital cycle for the winner. The risk is the later phase: CAVA’s own success invites copycats, and the adjacency to Chipotle and QSR value menus means the Mediterranean lunch occasion is contestable.

Whitespace. Management states it sees room for more than 1,000 CAVA restaurants in the U.S. by 2032 (from 459 today), having penetrated only 28 states. From 459 to 1,000+ over ~7 years implies ~80–90 net new units a year at a ~12–15% unit CAGR — an acceleration from the 72 opened in FY2025. It is plausible given (a) the limited current footprint, (b) new units opening above the fleet AUV, and © the Zoe’s/Chipotle precedent that an ethnic better-for-you concept can travel nationally — but it assumes new markets (Midwest, Southeast) replicate coastal economics, which is unproven.

Regulation and input costs. The sector faces rising minimum/living-wage mandates (California’s $20 fast-food wage law is a direct cost input for any operator there), food-safety regulation (the fresh-food model carries idiosyncratic outbreak tail risk — Chipotle’s 2015–18 E. coli crises are the cautionary precedent), immigration-enforcement risk to labor supply, and commodity volatility (chicken, beef/lamb, olive oil, dairy, produce). CAVA’s fresh, less-processed supply chain is a brand asset but a cost and food-safety liability relative to frozen-and-fried QSR.

Verdict: a structurally poor industry in which CAVA occupies the best-positioned format and is, so far, a genuine winner — and where, unusually, the capital cycle is currently culling its closest competitors rather than the reverse. Bad industry, excellent neighborhood, advantaged operator — better than commodity QSR, but a category whose very attractiveness guarantees future competition.

4. Competitive Position

The moat is real but narrow, early-stage, and unproven at scale. In Greenwald’s taxonomy, CAVA’s advantage is a combination of (a) a demand-side brand intangible — CAVA is the only national-scale Mediterranean fast-casual brand and effectively owns the category in the consumer’s mind, the way Chipotle owns fast-casual Mexican; (b) emerging economies of scale in supply chain — central production of dips, spreads, and dressing bases (now a dedicated facility) lowers unit COGS and compounds with restaurant count; and © a replicable, high-return operating system — the assembly-line throughput model, the people pipeline, and the consistency that lets new boxes open above the fleet AUV. It is the same moat archetype as Chipotle, but earlier, narrower, and far less proven — 459 versus ~3,800 boxes, ~$1.2B versus ~$12B of revenue, scale economies still being built rather than dominant. It is not a switching-cost or network-effect moat, and there is no real-estate-and-royalty toll.

Evidence the moat is genuine:

• Unit-economic premium. ~$2.9M AUV and ~25% restaurant margin are roughly double a typical fast-casual unit’s; ~50% first-year cash-on-cash on a ~$1.2M build is best-in-class-but-one. The honest test — would the financials deteriorate without the advantage? — passes: a no-moat operator cannot earn those returns.
• New-unit productivity. New restaurants opening at >100% productivity (above the existing fleet AUV, with the 2025 class trending above $3.0M) mean the box engine is not diluting as it scales — the single most important quality signal for a unit-growth story.
• The Sweetgreen contrast — the strongest single datum. Same category, same “better-for-you bowl” playbook, comparable vintage — yet Sweetgreen’s FY2025 SRS was −8% (deteriorating every quarter to −11.5% in Q4), restaurant margin fell to 15.2%, adjusted EBITDA went negative, net loss widened to ~$134M, and it slashed openings to ~15. CAVA’s +4% SRS / 24.4% margin / profitable / accelerating-units profile against that backdrop is conclusive evidence the advantage is operator-specific and real, not a category tailwind anyone can ride.

Where the moat is limited — pressure-tested:

• No switching costs, no network effects, weak captivity. A bowl is a discretionary, trivially substitutable purchase; CAVA must re-earn every lunch. The loyalty program (launched 2023) raises engagement but is not lock-in.
• The 2025 SRS collapse is the disconfirming evidence. SRS fell from +10.8% to +0.5% in four quarters — exactly what a no-captivity model looks like under consumer stress. The moat protects premium economics, not volume, against a weak consumer.
• Replicability by design. The assembly-line format is not proprietary; the moat is execution + brand, not a structural toll. The barrier is that replicating CAVA’s execution is hard — Sweetgreen proves it — but the concept is open.

Direct comparison. Versus Chipotle, CAVA has a similar (slightly lower) unit-economic profile but a far longer runway (~12% built toward its target versus CMG’s ~55%) and a less-proven, narrower moat; the roles are inverted — CMG is the maturing incumbent defending a saturating base, CAVA the upstart taking share. Versus Sweetgreen, CAVA is simply the better operator on every metric that matters. Versus the regional Mediterranean tail (Naf Naf, Roti, The Hummus & Pita Co.) and QSR value menus, CAVA has overwhelming scale, brand, and balance-sheet advantages.

