Starbucks Corporation (NASDAQ: SBUX) — The Turnaround Is Stronger; the Margin of Safety Is Not

Independent institutional research update. Report date: 2026-08-14. Reference price $108.55 at the August 13, 2026 close.

⚡ Claude’s Take

This block is a subjective opinion and is not investment advice. The institutional body below is deliberately position-free and carries no price target.

Verdict: HOLD / AVOID-HERE. The operating evidence is better than it was in June, but at $108.55 the shares still offer no execution discount. A more defensible accumulation zone is roughly $75–90, corresponding to about 21–25x the midpoint of management’s FY2028 EPS framework. Conviction: medium-high.

The Q3 print makes the turnaround harder to dismiss. Global comparable sales accelerated to 7.9%; U.S. comps were 7.9% with transactions up 4.2%; North America posted its first year-over-year margin expansion since early FY2024; and management raised FY2026 adjusted EPS guidance to $2.55–$2.65. China proceeds funded $1.8B of debt repayment in the quarter, resolving the prior report’s most important capital-allocation question in the right direction. This is now observed demand repair plus early earnings recovery, not merely a management plan.

Two qualifications stop me from paying the current price. First, management disclosed that roughly half—or slightly less—of the U.S. comp came from store-closure sales transfer and delivery growth. Those are economically useful, but they are not the same as organic visit growth at an unchanged store base. Second, Q3’s 14.4% adjusted consolidated margin excludes 390 basis points of restructuring, transaction and transformation costs; GAAP margin was 10.5%, while North America converted an 8.1% comp into only 30 basis points of reported margin expansion because labor, mix and restructuring absorbed most leverage. The bull traffic test passes; the margin test does not yet.

At $108.55, the stock trades at about 41.8x the midpoint of FY2026 adjusted EPS guidance and 29.5x the midpoint of management’s FY2028 $3.35–$4.00 framework—before discounting two years of execution. The current price itself implies approximately $4.34 of normalized EPS at 25x, above the framework midpoint and above its $4.00 high end. The own-history valuation screen is similarly demanding: P/E is at the 92.9th percentile and P/S at the 62.2nd percentile; P/B is meaningless because equity is negative. The single thing that would flip me bullish: clean, stable-base transaction growth plus a durable North America margin in the mid-teens, or a price that stops capitalizing that outcome in advance. The single thing that would flip me bearish: U.S. transactions turning negative for two consecutive quarters, or FY2027 GAAP margin failing to clear the low-double-digits despite the savings program. Tag: better proof, worse odds.

Stock Price Action — Five-Year Event Map

On AZI’s adjusted series, SBUX moved from $105.00 on August 16, 2021 to a $62.11 intraday low in May 2022, reached a $113.64 five-year intraday high in March 2025 and closed at $108.55. It is 4.5% below the five-year high and 3.4% above its starting close. The 52-week intraday range is $76.49–$110.51. Price moves are facts; attributions are interpretations.

1. Executive Summary

Starbucks is a globally scaled premium-coffee brand rebuilding earnings after a self-inflicted FY2025 trough. FY2025 operating income fell to $2.94B and margin to 7.9% as labor reinvestment, traffic deleverage and $892M of restructuring hit simultaneously. The first nine months of FY2026 show a genuine demand recovery: U.S. transactions were positive in all three quarters, Q3 global comps reached 7.9%, and management now expects roughly 6% global and slightly more than 6% U.S. comps for the year.

The evidence is strongest on sales and mixed on owner earnings. Q3 GAAP operating margin was 10.5%, up 60 basis points, versus 14.4% on management’s adjusted basis. North America margin was 13.6%, up 30 basis points, but year-to-date remains 11.8%, down 210 basis points. TTM FCF improved to $3.64B and now covers the $2.81B dividend 1.30x; however, capex fell to $1.34B from $2.31B in FY2025. Holding year-to-date capex at the prior-year pace would leave dividend coverage at only 0.83x. The balance sheet is materially safer after funded debt declined to $13.28B and net funded debt plus leases to $18.51B.

China is now a licensed/equity-method exposure. Starbucks received $3.1B, retained 40% of the venture, booked a $536.3M pre-tax gain below operating income, and used proceeds to reduce debt. This improves capital intensity and reported International margin, but it transfers rather than removes end-market risk: Luckin had 36,310 stores at June 2026, was adding units faster than revenue, and reported negative same-store sales.

The competitive advantage is narrow, not absent. Brand, habit, density, digital/stored value and procurement scale reinforce each other, especially in dense U.S. markets, while Channel Development and licensing convert the brand into high-return economics. Yet switching costs are close to zero, the recovery requires permanent labor and store investment, and competitors can scale cheaper formats. Filing-reconciled lease-adjusted ROIC is roughly 14% reported and 19% excluding clearly identified restructuring—better than the trough, but not fortress evidence.

The central debate has shifted from “is traffic recovering?” to “how much of the recovery is organic and how much margin can Starbucks retain?” At $108.55, approximately $124B of equity and $133.1B of lease-excluded enterprise value, the market prices a large portion of FY2028 success in advance. This memo makes no recommendation or price target outside Claude’s Take; it defines the embedded expectations and observable falsification tests.

2. Business Overview

What Starbucks is. Starbucks Corporation (founded 1971, Seattle; IPO 1992) is the world’s largest specialty-coffee retailer, with 41,304 stores across 90 markets at June 28, 2026. Following China deconsolidation, the system is now 33% company-operated and 67% licensed. North America contains 18,371 stores, including 11,149 company-operated; International contains 22,933, of which only 2,464 are directly operated. The company still carries labor, occupancy and product costs for its North American operating core, while the international and Channel Development layers are increasingly capital-light.

Three reportable segments (Q3 FY2026):

North America generated 79% of Q3 revenue and remains the dominant earnings driver. Channel Development is the cleanest asset-light stream. International’s 19.1% margin is not comparable with the old basis: converting China added roughly 800 basis points even though segment operating income fell 7%.

How it makes money. The direct business is a premium, beverage-led, high-frequency retailer. Q3 company-operated revenue was $7.51B, licensed-store revenue $1.20B and other revenue $616M. The annuity-like overlays matter disproportionately: Channel Development, international royalties, income from the China equity investment, and Starbucks Rewards. U.S. 90-day active Rewards membership was 35.8M; current stored-value and loyalty deferred revenue was $1.818B. Neither loyalty nor stored value is a contractual customer lock-in, but both lower friction and supply customer financing.

Verdict: Starbucks is now a hybrid: a labor- and lease-intensive North American retailer with an increasingly asset-light international and consumer-products perimeter. The mix shift improves capital efficiency, but the economic outcome still depends on company-operated U.S. throughput.

