Revenue Accelerates to +50% and Customers Cross 100 Million, While Delivery Costs Compress Margins and Management Warns on 2026

Results vs. Consensus

Luckin reports in renminbi with a USD convenience translation; we keep RMB as the currency of record. Coverage on the OTC ADR is thin, but enough estimates exist this quarter to frame a clear scorecard: a large revenue beat, an in-line bottom line, and a margin that missed the level the Street was implicitly carrying.

Year-over-Year Comparison

Sequential Comparison (vs. Q2 2025)

Quality of Beat/Miss

• Revenue: The cleanest part of the print and an unambiguous beat. Growth accelerated to +50.2% from +47.1% in Q2, GMV rose 48% to RMB17.3B, and the company added a record 42M new transacting customers to cross 100M monthly. Same-store sales of +14.4% confirm the acceleration is not purely new-unit-driven.
• Margins: The miss, and it is concentrated entirely in delivery. Delivery expense rose to 18.9% of revenue from 9.1% a year ago, a 980bps swing that more than consumed the combined leverage from materials (down 250bps), rent (down 220bps), marketing (down 90bps) and G&A (down 110bps). The CFO was explicit that improved efficiency was "completely offset" by the delivery line. Notably, delivery cost per order actually fell year over year; the pressure is mix, not unit inefficiency.
• EPS: Roughly in line at about US$0.65 non-GAAP, but GAAP net income of RMB1.28B fell 2.3% year over year, hit by both the operating-margin compression and a higher effective tax rate. This is the first time in the post-restructuring growth run that bottom-line dollars went backwards on a 50% revenue gain.

Segment Performance

The two-engine structure performed as it has all year: the self-operated base absorbs the delivery-margin hit directly, while the asset-light partnership channel grows fastest and dilutes the blended capital intensity. The difference this quarter is that the self-operated margin compression became impossible to miss.

Self-Operated Stores

Self-operated revenue grew 47.7% to RMB11.1B, but store-level operating profit grew only 10.2% to RMB1.94B, dropping the store-level margin to 17.5% from 21.0% in Q2 and roughly 23% a year ago. The arithmetic is the delivery mix: as platform-subsidized demand pulled an outsized share of volume into delivery, the per-cup economics softened even as cup counts and revenue surged. Same-store sales actually improved to +14.4%, helped by an ASP uptick from the delivery mix, so this is not a demand problem; it is a channel-margin problem.

"Same-store sales growth reached 14.4% for this quarter, driven by increased cup sales and ASP, reflecting the shift in volume mix towards delivery. Store-level operating profit grew 10% year over year to RMB1.9 billion, with self-operated store-level operating margin of 17.5%." — An Jing, CFO

Assessment: This is the cost of the land-grab, made explicit. The store-level margin give-back of roughly 600bps year over year is the price Luckin is paying to convert a subsidy window into customers and share. Whether that price was worth paying depends entirely on what the store margin recovers to once the delivery mix normalizes, which management says begins in Q4. This is the single most important line in the model to track for the next two prints.

Partnership Stores

Partnership revenue grew 62.3% to RMB3.8B, again the fastest line, and the partnership store count crossed 10,000 for the first time. This channel carries materials sales, profit share and delivery service fees, and it lets Luckin push into lower-tier geographies without bearing the store capex. Its continued outgrowth of the self-operated base is a structurally favorable mix shift on capital intensity, even though it shares exposure to the same delivery-fee dynamics.

"Revenues from partnership stores increased by 62% year over year to RMB3.8 billion, accounting for 25% of total net revenues... driven by higher material sales, profit sharing from strong partnership store performance, and increased delivery service fees." — An Jing, CFO

Assessment: Partnership is doing exactly what it should: growing faster than the company, broadening the footprint cheaply, and now a quarter of revenue. It is not a margin hedge against the delivery dynamic, but it is a capital-efficiency hedge, and the 10,000-store milestone is a genuine structural marker.

Key KPIs

Key Topics & Management Commentary

Overall Management Tone: Confident on growth and share, materially more candid on the near-term setup than last quarter. Where Q2's framing was "cautiously optimistic," Q3 carried an explicit list of forward headwinds, subsidy tapering, a high comparison base, seasonal slowdown and elevated bean costs, and a direct acknowledgment that same-store sales and margins will face pressure in Q4 and into 2026. The posture on the delivery-margin trade-off was unusually plain: it is temporary, expected, and the price of capturing share now.

1. The Delivery-Margin Trade-Off, Made Explicit

Management stopped hedging on the margin question. The CFO stated directly that the quarter's efficiency gains were "completely offset" by delivery expense as a share of revenue, and the CEO framed the elevated delivery mix as a deliberate, temporary cost of the current industry stage. Importantly, delivery cost per order fell year over year, so the pressure is the volume mix shifting to delivery, not deteriorating unit economics within delivery.