Verdict: a genuine, financially-validated moat (brand + emerging scale + a replicable high-return operating system) — durable enough to scale the concept profitably, but narrow, early, and without customer captivity, as the 2025 transaction-elasticity exposed. Durable advantage: yes, and currently widening versus a failing peer set. Impregnable: no — and priced as if it were.

5. Growth History and Forward Opportunities

A high-quality compounding record, one deceleration, and a reacceleration. CAVA’s revenue grew from ~$500M (2021) to $1,179.7M (2025), with the engine a combination of ~15–20% annual unit growth and same-restaurant sales:

Decomposition and the crux. The 2025 story is the SRS air-pocket: same-restaurant sales fell from +10.8% (Q1) to +0.5% (Q4) — a traffic-led consumer slowdown (the younger, lower-income demographic that pulled back across the better-for-you cohort and gutted Sweetgreen) — which, for a stock at ~20x sales, repriced the “perpetual double-digit comp” assumption and took the stock down ~75%. Then Q1-FY2026 reaccelerated to +9.7% SRS on +6.8% positive traffic — only ~2.9 points from price/mix, i.e., the high-quality traffic-led kind — prompting a guidance raise. The unresolved question, on which the valuation entirely hinges, is whether the FY2025 dip was cyclical (a transitory consumer soft patch, with Q1-26 the proof) or whether the concept’s comp algorithm has structurally downshifted toward mid-single-digits as the base matures and laps tougher comparisons. The evidence currently leans cyclical — but one quarter does not settle it.

Forward drivers:

• Unit growth (the dominant lever). 459 → 1,000+ by 2032, with new units opening above the fleet AUV and at >100% productivity — value creation comes overwhelmingly from box count at high incremental returns. FY2026 guidance is 75–77 net new units.
• New markets. Entry into the Midwest and Southeast extends the runway, but the economics outside the proven coastal/Sun Belt markets are not yet demonstrated.
• Menu innovation, loyalty, and digital. ~38% digital mix, a maturing loyalty program, and a doubled LTO/innovation cadence support comp without leaning on price.
• CPG and drive-thru optionality. The dips/spreads grocery business is a brand halo with modest standalone upside; “CAVA pickup”/drive-thru formats are an emerging margin and convenience lever.

Verdict: high-quality growth — unit-led, traffic-led, high-return, self-funded — with the single caveat that the SRS line proved cyclical and traffic-elastic in 2025. The growth is real and currently reaccelerating; the durability of the comp is the open question the valuation is paying full price for.

6. Financial Quality

Revenue and margins. Revenue compounded ~24% annually over five years to $1,179.7M (FY2025). The restaurant-level profit margin — the cleanest read on store economics — has run 24–25% since 2023 (24.8% → 25.0% → 24.4%), a best-in-class-but-one level that slipped 60bps in FY2025 on traffic deleverage and labor, then recovered to 25.1% in Q1-FY2026. Consolidated operating margin rose from −5.4% (2021) to −2.6% (2022) to +4.3% (2023) to +6.3% (2024) to +6.7% (2025), with Q1-FY2026 at 7.8% — a genuine, durable operating-leverage trend masked at the net-income line by tax noise (below). Gross/restaurant economics improve with scale; the consolidated margin is held down by G&A and pre-opening costs that should leverage as the base grows.

Quality-of-earnings flag #1 — the FY2024 tax benefit. FY2024 GAAP net income of $130.3M is not a clean number. Pretax income that year was only $59.9M; the reported net income was inflated by an ~$83.7M release of the deferred-tax-asset valuation allowance (the income-statement tax line was a $70.4M benefit, not an expense). Normalized at a real tax rate, FY2024 adjusted net income was ~$50.2M. FY2025 GAAP net income of $63.7M (with a more normal ~10% effective rate) therefore grew versus FY2024 adjusted earnings — the optical “earnings cut” from $130M to $64M is almost entirely the absence of the prior-year one-time tax benefit. Anyone computing a year-over-year EPS decline or a trailing P/E off the $1.10 FY2024 diluted EPS is being misled; the right read is operating income up $61.0M → $79.3M (+30%) and adjusted net income up ~$50M → ~$64M.