3. Industry Dynamics

Market boundary. The broad U.S. restaurant/foodservice sales ceiling is $1.55T in 2026. A narrower reproducible proxy is Census NAICS 722515—snack and nonalcoholic beverage bars—which generated $69.27B of employer and nonemployer revenue in 2023. Starbucks’ Q3 U.S. revenue annualizes to $27.56B and its 16,933 U.S. stores equal roughly one-fifth of the 85,140 employer establishments in that bucket. These are scale comparisons, not market share: dates and accounting bases differ, and coffee sold through QSR, convenience and grocery lies outside the bucket.

Profit pool. Growers carry crop/weather risk; store operators carry rent, hourly labor, equipment and throughput risk; brand owners spread sourcing, roasting, marketing and technology over large volumes. Starbucks’ Q3 margins locate the attractive economics: North America 13.6%, International 19.1% on its new licensed-heavy basis, and Channel Development 52.1%. Stores create brand and habit, but royalties and packaged goods monetize them with less capital.

Demand is habitual, not captive. U.S. transactions rose 4.2% after falling 3.7% a year earlier, leaving the two-year stack roughly flat. Pricing contributed less than one point to ticket. The consumer can still switch on every occasion among Starbucks, Dunkin, McCafé, convenience, local cafés and home brewing.

Supply remains the constraint. Starbucks’ U.S. units fell 2%, but industry capital is not withdrawing. Dutch Bros reached 1,225 shops in June, up 17.4%; McDonald’s operates more than 46,000 restaurants; small-box drive-through and pickup formats attack the occasion with lower occupancy/labor. North America is a late-expansion cycle with selective incumbent pruning, not broad contraction.

China is more adverse. Luckin reached 36,310 stores after adding 2,714 in Q2; revenue rose 28.5% while self-operated same-store sales fell 5.3%. Starbucks transferred 7,991 stores to Boyu and still aspires to as many as 20,000. Unit growth outpacing revenue and falling same-store productivity are classic capital-cycle warnings. Starbucks transferred operating risk; royalties and its 40% stake still depend on unit returns.

Labor is structural. Beverage-bar wages rose about 4%; service improvement requires permanent hours; roughly 5% of U.S. locations are unionized. Q3 tariff refunds largely offset first-nine-month tariff cost and are mostly received. Coffee may ease with Brazil supply, but purchase lags delay benefit.

Greenwald verdict: narrow-to-moderate U.S. advantage from density, habit and scale; no broad China moat. Marathon verdict: supply neutral-to-negative in North America and negative in China. Direct cafés are structurally mixed; brand/IP layers are attractive.

4. Competitive Position

Starbucks’ proposition combines premium beverages, customization, predictable execution, digital payment/order history, convenient locations and a “third place.” The ingredients are replicable; the combined system is harder to reproduce. Q3 food availability approached 99%, service-time targets were met on average across access points, and management reported brand consideration and purchase intent at five-year highs.

Captivity is moderate. U.S. Rewards had 35.8M active members and deferred revenue was $1.818B, but membership rose only modestly from Q1 and cohort visits/retention are undisclosed. Stored value lowers friction and supplies financing; it does not bind the next purchase.

Scale is local and load-bearing. Starbucks’ North American system is about fifteen times Dutch Bros’ size. Density supports convenience, procurement, roasting, distribution, marketing and technology. The disadvantage is complexity: overlap, mobile congestion, aging stores and labor. More than 1,000 uplifts were completed at roughly $150,000 each; management reports a transaction halo but no cohort return or cannibalization data.

Comp-quality caveat. The CFO said roughly half—or slightly less—of U.S. comp came from closure sales transfer and delivery. These are economic revenue, not equivalent to organic visits at unchanged stores. Closed units also leave the comp base, adding survivor effects.

Global purchasing/roasting gives a cost advantage over independents, but no proprietary input/process excludes large chains. Starbucks’ customized service can be structurally costlier.

Verdict: WEAK/NARROW competitive advantage. Brand, habit, density and scale reinforce each other and explain the recovery. They did not prevent FY2025’s collapse, do not lock in customers and require continuing labor, technology and store investment. The moat is valuable but maintenance-intensive.

5. Growth History and Forward Opportunities

Revenue rose from $29.1B in FY2021 to $37.2B in FY2025 while operating margin fell from 16.8% to 7.9%—growth without economic progress. FY2026 composition is better: U.S. transactions were +3.0%, +4.3% and +4.2% in Q1–Q3, versus -4% for FY2025.

Green Apron Service restores staffing, routines, coaching and store accountability. Two-thirds of North American company stores scored at least four positive “shots” on the internal GROW report, up more than 40 points since October. Availability, service times and leader tenure improved. These support management’s causal story but are unaudited and lack cohort economics.

Morning generated the largest absolute transaction gain; Refreshers, matcha, food and digital menu boards expand afternoon occasions. Food attach reached a Q3 record and Refreshers revenue grew double digits. Existing-store throughput is the best growth path if incremental transactions leverage labor already scheduled; it is less attractive if each step requires more hours.

The redesigned Rewards program and app delivery can deepen frequency. One in three members who tried a Free Mod Monday customization reordered one later. The missing evidence is incrementality: delivery is growing, but its inclusion in the comp bridge obscures channel shift versus new demand.

Uplifts exceeded 1,000 and the FY2026 goal rose to at least 1,500. Work occurs overnight and costs roughly $150,000, less than historic remodels. International is about 90% licensed after China, shifting unit growth to partners. Starbucks still expects 600–650 net FY2026 openings, while North American company-store growth may remain modest through FY2027.

Management now expects Q4 U.S. comps of at least 6.5%, FY2026 U.S. comps slightly above 6%, global comps near 6%, adjusted margin above 11% and adjusted EPS $2.55–$2.65. The FY2028 framework remains 13.5–15% margin and $3.35–$4.00 EPS, supported by $2B of gross savings. Gross savings must first absorb labor, wages and reinvestment.

Growth verdict: traffic, operations and asset mix improved. The highest-quality path is installed-base throughput plus licensed International growth. Stable-base transactions, uplift returns and post-labor margin—not unit count—are the required proof of value creation.

6. Financial Quality

FY2025 includes $892M restructuring and a $145.2M favorable litigation-settlement benefit. Adding back the former and subtracting the latter yields about 9.9% normalized margin, roughly 510 bps below FY2024. Q1–Q3 FY2026 GAAP margins were 9.0%, 8.7% and 10.5%; Q2/Q3 expanded year over year, but nine-month margin is still 9.4% versus 9.6%.

North America’s Q3 bridge was +340 bps sales leverage, +120 lapping the Leadership Experience and +110 lower inflation/refunds, offset by -240 restructuring, -190 labor and -100 mix. An 8.1% comp generated only 30 bps of GAAP expansion. International’s 550-bp expansion includes ~800 bps from China licensing; operating income fell 7%.