"The notable higher mix of our delivery orders has put some pressure on our margins, fully reflected in the decline of our third-quarter operating margin... the positive impact of our improved operational efficiency was completely offset by the significantly higher delivery expenses as a percentage of total revenues. But we view this as a temporary and expected impact." — Jinyi Guo, Co-founder & CEO

Assessment: Candor is the right move and the per-order efficiency point is genuinely reassuring, but "temporary and expected" is an assertion the next two quarters will test. The bull case requires the store margin to recover toward the low-20s as the mix normalizes; the bear case is that delivery becomes a permanent, higher share of the model and the margin resets lower for good.

2. The Subsidy Reversal and the 2026 Setup

The most consequential new disclosure was forward-looking. Management said the food-delivery platforms have begun "rapidly scaling back their subsidies," that promotions will turn more targeted and ROI-driven in 2026, and that this, combined with the high base the subsidies created this year, will pressure same-store sales next year. It added a second, separate cost headwind: elevated international green-coffee-bean prices with no sign of moderation.

"As platforms have already scaled back their subsidies and will shift towards a more refined approach next year, the industry's overall growth trajectory will differ from this year, and our same-store sales growth next year will also face challenges and pressure." — Jinyi Guo, Co-founder & CEO

Assessment: This is effectively a soft pre-announcement that 2026 same-store sales and possibly margins step down from 2025's subsidy-inflated pace. It is responsible disclosure, and it reframes the investment question from "how fast can Luckin grow" to "how gracefully does it lap a subsidy-distorted year." We take the bean-cost warning seriously as a distinct, supply-driven margin risk layered on top of the delivery mix.

3. The U.S. Relisting Intention

In the Q&A, management confirmed that the CEO had publicly raised the intention to pursue a U.S. main-board relisting (at the Xiamen Entrepreneurs Day conference). The company reiterated its commitment to the U.S. capital market while stressing there is "no specific timeline or schedule," with strategy execution the current priority.

"We remain committed to the U.S. capital market, though we currently have no specific timeline or schedule for U.S. listing on the mainboard. Our top priority at the current stage remains focusing on our strategy execution and business development." — Jinyi Guo, Co-founder & CEO

Assessment: This is a real, if unscheduled, re-rating catalyst. LKNCY trades at an OTC discount tied to the 2020 fraud and the subsequent pink-sheet listing; a successful main-board return would broaden the eligible investor base (index inclusion, institutional mandates that bar OTC names) and could compress that discount. We do not capitalize it today given the absence of a timeline, but it shifts the governance bear point from a one-way overhang to a two-way option.

4. Below-Store Operating Leverage Still Working

Lost in the delivery-margin headline is that every other cost line continued to leverage down: materials to 36.2% of revenue from 38.7%, store rent to 20.2% from 22.4%, marketing to 4.9% from 5.8%, and G&A to 5.2% from 6.3%. The structural efficiency thesis is intact; it is simply being masked by one line.

"Cost of materials as a percentage of total net revenues decreased to 36% from 39%... mainly due to our enhanced disciplined supply-chain advantages." — An Jing, CFO

Assessment: This is the coiled spring in the model. If and when the delivery mix normalizes, this accumulated below-store leverage is what should drive the margin recovery. The quarter is bearish on reported margin but quietly confirms the underlying cost engine is intact.

5. The Store and Customer Machine

Luckin added a record 3,008 net new stores to reach 29,214, with partnership crossing 10,000 units, and added 42M new transacting customers to push monthly customers past 100M for the first time. International reached 118 stores; management called Singapore an increasingly mature, replicable template and described U.S. performance as in line with expectations.

"We added over 42 million new transacting customers in the third quarter and achieved an average of over 110 million monthly transacting customers, both record highs. By quarter end, our cumulative transacting customer base surpassed 420 million." — Jinyi Guo, Co-founder & CEO

Assessment: The demand-side metrics are unambiguously strong and are the foundation of the "we bought durable share" argument. The open question remains retention as subsidies fade; the record acquisition is necessary but not sufficient evidence for that thesis.

6. Cash Generation Accelerating

Nine-month operating cash flow more than doubled to RMB5.53B from RMB2.60B a year ago, and the total cash position reached RMB9.35B. The store expansion remains entirely self-funded.

Assessment: The cash build continues to derisk the balance sheet and, with the relisting intent now public, raises the question of capital allocation more pointedly. A company generating this much cash with a rebuilt balance sheet and a stated capital-markets ambition is accumulating optionality on returns it has not yet discussed.

Guidance & Outlook

Luckin issues no formal numeric guidance, but Q3's qualitative outlook was the most directional in recent memory, and it points down on the near-term margin and same-store metrics even as unit growth stays strong.