Quality-of-earnings flag #2 — ROIC below cost of capital. Return on invested capital is only ~6% (ROIC 6.1% FY2025, 7.6% FY2023), comfortably below an ~8–9% cost of capital. This is the expected signature of an early-stage, company-owned unit-growth model: capital is deployed today into restaurants whose returns accrue over their multi-year life, and the consolidated ROIC is dragged by a large, recently-built, not-yet-fully-seasoned asset base plus ~$400M of low-returning cash. The new-unit cash-on-cash return (~50%) is what matters for value creation, and it is excellent — but the consolidated ROIC will remain sub-WACC until the unit base matures and G&A leverages. The thesis literally requires the 1,000-unit plan to convert that ~50% box-level return into a portfolio ROIC above WACC; today it has not.

Free cash flow. FY2025 operating cash flow was $184.8M against $158.7M of capex, for ~$26M of free cash flow — thin by design: CAVA is reinvesting essentially all operating cash into ~70–75 new restaurants a year. FCF will stay modest while unit growth runs at ~15–20%; the company is self-funding its growth with no need for external capital, which is the right answer for a ~50%-cash-on-cash reinvestment opportunity. (Note: reported OCF benefits from non-cash lease accounting and working-capital timing; the disciplined read is OCF-less-capex, ~$26M.)

Balance sheet — a fortress. As of Q1-FY2026 (April 2026): $403M of cash and short-term investments, zero funded debt (the ~$498M of “debt” on the balance sheet is entirely capitalized operating-lease obligations), total equity $810M, current ratio 2.65x, and a slightly negative cash-conversion cycle (restaurants collect cash before paying suppliers). CAVA carries an undrawn revolving credit facility. There is no liquidity, solvency, or refinancing risk — a meaningful differentiator versus leveraged or cash-burning peers (e.g., Sweetgreen).

Dilution and SBC. Share count rose from ~96.9M (2022) to ~116.4M (Q1-FY2026), but the overwhelming bulk of that is the IPO-mechanical conversion of ~95M preferred shares to common plus primary issuance (net IPO proceeds ~$336M); ongoing SBC dilution is modest. Stock-based compensation was $9.4M (2023) → $13.6M (2024) → $15.2M (2025), only ~1.2–1.3% of revenue — low for a recently-IPO’d growth restaurant. There is no chronic equity-funded burn.

Verdict: economics that genuinely improve with scale at the unit level (25% restaurant margins, ~50% cash-on-cash, rising operating margin, fortress net-cash balance sheet) — but a consolidated ROIC still below cost of capital and thin headline FCF, both inherent to an early-stage, self-funding unit-growth model. The quality is real; it is young, and the headline net-income and P/E figures must be normalized for the FY2024 tax benefit before any valuation conclusion.

7. Capital Allocation

The policy: 100% reinvestment into new units; no dividend, no buyback. This is the correct policy for a business compounding restaurant count at ~15–20% with ~50% first-year cash-on-cash returns — paying a dividend or repurchasing stock at ~120x earnings instead of opening ~50%-return boxes would be value-destructive. Capital allocation is therefore almost entirely a question of reinvestment quality, which the new-unit economics validate. The corollary, though, is that all shareholder value depends on that reinvestment ROIC — and, until 2026, no incentive metric held management accountable to it.

Incentive alignment — improving, but late. Through FY2025, the annual bonus was weighted 67% adjusted EBITDA / 33% revenue with no return-on-capital, restaurant-margin, or unit-economic metric, and long-term equity was time-vested RSUs and options only — no performance condition at all. A return metric appears for the first time with the 2026 grants: 50% PSUs / 50% RSUs, with the PSUs measured 50% on Adjusted ROIC and 50% on Adjusted Diluted EPS over a three-year period. This is a genuine and welcome upgrade — but it is recent, the “Adjusted ROIC” definition and its adjustments are undisclosed and bear watching, and for the first three public years the people deploying ~$150M/year of growth capex had no contractual stake in the returns on it. CEO Brett Schulman’s FY2025 total compensation was ~$3.86M and CFO Tricia Tolivar’s ~$2.30M; the say-on-pay vote passed with ~91% support (2025 meeting).

Insider behavior — the dominant capital-markets fact. Across the trailing Form 4 record, founders and private-equity backers sold ~$3.4B of stock, heavily clustered at the 2024 price peak ($124–$150): Artal/Invus ~$2.62B, Chairman Ron Shaich ~$637M, CEO Brett Schulman ~$92M, co-founder Xenohristos ~$36M. Most large-holder selling was via post-lockup secondary offerings and 10b5-1 plans in 2024; selling collapsed in 2025 as the price fell and resumed modestly in 2026 at $84–$90 as the stock recovered. Critically, there has been near-zero conviction buying — the only open-market purchases in 2025–26 are ~$0.5M of small, stock-ownership-guideline buys by two newly appointed executives plus a token 150-share CEO purchase. No founder or director stepped in to buy the 75% drawdown. This is not disqualifying — early-backer monetization after an IPO is normal, and 10b5-1 sales are diversification, not signal — but the complete absence of insider conviction at the lows, against ~$3.4B of distribution near the highs, is a fair tell about where management itself saw value.