Q3 GAAP operating income/margin was $980.4M/10.5%. Adding $302.6M restructuring, $44.1M transaction and $18.1M transformation cost yields adjusted $1.345B/14.4%. Reported EPS $0.91 includes a $536.3M China gain below operating income; adjusted EPS is $0.85. Adjusted margin shows potential but is a ceiling because excluded execution cost, mix and refunds are real.

TTM FCF $3.64B covers $2.81B dividends 1.30x. Yet capex fell 52% year to date; at $2.0–$2.5B normalized capex, dividend coverage is only 0.88–1.06x. Cash/investments were $3.92B, funded debt $13.28B, leases $9.16B and net obligations including leases $18.51B. Equity remains negative $7.67B, making P/B, ROE and debt/equity meaningless.

Lease-adjusted ROIC. A conservative operating-asset denominator averages PP&E, ROU assets, goodwill/intangibles, equity investments, other operating assets and working capital, without subtracting long-term deferred revenue. TTM NOPAT plus only imputed after-tax lease interest produces ~14.4% reported ROIC; adding clearly identified restructuring produces ~19.4%. Full rent is not added back.

Filing sweep. The 493-filing census reconciles all 397 required mirrored forms. There is no restatement, material weakness, auditor change or non-reliance filing; the 2024 10-Q/A only added an omitted 10b5-1-plan disclosure. Deloitte’s opinions were unqualified. Normalization still requires care:

• China held-for-sale accounting stopped D&A/ROU amortization before closing; the benefit is unquantified.
• China added $198.6M estimated incremental tax and $73.8M YTD transaction cost.
• Restructuring was $892M FY2025 and $415.8M YTD FY2026, with ~$230M still expected. Repeated closure cycles make part of it recurring owner cost.
• The proxy—not the 10-K legal note—identified a favorable $145.2M FY2025 litigation-settlement benefit, ~5% of GAAP operating profit.
• Post-close China affiliate flows were $52.5M revenue, $18.5M product cost and $30.7M receivable.

Liquidity is adequate: no CP/revolver balance, $3B unused facility and covenant compliance. Debt maturities are $1.5B FY2027, $1.08B FY2028 and $1.75B FY2029.

Verdict: demand, adjusted earnings and leverage are improving; owner earnings are not fully repaired. Year-to-date GAAP margin is down, FCF depends on low capex, and the core remains labor/lease intensive.

7. Capital Allocation

FY2021–FY2025 distributions exceeded FCF by $1.924B. Buybacks averaged $105.99; the error was poor opportunity-cost timing and weaker resilience, not a large nominal mark-to-market loss. Negative equity records that cumulative return exceeded retained earnings.

Current behavior is better. No repurchases occurred in FY2025 or FY2026 YTD. China generated $3.1B of consideration; Starbucks retained 40% and licensing rights, then repaid $2.816B of debt YTD including a $1.316B tender. Estimated annual pretax coupon savings, including scheduled maturities, are roughly $124M. This resolves the prior report’s key open question in the correct direction.

The $0.62 quarterly dividend has reported TTM coverage of 1.30x but only 0.88–1.06x under $2.0–$2.5B capex. The rational policy is to hold it flat until store investment and coverage normalize rather than protect a raise streak at any cost.

Nestlé was the clear partnership win; Teavana the clear loss; China control followed by divestiture is a strategic round trip. Boyu is sensible today but acknowledges poor stand-alone China returns.

Incentives have downside through adjusted revenue/income, EPS, comps and relative TSR, but omit FCF, ROIC and leverage. Pay scale and aircraft use weaken optics. Insider evidence is neutral: seven code-P rows totaled $22.0M over five years, mostly the stale 2022 cluster; the latest was Knudstorp’s $994,500 purchase in November 2025. All 2026 sales were disclosed 10b5-1 plans; no executive has bought since Schultz in 2022.

Verdict: B- / improving. Reinvest, de-risk China, reduce debt and stop buybacks is the right hierarchy. Management credibility is improving, not fully proven; the test is funding both normalized capex and dividend without reversing deleveraging.

8. Changes and Headwinds — Last Two Years

The traffic bounce is now sustained enough to weaken the prior bear case. It is not fully falsified: GAAP margin is 10.5%, stable-base organic traffic is undisclosed, and year-to-date North America margin remains below Q3. Repeated restructuring, further closures and a large Corporate loss show the steady state is not yet reached.

Supplemental Analysis: What Q3 Proves—and What It Still Does Not

Q3 FY2026 operating bridge: better demand, but three recoveries are embedded in one comp

Q3 was the strongest operating evidence since the turnaround began, but the 7.9% U.S. comparable-sales result should be decomposed before treating it as a single organic signal. U.S. transactions increased 4.2% and ticket increased 3.6%. Pricing contributed less than one percentage point to ticket, so most of the ticket increase came from delivery mix, record food attachment and paid beverage modifications. Management also said roughly half—or slightly less than half—of the total U.S. comp came from transferred sales after store closures plus delivery growth; extended hours contributed only a few basis points. The remainder came from store performance, menu and innovation. Delivery therefore appears in both the causal explanation for ticket and the broader closure/delivery bridge. It should not be counted twice as evidence of stronger in-store frequency. (Starbucks Q3 FY2026 earnings call transcript, July 29, 2026; Q3 earnings release, July 29, 2026)

This decomposition reveals three distinct recoveries. First, a store-execution recovery is visible in positive morning and afternoon transactions, food availability near 99%, service-time targets being met across every access point, and two-thirds of North American company-operated stores recording at least four positive “shots” in the GROW operating score. That GROW proportion increased about 40 percentage points since October, while stores led by managers with at least two years’ tenure performed seven percentage points better year over year on the score. Morning produced the largest absolute transaction gain, consistent with restored throughput and availability mattering most in the chain’s habitual daypart. Refreshers grew U.S. revenue at a double-digit rate and food attach reached a record, providing product evidence beyond transferred traffic. These measures are directionally consistent with labor and process changes working.

Second, a portfolio recovery is occurring as sales migrate from closed stores to surviving stores. It is economically rational to close structurally weak units and retain some of their demand, but this raises comparable sales without necessarily increasing customer occasions for the system. Permanent closures also leave the comparable-store base after the month of closure. North American store count fell by 363 year over year, and the U.S. count declined 2%. Some of Q3’s leverage is consequently the arithmetic of concentrating existing demand into a smaller denominator. That can improve per-store returns, but it is closer to capital rationalization than market-share growth.