"These factors will introduce new dynamics and create headwinds for the industry and pose challenges to our fourth quarter or even next year's business development... we will strive to maintain a healthy and sustainable profit profile." — Jinyi Guo, Co-founder & CEO

Implied trajectory: Expect revenue growth to decelerate from the +50% Q3 pace as the subsidy-inflated base gets harder, same-store sales growth to moderate in Q4 and through 2026, and reported margins to remain under pressure from both the delivery mix and elevated bean costs before any recovery from mix normalization.

Guidance style: A clear step toward conservatism. Management spent Q2 under-committing on margin while letting growth surprise; in Q3 it pre-warned on both growth and margin for the first time this cycle. When a management team that habitually sandbags starts flagging headwinds out loud, the signal is worth weighting.

Analyst Q&A Highlights

Three questions were taken, and all went to the heart of the debate: the durability of the subsidy-driven growth and its margin cost, how management balances scale against profitability, and the capital-markets path. The exchanges were more substantive than Q2's, and the relisting question in particular drew out a disclosure that was not in the prepared remarks.

Subsidy Impact: Seasonal Pause or Structural High Base

The first question pressed directly on whether the Q4 subsidy fade is seasonal or structural, and whether 2026 faces a high-base problem. Management answered with a framework: coffee is fundamentally a pickup, location-based product; delivery is a supplemental channel that is structurally lower-margin and will recede over time, so 2026 same-store sales will face pressure as the subsidy base laps.

Q: "The delivery subsidy definitely helps our revenue grow very strongly... but it also has a negative impact on our margin. Going to next year, should we worry about the high base effect?"
— Ethan Wang, CLSA

A: "Coffee is inherently a location-based and store-driven consumer product... pickup will remain the primary consumption format over the long term, but delivery will serve more as a supplemental channel... As platforms have already scaled back their subsidies and will shift towards a more refined approach next year... our same-store sales growth next year will also face challenges and pressure."
— Jinyi Guo, Co-founder & CEO

Assessment: Management both validated the bear's high-base concern and gave the bull a structural counter, that the natural return to pickup restores margin over time and that a scale player with owned fulfillment stays the platforms' preferred partner. It is a coherent answer that nonetheless concedes 2026 same-store sales decelerate. That concession is the takeaway.

Balancing Scale, Same-Store Sales and Profit

The second exchange asked how management weighs the competing objectives of scale, same-store growth and profitability through an unstable external environment. The answer was an unambiguous prioritization: growth and market share come first at this stage of China's coffee market, store quality is protected as the top operational priority post-2020, and margins are managed for "healthy and sustainable" rather than maximized.

Q: "How will we balance different targets, including scale, same-store sales growth and profit?"
— Sijie Lin, CICC

A: "Business growth and market-share expansion remain our strategic priorities at this stage... even if same-store performance shows some fluctuations, the overall trajectory remains within our expectations. At the same time, we will strive to maintain healthy and sustainable profit levels and remain confident in our long-term profitability potential."
— Jinyi Guo, Co-founder & CEO

Assessment: This is the clearest statement yet of the capital-allocation philosophy: Luckin will trade near-term margin for share while the category is consolidating. For an investor, it sets expectations correctly, the company is not optimizing for the next two quarters' margin, and it should be modeled accordingly.

U.S. Relisting Status

The third question asked for an update on the CEO's publicly floated intention to pursue a U.S. main-board relisting. Management confirmed the commitment to the U.S. capital market but explicitly declined to attach any timeline, framing business execution as the current priority.

Q: "Dr. Guo mentioned the company's intention to pursue a relisting on a major U.S. exchange. Could management share an update on the current status of this initiative?"
— Huayi Li, Securities

A: "We remain committed to the U.S. capital market, though we currently have no specific timeline or schedule for U.S. listing on the mainboard. Our top priority at the current stage remains focusing on our strategy execution and business development."
— Jinyi Guo, Co-founder & CEO

Assessment: The intention is now on the record from management, not just a conference soundbite, which is meaningful, but the "no timeline" qualifier keeps it firmly in the optionality bucket. It is a catalyst to monitor, not to underwrite.

What They're NOT Saying

1. A quantified store-margin recovery path: Management says the delivery margin drag is "temporary" but will not say what self-operated store margin normalizes to, or when. The entire bull case hinges on a recovery that is asserted, not sized.
2. Retention data on subsidy-acquired customers: A second straight quarter of record acquisition with no cohort retention or repeat-rate disclosure. As subsidies fade, this is the number that determines whether the share was bought or rented.
3. 2026 revenue or same-store framework: Management warned on 2026 but offered no ranges, leaving the magnitude of the deceleration entirely to the Street's imagination.
4. Capital allocation despite RMB9.35B cash and a relisting ambition: Still no discussion of buybacks or dividends. With cash building fast and a capital-markets initiative now public, the silence on returns is increasingly conspicuous.
5. Competitive pricing detail: No engagement on ASP ex-delivery or on the price war with low-cost challengers; "complex environment" continues to do a lot of work.