M&A and use of proceeds. The defining transaction was the 2018 Zoe’s Kitchen acquisition (~$300M), which provided the real-estate platform that made CAVA a national chain — a strategically excellent deal whose locations have all been converted or closed. There has been no material M&A since (only a small ~$5M convertible-note investment in FY2025). IPO proceeds (~$336M net) went to general corporate purposes and reduced reliance on the credit facility. Capex has run $139M (2023) → $108M (2024) → $159M (2025), tracking the new-unit cadence at a ~$1.2M build cost.

Verdict: capital allocation is sound in substance — disciplined reinvestment into a high-return unit pipeline, a fortress balance sheet, no value-destructive buybacks at a rich multiple, and an excellent founding-era acquisition — but governance lagged the substance: ROIC entered the incentive structure only in 2026, and insiders have been heavy net sellers (~$3.4B at the highs) with no conviction buying at the lows. The reinvestment is the right call; the alignment is newly, not durably, established.

8. Changes and Headwinds — Last Two Years

Strategic and operational. The defining change of the period is the 2025 SRS deceleration and 2026 reacceleration: the comp algorithm slowed from double digits to near-flat through 2025 on a consumer pullback, then reaccelerated to +9.7% with positive traffic in Q1-FY2026, accompanied by a guidance raise. Unit growth continued throughout (72 net new in FY2025, 75–77 guided for FY2026), and new-unit productivity stayed above 100% — the growth engine never broke; the comp wobbled and recovered.

Leadership. COO Jennifer Somers departed in September 2025, and Douglas W. Thompson was appointed COO in January 2026 — a notable operational-leadership transition at a company scaling unit count rapidly (Thompson subsequently made small open-market purchases). Founders remain in place: Schulman as CEO, Xenohristos as Chief Concept Officer, Shaich as Chairman. A director retirement (Kochevar, 2026) and the compensation-program overhaul (2026 PSUs) round out the governance changes.

Financing. A routine credit-facility amendment (March 2026, JPMorgan as administrative agent) was the only material financing event — CAVA remains net-cash with an undrawn revolver. There were no restatements, impairment 8-Ks, or guidance-cut 8-Ks in the corpus.

Headwinds. (1) Consumer cyclicality — the 2025 air-pocket proved CAVA’s younger, value-conscious customer base is traffic-elastic; a macro or employment downturn would pressure comps again. (2) Cost inflation — labor (minimum-wage mandates), proteins, olive oil, and energy (management flagged a 20–40bp restaurant-margin energy headwind into FY2026 guidance on geopolitical/oil uncertainty). (3) New-market execution risk — the 1,000-unit plan requires unproven Midwest/Southeast economics. (4) Competitive encroachment — success invites copycats and Chipotle/QSR adjacency. (5) Food-safety tail — the fresh model carries idiosyncratic outbreak risk. (6) The valuation itself — a ~120–140x forward-earnings multiple is its own headwind: it leaves no room for error, as 2025 demonstrated.

Verdict: the past two years strengthened the operating thesis (the growth engine proved resilient through a comp air-pocket and reaccelerated, the balance sheet stayed pristine, and incentive alignment improved) while sharpening the central risk (a traffic-elastic comp against a perfection-priced multiple). On balance, the business is stronger; the stock is more, not less, demanding.

9. Risk Analysis (Risk Matrix)

Summary. The dominant risk is valuation — not solvency, not the business. CAVA is a financially sound, growing, net-cash company; the realistic downside is a multiple de-rating triggered by a comp stumble (as in 2025), not impairment or insolvency. The asymmetry is unfavorable at this price: limited fundamental risk, but high price risk.

10. Valuation Discussion (Embedded Expectations)

Where the multiple sits. At ~$83.40 (116.4M shares → ~$9.7B market cap; ~$403M net cash → ~$9.3B EV ex-leases), CAVA trades at:

The AZI own-history valuation percentile reads only ~44th (composite), which is misleading and must be discounted: CAVA’s trading history is barely ~2.5 years and includes the 2024 bubble (EV/sales peaked near 20x), so “mid-range of its own history” still means “extremely rich in absolute terms.” The disciplined read is the absolute multiple against the growth and returns — and on that basis CAVA is the most expensive scaled restaurant in the market.