Third, a channel-and-mix recovery is raising ticket. Delivery adds convenience and may recruit incremental occasions, while food and modifications increase basket size. Yet delivery carries fees and platform economics, and richer product mix can add labor and ingredient complexity. Q3’s North America margin bridge confirms the trade-off. Sales leverage contributed approximately 340 basis points, lapping the prior-year Leadership Experience added 120 basis points, and lower inflation plus tariff refunds added 110 basis points. Those 570 basis points were almost entirely consumed by restructuring at 240 basis points, added labor at 190 basis points, and product mix at 100 basis points. Rounding explains the bridge to the reported 30-basis-point expansion to 13.6%. (Starbucks Q3 FY2026 Form 10-Q, filed July 29, 2026)

The quality of that 30-basis-point gain is mixed. Sales leverage is the strongest recurring component. Lapping the Leadership Experience is a comparison benefit, not a new productivity source. Tariff refunds are nonrecurring and lower inflation may reverse. Conversely, labor investment is not an item to normalize away: it is part of the service proposition now required to retain traffic. Product mix may also remain a recurring headwind if customized beverages and food are central to ticket growth. Restructuring is unusual quarter to quarter, but Starbucks recorded major closure programs in FY2020–FY2021 and again in FY2025–FY2027; repeated portfolio correction is partly an owner cost, not wholly exceptional. Year-to-date North America margin of 11.8%, down 210 basis points, is therefore a more conservative anchor than the single-quarter 13.6% result.

The operating test for Q4 and FY2027 is not simply whether comps remain positive. It is whether management can disclose or demonstrate a stable-base transaction gain after isolating closure transfer and delivery, and whether sales leverage begins to exceed the ongoing labor, product and upkeep burden. More than 1,000 store uplifts had been completed by Q3, with at least 1,500 targeted for FY2026 at approximately $150,000 each—at least $225 million of implied capital for the target cohort. Management describes returns and a neighboring-store transaction “halo” as strong, but gives no cohort sales lift, contribution profit, cash payback or cannibalization. Until those data emerge, the uplift program is credible operational evidence but not yet proven store-level economics.

Greenwald moat test: the evidence supports local scale plus habit, not hard captivity

Under Greenwald’s taxonomy, durable advantages must come from customer captivity, lower costs, or economies of scale that entrants cannot efficiently replicate. Starbucks has elements of all three, but only the combination of local density and habit looks meaningfully defensible.

Demand captivity is real but soft. The chain had 35.8 million U.S. 90-day active Rewards members in Q3, and Rewards historically represented roughly 60% of U.S. company-operated revenue. Current deferred revenue, largely stored value and loyalty obligations, was $1.818 billion at June 28. A familiar nearby location, stored payment credentials, personalized offers, Stars and a known menu reduce search and transaction friction. Management’s brand tracking showed five-year highs in affinity, consideration and purchase intent, with growth across generations, income groups, members and non-members. One in three members who tried a modification during “Free Mod Mondays” later reordered one, suggesting promotions can seed paid habits. (Starbucks FY2025 Form 10-K, filed November 14, 2025; Q3 transcript)

The disconfirmation is equally important. A consumer incurs essentially no contractual or technical cost by buying coffee elsewhere. Competing apps are free, routines can change quickly, and Rewards actives rose only modestly from 35.5 million in Q1 despite the sharp comp recovery. Starbucks publishes neither visits per active member nor retention, frequency, control-group lift or member-cohort economics. Internal brand and GROW measures have no disclosed methodology. Most importantly, Q3 FY2025 U.S. transactions had declined 3.7%; compounding that with Q3 FY2026’s 4.2% increase leaves transactions only about 0.3% above the two-year-ago level. The recovery is genuine, but it has restored rather than materially expanded traffic so far.

Cost advantage exists relative to independents, not necessarily to scaled chains. Global green-coffee purchasing, roasting, logistics, product development, marketing and technology can be spread over 41,304 stores. Channel Development demonstrates the value of the brand and system without the full café cost base: Q3 revenue rose 22% to $588 million and operating margin was 52.1%, albeit with help from tariff refunds. The licensed network similarly earns royalty and product economics while partners carry more local capital. There is, however, no proprietary coffee input, process or mobile technology that another large chain cannot access. Starbucks’ labor-intensive customization and third-place promise can make its operating model structurally more expensive than a limited-menu drive-through or a franchise-heavy QSR system.

Economies of scale are strongest locally, where they reinforce captivity. Starbucks’ 18,371 North American stores make a Rewards relationship more useful because the brand is easy to find; density also supports regional distribution, advertising and digital investment. That installed base is about fifteen times Dutch Bros’ 1,225 shops. Yet national scale alone is not the barrier: coffee retail is won trade area by trade area, and challengers can build clusters with different box formats. Aging and overlapping stores can turn scale into complexity, as the closure and uplift programs demonstrate.

Financial outcomes keep the moat conclusion disciplined. A fortress franchise should normally exhibit stable margins and require modest defensive capital. Starbucks’ GAAP operating margin fell from 16.8% in FY2021 to 7.9% in FY2025, or approximately 9.9% after adding back restructuring and subtracting the favorable litigation-settlement benefit. Q3 shows latent operating leverage, but it also shows that a high-single-digit comp produced only 30 basis points of North America margin expansion after labor, mix and repair costs. The best reading is a narrow moat: valuable local scale and habitual demand, enhanced by an asset-light brand layer, but no hard switching costs and no unique input monopoly. Disconfirmation would come from stable-base traffic losing momentum after closure benefits fade, uplift returns failing to cover capital, or challengers sustaining rapid unit growth without declining returns. Confirmation would require several years of share-stable transaction growth and durable post-investment margin, not another strong quarter alone.

Capital-cycle comparison: Starbucks is pruning while competitors continue to add supply

The supply side is less favorable than Starbucks’ own store rationalization suggests. In North America, capacity is migrating rather than exiting. Dutch Bros ended June with 1,225 shops, up 182 or 17.4% year over year. Q2 system comparable sales rose 5.8%, transactions increased 1.7%, trailing-12-month system AUV reached $2.193 million, and company-operated shop contribution margin was 30.6%. Company-operated comps were still stronger at 8.3%, including 3.4% transaction growth. Its contribution margin excludes central expenses and cannot be compared directly with Starbucks’ segment margin, but the combination of rapid openings, positive traffic and rising AUV is exactly what would be expected when incremental capital still finds attractive sites. Dutch Rewards represented 73% of transactions, a further warning that Starbucks does not have exclusive ownership of beverage loyalty. (Dutch Bros Q2 FY2026 Form 10-Q, filed August 6, 2026)

McDonald’s provides a different comparison. It had 46,028 restaurants at June 30 and approximately 95% were franchised. Q2 U.S. comps rose only 0.8% as higher check offset negative guest counts, so Starbucks currently has the better headline traffic signal. But McDonald’s 46.2% consolidated operating margin reflects a structurally more asset-light, franchise- and real-estate-backed model, and its value positioning competes for the same breakfast and beverage occasions. Starbucks therefore faces pressure from both a fast-growing specialist with efficient drive-through boxes and a vast QSR network that can defend value without matching the café experience. (McDonald’s Q2 2026 Form 10-Q, filed August 7, 2026)