Market Reaction

• Pre-print setup: LKNCY closed at $37.80 on November 14, up roughly 47% year to date but down about 5.7% over the trailing 30 days, well off its 52-week closing high of $43.03. The recent pullback meant the stock entered the print with some of the margin concern already discounted.
• Print-day session (November 17, BMO): Shares gapped down 2.4% to open at $36.90 as the market first read the margin compression, then reversed hard through the session to close at $40.00, up 5.8% on the day and at the top of a wide $36.72 to $40.22 range, on volume of 5.4M versus a 2.4M 30-day average (2.3x).
• Index context: The S&P 500 fell 0.9% on the session, so the rally was decisively stock-specific.

The intraday reversal from -2.4% to +5.8% is the informative part. The market opened focused on the margin miss, then re-weighted toward the 50% revenue acceleration, the 100M-customer milestone and the newly explicit relisting optionality, deciding on balance that the share-grab and the re-rating option outweigh a margin compression that was, by November, broadly expected. The recent 30-day de-rate into the print also left room for the relief rally. It is a constructive reaction, but it is a reaction to optionality and growth, not to profitability.

Street Perspective

Debate: Is the Margin Compression Temporary or the New Normal?

Bull view: The optimists take management at its word that the delivery mix is a transient feature of a subsidy-distorted year, point to the per-order efficiency gains and the intact below-store leverage, and argue the store margin recovers toward the low-20s as the channel normalizes in 2026.

Bear view: The skeptics argue delivery is now structurally embedded in Chinese coffee consumption, that 18.9% of revenue going to delivery is unlikely to fully reverse, and that elevated bean costs add a second, supply-driven leg to the margin reset.

Our take: The truth is probably in between: the store margin recovers partially but not all the way, settling above Q3's trough but below the 21% of Q2, because some of the delivery shift is permanent. That is a perfectly investable outcome, but it argues against capitalizing either the 2025 growth rate or a full margin recovery. Hold.

Debate: Does the Relisting Change the Investment Case?

Bull view: The bulls see a U.S. main-board return as a step-change re-rating catalyst that would unlock the OTC discount, broaden the investor base and potentially enable index inclusion, on a company whose fundamentals already justify a premium growth multiple.

Bear view: The bears note there is no timeline, that a relisting would invite far more rigorous scrutiny of a company with a fraud history, and that the discount exists for governance reasons that a listing venue alone does not cure.

Our take: It is a real option with asymmetric, hard-to-time payoff. We treat it as a reason not to be short and a reason the governance bear point is now two-sided, but not yet a reason to pay up. If a concrete timeline emerges, the calculus changes.

Debate: Has the Growth Already Peaked?

Bull view: Bulls argue 50% growth with a 100M-customer base and a multi-year store runway means the deceleration is from an exceptional level to merely a strong one, and that share gains during the subsidy war are permanent.

Bear view: Bears contend Q3 is the high-water mark, that the 2026 same-store warning plus a hard base means estimate revisions head lower from here, and that a decelerating-and-margin-pressured story does not support the current multiple.

Our take: We side with the view that growth decelerates meaningfully into 2026, which management itself telegraphed. That does not make Luckin a short, the franchise is too strong and the cash too real, but it does cap the upside until the deceleration is in numbers and the margin trough is visible. Maintaining Hold.

Model Update Needed

Valuation impact: We hold our balanced stance. The model now carries a sharper 2026 deceleration in same-store sales and a margin that troughs in Q3/Q4 2025 and recovers only partially as the delivery mix normalizes. The relisting is a discrete upside option we do not capitalize. A move to Outperform would require either visible margin-trough confirmation with a recovery path or a concrete relisting timeline; a move to Underperform would require evidence the store margin reset is permanent and the subsidy-acquired customers churn out.

Thesis Scorecard Post-Earnings

Grading the standing thesis established at the Q2 initiation.

Overall: The thesis is playing out close to script. The bull pillars on scale, runway and cash are all reconfirmed, a new relisting option has appeared, and the central bear point (delivery margin) has moved from "emerging" to "materializing" exactly as anticipated. Net, the thesis is unchanged: a great franchise priced for a lot, with a known margin reset underway and a known 2026 deceleration ahead.

Action: Maintain Hold. The quality and the optionality keep us from turning negative; the materializing margin compression and the management-flagged 2026 step-down keep us from turning positive. We want the margin trough confirmed or a relisting timeline before changing the call.