Embedded-expectations analysis — what the price requires. A ~$9.3B EV on a business generating ~$170M of EBITDA and ~$26M of FCF is not justifiable on current cash flows; it is a call option on the 2032 unit base. Reverse-engineering the bull math: if CAVA reaches ~1,000+ restaurants by 2032 at ~$3.0M AUV, that is ~$3.0B+ of revenue; at a ~25% restaurant margin and a corporate EBITDA margin scaling toward ~18–20% (operating leverage on G&A), that is ~$550–600M of EBITDA and perhaps ~$300M+ of net income. At today’s ~$9.3B EV, the market is paying ~16x 2032 EBITDA and ~30x 2032 earnings — a reasonable multiple for a then-still-growing compounder, but one that requires (a) the full unit plan executes on schedule, (b) comps stay positive (mid-to-high single digit) throughout, © margins hold or expand, and (d) no air-pocket de-rates the stock along the way. In short, the price already capitalizes the bull case. The market is underwriting flawless execution; it is not underwriting another 2025.

What the market is pricing correctly vs. incorrectly. Correctly: that CAVA’s unit economics are best-in-class, the runway is long, the balance sheet is pristine, and the FY2025 dip was (on Q1-26 evidence) more cyclical than structural. Potentially incorrectly: that the comp algorithm can sustain high-single-digit traffic-led growth and that the multiple should stay near these levels — both of which 2025 showed are fragile assumptions.

Scenario analysis (illustrative; not price targets):

• Bear (~$45–55): the FY2025 comp deceleration recurs, new-market economics disappoint, and the multiple compresses toward ~4x sales / ~30x EBITDA — i.e., back to where the stock actually traded in November 2025 / February 2026. ~−35 to −45%.
• Base (~$60–80): CAVA executes ~15–18% unit growth and mid-single-digit comps, holds ~24–25% restaurant margins; by ~2028 revenue is ~$1.8–2.0B and EBITDA ~$320M; the market keeps a ~35–40x EBITDA multiple on a still-long runway, leaving the stock roughly flat-to-modestly-higher over time. Fair value zone today ~$55–75.
• Bull (~$120–150+): the 1,000-unit-by-2032 path holds, comps run high-single-digit, margins expand toward 27%, and the market continues to pay a premium growth multiple; the stock compounds from here.

Verdict. CAVA is a great business at a price that has already priced the bull case. The embedded expectation is flawless execution with no allowance for a comp stumble — an unfavorable asymmetry at ~$83, where the upside requires everything to go right and the downside (~−40%) requires only a single soft quarter, exactly the event that occurred in 2025. No price target; no recommendation.

11. Variant Perception

Consensus. CAVA is widely held as a premier, must-own fast-casual growth name — “the next Chipotle” — with sell-side sentiment constructive (UBS upgraded to Buy at a $90 target on June 10, 2026, and the tape has rallied ~92% off the November-2025 low). Consensus accepts that the stock is expensive but argues the runway, unit economics, and reaccelerating comps justify a premium that “you always have to pay” for the best growth stories.

The strongest bull case. This is the rare genuine compounder: a category-defining brand with ~50%-cash-on-cash unit economics, a fortress net-cash balance sheet, a runway to more than double the store base, new units opening above the fleet AUV, reaccelerating traffic-led comps (+6.8% traffic in Q1-26), and a failing peer set (Sweetgreen) that proves the moat is real. Buy the best operator in a structurally attractive format with a decade of unit growth ahead, pay up, and let it compound — the multiple looks expensive every year and the stock wins anyway.

The strongest bear case. The price is the thesis-killer. At ~7.2x sales / ~50x forward EBITDA / ~120–140x forward earnings, CAVA capitalizes flawless execution with no margin of safety; the consolidated ROIC is ~6% (below WACC) and only becomes attractive if the 2032 plan executes perfectly; the comp line is demonstrably traffic-elastic and cyclical (the 2025 air-pocket halved the stock and could recur); insiders sold ~$3.4B near the highs with zero conviction buying at the lows; and the concept is replicable, with the capital cycle guaranteed to draw competition as CAVA’s success becomes obvious. A single soft quarter de-rates the multiple 40%+, as 2025 proved.

The 3–5 assumptions that matter most:

1. Comp durability — can CAVA sustain mid-to-high-single-digit traffic-led SRS, or was 2025 the first sign of structural deceleration? (The whole multiple rests here.)
2. Unit-growth runway and new-market economics — does 459 → 1,000+ execute at preserved ~50% cash-on-cash, including in unproven geographies?
3. Margin trajectory — do restaurant margins hold/expand (24–25%+) and does corporate EBITDA margin leverage with scale, lifting consolidated ROIC above WACC?
4. Multiple persistence — will the market keep paying ~7x sales / ~50x EBITDA, or does any stumble reset it toward the 4–5x sales it traded at seven months ago?
5. Competitive intensity — does CAVA’s lead widen (Sweetgreen-style failures) or compress (new well-capitalized entrants / CMG-QSR adjacency)?