China is further into the expansion phase. Luckin ended Q2 with 36,310 stores after adding 2,714 net units in one quarter. Revenue increased 28.5% and monthly transacting customers rose 22.9% to 112.7 million, but self-operated same-store sales fell 5.3% and store-level margin slipped to 21.3%. At year-end 2025, 99.1% of Luckin’s self-operated stores were compact 20–60 square-meter pickup units. Unit growth running ahead of revenue while same-store sales decline is classic capital-cycle evidence of cannibalization or saturation, yet a 21%-plus store margin indicates the market has not reached the capacity-withdrawal stage. Starbucks’ 7,991-store China venture is far smaller and its ambition to reach as many as 20,000 stores would add to supply, not relieve it. (Luckin Q2 2026 results, August 3, 2026; Luckin FY2025 Form 20-F)

The capital-cycle conclusion is counterintuitive: Starbucks’ pruning improves its own average-store economics, but does not create category scarcity. North American supply continues to enter through small boxes, drive-throughs, licensed stores and QSR beverage platforms; China remains in a capacity boom. The most attractive position is increasingly the owner of brand, IP, product and royalty economics rather than the funder of every café. Starbucks’ Channel Development margins and China re-partnering reflect that logic.

China: improved capital intensity, new accounting basis, unresolved underlying returns

On March 30, 2026, Boyu Capital acquired 60% of the China retail business at an approximately $4.0 billion cash-free/debt-free enterprise value. Starbucks received $3.1 billion of total consideration, including its share of joint-venture debt proceeds, and retained a 40% equity-method interest initially valued at about $1.2 billion. It derecognized $3.4 billion of net assets, recycled $282.8 million of cumulative translation losses and $99.7 million of net-investment-hedge losses, and recognized a $536.3 million pretax divestiture gain. Estimated incremental tax was $198.6 million, while transaction costs were $44.1 million in Q3 and $73.8 million year to date. These are transaction accounting items, not operating earnings. (Starbucks Q3 FY2026 Form 10-Q)

The comparability break began before closing. While China was classified as held for sale in Q1 and Q2, Starbucks ceased depreciation of its property, plant and equipment and amortization of lease right-of-use assets in the disposal group. The company did not quantify the resulting benefit. After closing, 7,991 China stores became licensed rather than company-operated for Starbucks’ reporting purposes. International Q3 margin rose 550 basis points to 19.1%, but the ownership-model change contributed approximately 800 basis points and International operating income actually declined 7.3%. The margin increase is therefore a new accounting perimeter, not evidence that China stores suddenly became more productive.

Future China economics have three components: royalty and product profit recorded in revenue, 40% of joint-venture earnings recorded through equity-method accounting, and the value or risk of the retained interest. In the partial post-close quarter, Starbucks recorded $52.5 million of sales and royalty revenue from the JV, $18.5 million of related product and distribution cost, and a $30.7 million receivable. The implied $34 million gross spread cannot be annualized confidently because the filing does not separate royalties from product sales, identify central costs, or provide a full-quarter basis. Nor does Starbucks disclose JV AUV, same-store sales, unit-level return on invested capital, royalty rate, debt, interest burden or distributions.

Strategically, the transaction transfers lease, labor and new-unit funding risk to a locally connected partner while preserving brand ownership and upside. Financially, it is not a clean franchise annuity yet: Starbucks still owns 40%, and royalties ultimately depend on store productivity in an over-supplied market. Using proceeds for $2.816 billion of year-to-date debt repayment, including a $1.316 billion tender, was a prudent conversion of transaction liquidity into lower fixed claims. But the disposal cannot establish that the earlier move to full China ownership earned an attractive return because intervening capex, cash flows, FX and distributions are not disclosed on a comparable perimeter.

Management credibility: execution evidence is ahead of owner-earnings evidence

Management credibility has improved on observable actions. U.S. transactions have been positive for three consecutive quarters; Q3 operating metrics strengthened; China closed on announced terms; proceeds were directed to debt rather than buybacks; and FY2026 adjusted EPS guidance was raised to $2.55–$2.65. The Q3 call also deserves credit for acknowledging that closure transfer and delivery produced roughly half or slightly less of the U.S. comp. That disclosure makes the headline less flattering but the narrative more testable.

The remaining credibility gap is translation from operational claims to durable owner earnings. Management calls uplift returns strong but supplies no cohort economics. Internal brand, customer-connection and GROW scores are not accompanied by methodologies or raw series. The proxy credits management with more than $2 billion of P&L cost reductions, but the filings do not reconcile gross savings with reinvestment and retained margin or free cash flow. Repeated restructuring also complicates adjusted results: FY2025 recorded $892 million of restructuring and impairment; FY2026 had recorded another $415.8 million through Q3, with approximately $230 million more expected through the first half of FY2027. Adjusting each program away helps compare quarters, but overstates through-cycle earnings if large portfolio repairs recur. (Starbucks FY2026 proxy, filed January 26, 2026; Starbucks May 2026 restructuring Form 8-K)

Incentives are meaningful but incomplete. FY2025 adjusted operating income achieved only 76.8% of target and produced a zero factor for that portion of the annual plan, yet the board awarded most named executives a 100% individual factor, lifting the overall payout to 54.8% of target. The June 2025 Back to Starbucks award requires at least $800 million of cumulative FY2026–FY2027 operating-expense savings before strategy and relative-TSR adjustments, but has no explicit free-cash-flow, ROIC, leverage or dividend-coverage metric. This design can reward gross cost removal before proving that savings survive reinvestment. The proper credibility standard is thus asymmetrical: reported traffic and transaction execution deserve greater weight now, while margin, savings and return claims require external financial confirmation.

Bull/bear evidence scorecard

The prior bear case is now partially falsified: positive transactions and a North America margin above 12% occurred together in Q3, earlier than a strict bear view allowed. The prior bull case is also not fully proven, because the margin result is not yet sustained, the comparable-sales bridge contains substantial transfer and delivery effects, and uplift economics remain undisclosed. The decisive next evidence is stable-base organic traffic plus repeatable margin after labor, product mix, normal inflation and recurring portfolio upkeep. That standard keeps the Q3 improvement in proper proportion: it is substantial operating progress, but not yet proof that Starbucks has restored its historical earnings architecture.

9. Risk Analysis

The dominant risk is correlated earnings and multiple compression. If traffic fades, labor deleverage slows margin recovery and the market can stop capitalizing FY2028 success at a premium. The upside risk is symmetric: retained savings and mix could lift earnings above the framework.

10. Valuation Discussion (Embedded Expectations)

At $108.55 and 1.1398B shares, equity value is ~$123.7B. Adding $13.28B funded debt and subtracting $3.92B cash/investments gives conventional EV ~$133.1B. Leases are excluded; adding the $9.16B liability is appropriate only with lease-adjusted earnings.

AZI’s $1.3102 TTM EPS differs from filing-derived ~$1.74 due to vendor convention/period mix; the percentile is used for rank only, not absolute P/E.

Current peer screen. Refreshed August 13 equity multiples and lease-inclusive ROIC.ai EV multiples are useful only with business-model caveats.

SBUX’s 41.8x FY2026 adjusted P/E exceeds CMG and mature franchisors; its 29.5x FY2028 midpoint merely reaches today’s CMG multiple. EV/Sales is misleading because franchise revenue is a high-margin slice of system sales while Starbucks consolidates most North American store revenue.

Reverse-DCF cross-check. Using equity value, levered CFO-minus-capex FCF, ten years, an 8% equity discount rate and 3% terminal growth, reported $3.642B TTM FCF requires approximately 9.3% annual growth for ten years. Starting from a $2.680B capex-normalized stress requires 13.2%. A no-transition perpetuity requires $6.19B next-period FCF, 70% above reported and 131% above normalized. This is not a forecast; it shows that terminal durability beyond FY2028 is load-bearing.

Reverse EPS.

TTM operating income is $2.98B. A 13.5–15% margin on ~$40–42B revenue implies ~$5.4–$6.3B, roughly doubling profit. The price underwrites much of that bridge.

These are sensitivity assumptions, not targets. Earnings and multiple are correlated: failure removes both, while success must preserve a premium. McDonald’s/franchisors deserve higher margins and lower capital intensity; CMG/BROS deserve faster units. Starbucks sits between, and neither P/E nor EV/EBITDA alone captures leases/capex.

Verdict: more credible turnaround, still efficiently priced to rich. Upside requires earnings above the current FY2028 framework rather than mere delivery of its midpoint.

11. Variant Perception

The tape already agrees with the turnaround. SBUX is up 30.9% YTD and 19.3% over twelve months, above its 21/50/200-day EMAs by 3.3%/4.7%/11.5%.

Negative incremental Momentum, zeroed Quality and only 22% explained variance indicate a company-specific rerating, not a crowded momentum factor. Short interest is modest. Positioning verdict: strong trend, neither one-way factor street nor falling knife.

Bull: three positive-transaction quarters; operating KPIs improve; savings/mix can double trough profit; China is de-risked; debt fell; guidance rose.

Bear: comp provenance is mixed; 8.1% NA comp produced 30 bps margin; adjusted margin excludes 390 bps; FCF depends on capex; China/North America supply is aggressive; price is 29.5x FY2028 midpoint.

Variant view. The under-discussed issue is comp conversion. Transfer, delivery, ticket/mix and survivor effects can lift reported comps without unchanged-store frequency. Conversely, closing bad units and routing volume through a fixed network creates real value. Stable-base transactions and post-labor cash margin decide the debate.

Falsification. Bull fails with two negative U.S. transaction quarters or FY2027 GAAP margin below 11%. Bear fails with two more quarters of stable-base mid-single-digit comps/positive transactions, GAAP margin above 12% and dividend coverage after normalized capex.

12. Fact vs. Interpretation Table

13. Open Questions

1. What portion of transactions is closure transfer, delivery, member/nonmember frequency and incremental café demand?
2. What are uplift cohort sales, margin, cannibalization and cash returns?
3. What is steady-state North America margin after permanent labor?
4. How much of $2B gross savings is realized, retained and reinvested?
5. What is normalized maintenance plus growth capex?
6. Can $2.48 dividends be funded after that capex without borrowing?
7. What are China JV royalty, product and equity-income economics?
8. Do JV unit returns improve while China capacity expands?
9. What are Rewards retention, visits and spend by tier/cohort?
10. Will FY2027 systems modernization improve throughput without disruption?
11. How many North American closures remain and what transfer do they create?
12. What produced the $145.2M FY2025 favorable litigation settlement?

14. What Must Be True

Bull: transactions stay positive after transfers/weak bases fade; NA normalized margin reaches mid-teens; GAAP margin clears 12% by FY2027 and approaches 13.5%; at least half of gross savings remains; normalized FCF covers dividend and stores; China/licensing maintains standards. Falsification: two negative transaction quarters, FY2027 margin below 11%, or normalized FCF uncovered.

Bear: comps fade; labor holds NA margin near 11–13%; savings are reinvested and capex rises; China supply dilutes returns; valuation moves toward mature-QSR levels. Falsification: stable-base MSD comps/positive transactions for two more quarters, GAAP margin above 12% and post-capex dividend coverage >1x.

The thesis resolves on stable-base U.S. transactions, GAAP margin and FCF after normalized capex. Current evidence only partly clears the first hurdle.

15. Source Appendix

See the separate Source and Diligence Appendices. Core evidence is the FY2025 10-K, Q1–Q3 FY2026 10-Qs, Q3 release/call, FY2026 proxy, five-year SEC/insider corpus and current competitor filings. AZI supplies adjusted price and own-history valuation ranks; ROIC.ai and FactorsToday are reconciled third-party analytical cross-checks.

The institutional body has no recommendation or price target; the labeled Claude’s Take is a subjective opinion.

APPENDIX A — Standard Diligence Questionnaire

Supplemental to the research memo. Report date 2026-08-14. Fact, interpretation and assumption labels are applied where material.

General

What is the central debate? Whether Starbucks’ transaction-led recovery represents durable, stable-base habit repair and whether sales leverage can overcome permanent labor, store reinvestment and licensed-mix changes. Q3 weakened the “temporary bounce” case, but management’s disclosure that roughly half or slightly less of U.S. comp came from closure transfer and delivery prevents a clean conclusion.

What changed since the June report? Q3 global comps accelerated to 7.9%; U.S. transactions remained above 4%; adjusted EPS guidance rose to $2.55–$2.65; China closed and deconsolidated; $2.816B of debt was repaid year to date; and the stock rose to $108.55. The prior traffic concern weakened and the capital-allocation question resolved favorably. The valuation and GAAP-margin questions intensified.

Cyclicality and Earnings Nature

High or low earnings? GAAP earnings remain near a deliberate trough. TTM GAAP operating margin is 7.8% versus 15.0% in FY2024; Q3 improved to 10.5%. Management-adjusted Q3 margin of 14.4% shows potential but excludes restructuring, transaction and transformation costs.

External or internal? Predominantly internal. FY2025 combined operational deterioration with chosen labor reinvestment and restructuring. Coffee, tariffs, wage inflation and consumer caution are external overlays. The recovery is being driven by staffing, queue management, availability, marketing and portfolio pruning.

Revenue stability? Revenue is stable, but China deconsolidation makes reported growth hard to read. Q3 revenue fell 1% while North America grew 7%. Comparable sales and segment operating income are more informative than consolidated revenue until the new basis laps.