Falsification tests. Bull falsified if: SRS rolls back toward flat for two-plus quarters while new-unit AUVs/returns visibly fade — saturation, not a soft patch. Bear falsified if: CAVA delivers several more quarters of high-single-digit traffic-led comps with restaurant margins expanding toward 26%+ and consolidated ROIC climbing toward double digits — proving the model scales into its multiple.

The factor-positioning read. FactorsToday shows CAVA as a high-beta (~1.78), high-idiosyncratic-vol (~55% annualized) name with no clean style identity — no Momentum, Growth, or Quality factor loading survives the model’s regularization, Value is slightly negative, and the R² is only ~0.27, meaning the stock trades overwhelmingly on its own idiosyncratic story rather than as a factor vehicle. The risk-adjusted track record confirms the round-trip: a ~−71% three-year max drawdown but a ~+82% annualized six-month return (Sharpe ~1.25) — i.e., momentum has turned back up, the stock is ~92% off its low and ~45% off its peak, and this is a re-rating recovery, not a falling knife. That is precisely why the variant view is not “catch the falling knife” but “wait for the next air-pocket”: the tape currently agrees with the bulls, the consensus is constructive, and the asymmetry at ~$83 is unattractive — the contrarian, evidence-based stance is patience, not capitulation to the rally.

12. Fact vs. Interpretation Table

13. Open Questions

1. Is the FY2025 SRS deceleration cyclical or structural? Q1-26 (+9.7%) argues cyclical, but two-to-three more quarters of high-single-digit traffic-led comps are needed to settle it. The single most important open question.
2. Do new-market (Midwest/Southeast) unit economics match the proven coastal/Sun Belt boxes at ~$3M AUV and ~50% cash-on-cash, or do they dilute as the easy markets fill in?
3. When does consolidated ROIC cross WACC? The unit-level ~50% return is excellent, but the portfolio ROIC is ~6%; the path and timing to a double-digit consolidated ROIC are unmodeled by management publicly.
4. How is “Adjusted ROIC” defined in the 2026 PSU plan, and how aggressive are the adjustments? Governance just put ROIC into pay — the definition determines whether it is a real hurdle.
5. What is the durable corporate EBITDA / operating-margin ceiling as G&A leverages — 18%? 20%? — and how much of the bull case depends on margin expansion versus unit count?
6. Will the remaining PE overhang (Artal ~8%) and ongoing 10b5-1 insider sales pressure the stock, and does any insider ever buy on weakness?
7. How real is the CPG / drive-thru optionality as a future growth or margin lever, versus a brand-halo footnote?

14. What Must Be True

For the bull case (own it here and compound):

• CAVA sustains mid-to-high-single-digit, traffic-led same-restaurant sales through cycles — the 2025 air-pocket was a soft patch, not a structural downshift.
• The unit base scales from 459 toward 1,000+ by 2032 at preserved ~50% cash-on-cash returns, including in unproven new markets.
• Restaurant margins hold/expand toward 26%+ and corporate operating leverage lifts consolidated ROIC above WACC, converting the box-level return into portfolio value.
• The market continues to pay a premium multiple (~6–7x sales / ~45–50x EBITDA) as the runway shrinks.
• Falsification test: two-plus consecutive quarters of SRS rolling toward flat while new-unit AUVs/returns visibly fade — saturation, not cyclicality — falsifies the bull.

For the bear case (the price de-rates):

• The comp algorithm structurally downshifts toward low-single-digits as the base matures and laps tough comparisons, recreating a 2025-style traffic air-pocket.
• New-market economics disappoint, slowing high-return unit growth and forcing the 1,000-unit timeline out.
• The multiple compresses toward ~4–5x sales / ~30x EBITDA (where it traded in late-2025/early-2026), delivering ~−35–45% even if the business merely grows steadily.
• Falsification test: several quarters of high-single-digit traffic-led comps with margins expanding toward 26%+ and ROIC climbing toward double digits — proof the model scales into its multiple — falsifies the bear.

The two cases are not symmetric in the business (CAVA is a high-quality, growing, net-cash company in either) — they are a debate about the durability of the comp and the persistence of the multiple. At ~$83 the price sides decisively with the bull; the margin of safety sides with patience.