Quality of demand? Better than FY2025 because transactions are positive and pricing contributed less than one point to ticket. Less clean than the headline because delivery, closure transfer, sales mix and survivor effects are not separately disclosed.

Business Quality and Competitive Moat

Do brands matter? Yes. Brand recognition, ritual, customization, digital convenience and density support a premium and repeat use. The brand is economically visible in 52.1% Channel Development margin and the ability to license globally.

Switching costs? Low. Rewards and stored value reduce friction but do not contractually bind customers. Consumer captivity is habit-based and leaky.

Scale advantage? Meaningful in local density, procurement, roasting, distribution, technology and marketing. Not unique against McDonald’s or large packaged-goods systems. Scale also creates overlap, complexity and a large refurbishment burden.

Competitive direction? More competitive. Dutch Bros is expanding U.S. capacity at a mid-teens rate; Luckin reached 36,310 stores and is adding units faster than revenue; alternative QSR and convenience formats compete on value and speed. Starbucks has a narrow U.S. moat and no demonstrated broad China moat.

Industry return structure? Best at brand/IP, royalties and packaged goods; weaker at direct cafés bearing labor and occupancy. China is in an expansion phase with adverse capital-cycle signals. North America is late-cycle with selective pruning, not broad capacity withdrawal.

Financial Condition

Five-year economics? FY2021–FY2025 revenue rose 28%, while GAAP operating income fell 40%. This rejects the proposition that recent scale automatically improved economics.

Cash conversion? TTM CFO is $4.99B and FCF $3.64B. The China gain is reversed in operating cash flow; sale proceeds appear in investing cash flow. Working-capital and tax timing helped nine-month CFO.

Dividend safety? Reported TTM coverage is 1.30x. Under the analyst assumption of normalized annual capex of $2.0–$2.5B, coverage is approximately 0.88–1.06x. The dividend is fundable today but not yet comfortably covered after normalized store investment.

Leverage? Funded debt is $13.28B and operating leases $9.16B. Cash/investments are $3.92B, yielding $18.51B of net funded obligations including leases. Debt and leases fell materially after China, improving resilience.

Negative equity? Negative $7.67B, largely created by cumulative distributions exceeding retained earnings. It makes ROE, debt/equity and P/B unusable but is not evidence of insolvency.

Off-balance-sheet items? Operating leases are recognized. The important analytical adjustment is to treat them as debt-like while adding back only imputed lease interest in ROIC, not full rent.

ROIC? Filing-reconciled lease-adjusted ROIC is approximately 14.4% reported and 19.4% excluding clearly identified restructuring. Definition sensitivity is high because negative equity and customer financing distort conventional denominators.

Capital Allocation and Management

Historical record? Poor. FY2021–FY2025 dividends and repurchases exceeded FCF by about $1.9B. Buybacks averaged approximately $105.99; the problem is opportunity cost and balance-sheet timing, not a large nominal mark-to-market loss.

Current record? Improving. Management stopped buybacks, invested in stores, sold 60% of China on reasonable headline economics and used cash for debt. The debt tender and maturities should save roughly $124M of annual pretax coupon expense.

M&A? Nestlé alliance is the clear win; Teavana the clear loss; China control followed by divestiture is a strategic round trip. Boyu is a sensible response to current conditions.

Incentives? Financial metrics and relative TSR create real downside in awards. The omission of FCF, ROIC and leverage from primary incentive metrics is a weakness. Pay scale and aircraft perks remain governance concerns.

Insider activity? Neutral. One material open-market director purchase in November 2025; no broad discretionary buying by management. Most activity is grants, vesting and routine sales.

Valuation and Market Data

Current multiples? Approximately 41.8x FY2026 adjusted guide midpoint, 29.5x FY2028 framework midpoint, 34.0x TTM FCF and 28.7x conventional lease-excluded EV/TTM EBITDA, using approximately $133.1B EV and filing-derived TTM EBITDA of $4.64B. Each denominator has a caveat: FY2026 is a recovery year, FY2028 is not yet delivered, FCF benefits from low capex, and EBITDA sits at a restructuring trough.

Own-history context? AZI places P/E at the 92.9th percentile and P/S at the 62.2nd percentile as of August 13. Composite is 77.5th based on two valid components; P/B is null because equity is negative.

What does the price imply? $4.34 of normalized EPS at 25x, above management’s $4.00 FY2028 framework high end. Alternatively, the framework midpoint requires a 29.5x multiple. Both express the same embedded optimism.

ADR / K-1? No. NASDAQ common stock.

Dividend yield? About 2.28% on $2.48 annualized.

Risks and Downside

Primary downside? Organic comps slow, permanent labor caps margin, capex normalizes, and both earnings and multiple compress. China JV, union, commodity and implementation risks matter but are secondary to comp provenance, margin and valuation.

Catastrophic risk? Low. Starbucks has a global brand, strong CFO and improved debt profile. Lease obligations and negative equity reduce flexibility but do not imply imminent solvency risk.

Key-person risk? Meaningful. The largest five-year daily price gain occurred when Niccol was named, and the current operating system is closely identified with him.

Recent Events and Monitoring

Next decisive evidence: Q4 U.S. comps versus the at-least-6.5% guide; stable-base transaction disclosure; GAAP versus adjusted margin; full-year capex; dividend action; FY2027 technology plan; China royalty/equity income; and additional closures.

Bull falsification: two consecutive negative U.S. transaction quarters, FY2027 GAAP margin below 11%, or normalized FCF again failing to cover the dividend.

Bear falsification: stable-base mid-single-digit comps with positive transactions for two more quarters, GAAP margin above 12%, and dividend coverage above 1.0x after normalized capex.

APPENDIX B — Source Appendix

Report date 2026-08-14. Primary sources first.

Starbucks Primary Sources

1. Q3 FY2026 Form 10-Q, filed 2026-07-29, quarter ended 2026-06-28. Consolidated statements; China Note 2; leases Note 7; debt Note 8; equity/dividends/SBC; segment Note 16; MD&A and liquidity. https://www.sec.gov/Archives/edgar/data/829224/000082922426000130/sbux-20260628.htm
2. Q3 FY2026 earnings release, Exhibit 99.1, 2026-07-29. Comps, store counts, guidance, segment results and GAAP/non-GAAP reconciliation. https://www.sec.gov/Archives/edgar/data/829224/000082922426000129/sbux-06282026xearningsrele.htm
3. Q3 FY2026 earnings-call transcript, 2026-07-29. Management’s comp bridge, service/availability, Rewards, uplift cost, China economics, tariff normalization and FY2026 outlook. https://s203.q4cdn.com/326826266/files/doc_financials/2026/q3/Q3-FY26-SBUX-Transcript.pdf
4. FY2025 Form 10-K, filed 2025-11-14. FY2023–FY2025 statements; store counts; restructuring Note 18; segments; leases, debt, deferred revenue and risk factors. https://www.sec.gov/Archives/edgar/data/829224/000082922425000114/sbux-20250928.htm
5. FY2022 Form 10-K, filed 2022-11-18. FY2021–FY2022 statements and segment history. https://www.sec.gov/Archives/edgar/data/829224/000082922422000058/sbux-20221002.htm
6. Q1 FY2026 Form 10-Q, filed 2026-01-28, quarter ended 2025-12-28, https://www.sec.gov/Archives/edgar/data/829224/000082922426000011/sbux-20251228.htm; and Q2 FY2026 Form 10-Q, filed 2026-04-28, quarter ended 2026-03-29, https://www.sec.gov/Archives/edgar/data/829224/000082922426000080/sbux-20260329.htm. Quarterly comp/margin sequence and China held-for-sale detail.
7. China JV completion release / 8-K exhibit, 2026-04-02. Boyu purchase, 40% retained stake and licensing structure. https://www.sec.gov/Archives/edgar/data/829224/000082922426000064/sbux-04022026xexhibit991.htm
8. FY2026 DEF 14A, filed 2026-01-26. Compensation, incentive metrics, governance, pay ratio and ownership. https://www.sec.gov/Archives/edgar/data/829224/000121390026007780/ea0270973-01.htm
9. SEC Companyfacts, accessed 2026-08-14. Cross-check for revenue, operating income, net income, CFO, capex, dividends, repurchases, debt, leases, SBC and shares. https://data.sec.gov/api/xbrl/companyfacts/CIK0000829224.json
10. Five-year SEC filing corpus. Census reviewed: five 10-Ks, fifteen 10-Qs plus one amendment, sixty-seven 8-Ks plus two amendments, 208 Forms 4 plus amendments, and related proxy/ownership filings. Public index: https://www.sec.gov/edgar/browse/?CIK=829224

Strategy and Management Sources

11. Starbucks 2026 Investor Day, Starbucks IR, event dated 2026-01-29. FY2028 framework, $2B gross savings, Green Apron Service and unit/margin plan. https://investor.starbucks.com/events-presentations/event-details/2026/Starbucks-2026-Investor-Day-2026-w0k0vvnJag/default.aspx
12. Q2 FY2026 earnings call, Starbucks IR, 2026-04-28, https://investor.starbucks.com/events-presentations/event-details/2026/Starbucks-Q2-Fiscal-Year-2026-Earnings-Conference-Call-2026-i5MwCsdAaW/default.aspx; and Bernstein 42nd Annual Strategic Decisions Conference, Starbucks IR, 2026-05-28, https://investor.starbucks.com/events-presentations/event-details/2026/Bernstein-42nd-Annual-Strategic-Decisions-Conference-2026-D2IMXfaaFW/default.aspx. Operating leverage, fixed-cost and procurement discussion.
13. Starbucks leadership appointment, 2024-08-13. Brian Niccol appointment. https://investor.starbucks.com/news/financial-releases/news-details/2024/Starbucks-names-Brian-Niccol-as-Chairman-and-Chief-Executive-Officer/default.aspx

Competitor and Industry Primary Sources

14. Dutch Bros Q2 FY2026 Form 10-Q, filed 2026-08-06. Shop count, comps, transactions, AUV, Rewards and shop contribution. https://www.sec.gov/Archives/edgar/data/1866581/000186658126000133/bros-20260630.htm
15. Luckin Coffee Q2 FY2026 results, 2026-08-03. Stores, revenue, customers, same-store sales and store-level margin. https://investor.luckincoffee.com/news-releases/news-release-details/luckin-coffee-announces-second-quarter-2026-financial-results
16. McDonald’s Q2 FY2026 Form 10-Q, filed 2026-08-07. Restaurant count, ownership mix, comps and consolidated margin. https://www.sec.gov/Archives/edgar/data/63908/000006390826000073/mcd-20260630.htm
17. U.S. Census NAICS 722515, 2023 employer revenue, https://data.census.gov/table/AIESBASICTIMESERIES.AIES00BASIC?codeset=naics\~722515&g=010XX00US; nonemployer revenue, https://data.census.gov/table/AIESNONEMP2023.AIES00NONEMP?codeset=naics\~722515&g=010XX00US; establishments, https://data.census.gov/profile/722515_-_Snack_and_Nonalcoholic_Beverage_Bars?codeset=naics\~722515; and definition, https://www.census.gov/naics/?details=722515&input=722515&year=2022. Accessed 2026-08-14.
18. National Restaurant Association 2026 State of the Industry, https://restaurant.org/research-and-media/research/research-reports/state-of-the-industry; and July 22 update, https://restaurant.org/research-and-media/research/restaurant-economic-insights/analysis-commentary/restaurants-remain-resilient-despite-challenging-business-conditions/. Accessed 2026-08-14.
19. International Coffee Organization Coffee Market Report, June 2026, https://ico.org/resources/coffee-market-report-statistics-section/; and USDA Brazil Coffee Annual, BR2026-0025, https://apps.fas.usda.gov/newgainapi/api/Report/DownloadReportByFileName?fileName=Coffee+Annual_Brasilia_Brazil_BR2026-0025.pdf. Accessed 2026-08-14.
20. BLS Employment and Earnings, May 2026, https://www.bls.gov/ces/data/employment-and-earnings/2026/table3a_202605.htm; California minimum wage, https://dir.ca.gov/dlse/minimum_wage.htm; fast-food rule, https://www.dir.ca.gov/t8/11000_1.html; and NLRB decisions, https://www.nlrb.gov/cases-decisions/weekly-summaries-decisions/summary-of-nlrb-decisions-for-june-22-26-2026. Accessed 2026-08-14.

Market and Analytical Sources

21. AZI adjusted price CSV, downloaded 2026-08-13. Daily adjusted OHLCV, EMAs, beta and alpha. https://azitrading.com/controls/download-data.php?t=SBUX
22. AZI valuation_index, 2026-08-13, retrieved through the authenticated fundamentals endpoint and archived in the dated research log. Own-history P/E and P/S percentile ranks only; statement arrays and snapshot EV were not used.
23. ROIC.ai, accessed 2026-08-14. TTM statements, enterprise value, ratios and Q3 transcript. Reconciled to SEC; filing values take precedence where ROIC operating income excluded charges. https://www.roic.ai/quote/NASDAQ:SBUX
24. FactorsToday, accessed 2026-08-14. Stock loadings, risk record, factor regime, related stocks and specific volatility. https://www.factorstoday.com/api/stock-loadings/SBUX

Unless otherwise stated, web sources were accessed 2026-08-14. Every material financial number is traceable to a filing or a transparent calculation. Management commentary is treated as hypothesis and tested against filed results and competitor evidence. Third-party computations are dated, basis-labeled and subordinated to primary sources.