15. Source Appendix

See the separate Source Appendix (CAVA_source_appendix.md) and the Standard Diligence Questionnaire (CAVA_diligence_appendix.md) accompanying this memo. Primary sources: CAVA Group FY2025 Form 10-K (filed 2026-02-25, CIK 0001639438), Q1-FY2026 Form 10-Q and earnings release/transcript (2026-05-19), DEF 14A proxy (2026-04-24), the trailing Form 3/4/8-K corpus (2023–2026), ROIC.ai fundamentals/valuation data, FactorsToday factor model, public market data, and a comparative read of Chipotle’s public filings for fast-casual industry and unit-economic context.

This analysis carries no investment recommendation and no price target; valuation is discussed solely as embedded expectations and scenarios. The single, clearly-labeled exception is the opening opinion block, which is the author’s own subjective view and general information only — not investment advice.

APPENDIX A — Standard Diligence Questionnaire — CAVA Group, Inc. (NYSE: CAVA)

Supplemental to the research memo (not counted toward its length standard). Grounded in the underlying analysis; Fact / Interpretation / Assumption labeled where it matters. Report date 2026-06-27.

General

What thoughtful questions have other investors asked about this company? The central debate is whether CAVA is “the next Chipotle” (a decade-long compounder worth any premium) or a hyper-growth name whose multiple is unsustainable. The most-discussed questions: (1) Is the 2025 same-restaurant-sales (SRS) deceleration cyclical or the start of structural maturation? (2) Can the 1,000-unit-by-2032 target be hit at preserved ~50% cash-on-cash returns, including in unproven new markets? (3) When does consolidated ROIC (~6%) cross the cost of capital? (4) Is ~7x sales / ~50x EBITDA defensible? (5) Why have insiders sold ~$3.4B with no conviction buying at the lows?

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Interpretation: Neither extreme. Restaurant-level margins (24.4%) are near a normal level (off a 25.0% high); the comp line had a cyclical low in late 2025 (+0.5% Q4) and is reaccelerating (+9.7% Q1-26). Earnings are early in a unit-growth ramp, not cyclically peaked. The FY2024 GAAP net income ($130.3M) was a one-time high due to an ~$83.7M deferred-tax valuation-allowance release — not a recurring level.

Driven by external environment or internal actions? Both. Unit growth and margin structure are internally driven (best-in-class execution); the 2025 comp dip was an external consumer/traffic slowdown (the same one that gutted Sweetgreen).

How stable are revenues? Fact: Restaurant revenue is recurring in aggregate (daily discretionary purchases) but not contractually sticky — no switching costs, traffic-elastic. The 2025 SRS path (+10.8%→+0.5%) shows volume can swing materially with the consumer.

Outlook for products/services; how big will the market be? Interpretation: Mediterranean fast-casual is an early, growing category; CAVA targets 1,000+ U.S. units by 2032 (from 459) — a ~12–15% unit CAGR. The U.S. restaurant TAM is ~$1.1T; CAVA’s served market is the better-for-you fast-casual share-gainer. Growing, domestic, with international optionality untapped.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Interpretation: More — category attractiveness draws capital (Marathon cycle) — but CAVA’s closest competitor (Sweetgreen) is currently retrenching, so CAVA’s relative position is widening near-term.

How profitable is the business (ROIC, ROE)? Fact: Consolidated ROIC ~6% (below WACC); restaurant-level returns ~50% cash-on-cash (excellent). ROE is distorted by the negative accumulated deficit / recent IPO equity. The unit is highly profitable; the consolidated entity is early-stage.

How profitable is the industry — competitors, barriers? Interpretation: Industry: poor (thin median margins, high failure). Barriers to entry: low; barriers to scale: high. CAVA’s barrier is execution + brand, not a structural toll.

Can the business be easily understood? Yes — a clean single-concept company-owned restaurant: revenue = units × AUV × comp.

Undermined by foreign low-cost labor? No — a domestic, service-and-location business.

Do brands matter? Yes — CAVA’s category-defining Mediterranean brand is the core demand-side intangible.

Nature of competition / switching costs? Compete on food quality, value, convenience, brand. Fact: Customer switching costs are effectively zero; CAVA must re-earn every visit.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? The brand intangible and the new-unit-development capability are not capitalized. The ~$285M of annual restaurant-level profit reflects an under-recognized internally-built store portfolio.

Off-balance-sheet liabilities? None material; operating leases ARE capitalized (~$498M, the entirety of “debt”). Standard restaurant lease obligations.

How conservative is the accounting? Interpretation: Generally clean, with one watch item — the FY2024 deferred-tax valuation-allowance release inflated GAAP net income; normalize it. SBC is modest (~1.2% of revenue). No reverse factoring or revenue-recognition red flags identified.

How CapEx-hungry? Very — ~$159M/year (~13% of revenue), ~$1.2M per new restaurant; this is growth capex (~70–75 new units/yr), not maintenance. Maintenance capex is a small fraction; FCF would be far higher if growth stopped.

Capital Allocation & Management

FCF generation and use; philosophy? Fact: FY2025 FCF ~$26M (OCF $184.8M − capex $158.7M); ~100% reinvested into new units. No dividend, no buyback — the correct policy at ~50% cash-on-cash and a ~120x multiple.

Significant acquisitions recently? None since the 2018 Zoe’s Kitchen deal (~$300M, strategically excellent, now fully converted). Only a small ~$5M convertible-note investment in FY2025.

Buying back shares? No buyback program.

Issuing large amounts of stock to insiders? SBC modest (~1.2% of revenue); share-count growth is IPO-mechanical (preferred conversion), not chronic dilution.

Compensation policy / incentives? Fact: FY2025 bonus = 67% adj EBITDA / 33% revenue, no return metric; LTI was time-vested only. ROIC enters comp for the first time with 2026 PSUs (50% Adj ROIC / 50% Adj EPS). Say-on-pay passed ~91%. CEO FY2025 comp ~$3.86M.

Motivations of management? Interpretation: Founder-led (Schulman CEO since 2010; Xenohristos CCO; Shaich Chairman). Aligned by large historical equity stakes — but founders/PE sold ~$3.4B near the 2024 highs with near-zero conviction buying since, a fair tell on perceived value.

Valuation & Market Data

ADR, MLP, or K-1 issuer? No — a U.S. C-corp common stock (single class), NYSE-listed.

Dividend policy? None (no dividend).

How profitable? Restaurant-level: excellent (25%). Consolidated: thin (6.7% operating margin, ~5% net), early-stage.

Net income diverging from cash from operations? Fact: OCF ($184.8M) >> net income ($63.7M) — normal for a restaurant (D&A, lease accounting, negative cash-conversion cycle). FCF is far below OCF because of growth capex. No adverse divergence.

Risks & Downside

What would cause the stock to decline? Primarily a multiple de-rating on an SRS stumble (as in 2025, −75%); secondarily margin compression, new-market disappointment, or a food-safety event.

Risk of a catastrophic loss? Low — net-cash, profitable, growing; a brand-destroying food-safety event is the main tail.

Chance of a total loss? Very low — no leverage, no solvency risk; total loss would require simultaneous brand destruction and balance-sheet failure (remote).

Recent News & Events

Has the business environment changed recently? Yes, favorably at the margin — Q1-FY2026 SRS reaccelerated to +9.7% on +6.8% traffic, management raised guidance, and UBS upgraded to Buy ($90 target, June 10, 2026).

Significant acquisitions? None recent.

Change in accounting policies? None material.

Recent changes — markets, facilities, management? COO transition (Somers out Sep-2025, Thompson in Jan-2026); new ~$30M Virginia dips/spreads production facility (Mar-2026); 2026 PSU/ROIC compensation program; ongoing entry into Midwest/Southeast markets; routine credit-facility amendment (Mar-2026).

APPENDIX B — Source Appendix — CAVA Group, Inc. (NYSE: CAVA)

Report date 2026-06-27. Primary sources prioritized over secondary. Facts traced to filings, company data, and third-party quantitative feeds; management commentary treated as hypothesis and validated against filings/financials.

Primary — SEC Filings (CIK 0001639438; mirrored locally in output/CAVA/sources/)

Primary — Company / Market Data

Secondary — Industry & Peer Context

Data quality / reconciliation notes

• FY2024 net income normalization. GAAP net income $130.3M includes an ~$83.7M deferred-tax valuation-allowance release (income-tax line a $70.4M benefit; pretax income only $59.9M). Adjusted net income ~$50.2M. Trailing P/E and YoY-EPS comparisons must use the normalized figure.
• EV computed at the live price. Market cap = 116.4M shares × $83.40 ≈ $9.71B; net cash ~$403M (cash + short-term investments, no funded debt) → EV (ex-capitalized-leases) ~$9.3B. ROIC.ai’s snapshot EV is struck near the prior quarter-end price — re-derived at the live price for all current multiples.
• AZI own-history percentile discounted. CAVA’s ~2.5-year trading history includes the 2024 bubble (EV/sales ~20x), so the 44th-percentile composite understates absolute richness; valuation conclusions use absolute multiples vs. growth/returns.
• FactorsToday related-stocks unreliable here — returned industrials (WAB/CEG/HWM), not restaurants; factor-similar-peer cross-check not used.
• Management commentary (guidance, AUV, cash-on-cash, 1,000-unit target) treated as hypothesis and reconciled to filings/financials where possible; build-cost and cash-on-cash figures are management-sourced and flagged as such.