On August 19, 2026, Unitree Technology made its debut on the STAR Market, pricing its IPO at 150.80 yuan per share, which translated to a market capitalization of roughly 61 billion yuan. The stock surged dramatically during its first trading day, hitting an intraday peak of 1,100 yuan—a market value of about 444.9 billion yuan—before closing at a still-lofty 341.8 billion yuan. However, the stock has since retreated sharply: it closed at 687 yuan on August 20, 672.41 yuan on August 21, and 603.08 yuan by August 24, bringing its total market capitalization to approximately 243.9 billion yuan. Over just four trading sessions, the share price has fallen more than 45% from its intraday high, erasing over 200 billion yuan in market value, yet it remains roughly 300% above its IPO price.
Short-term share prices are inevitably swayed by scarce float, market sentiment, and trading volume, but the valuation question runs much deeper: why is a robotics company with 2025 revenue of approximately 1.7 billion yuan and first-half 2026 revenue of about 1.15 billion yuan still priced at roughly 240 billion yuan even after a rapid correction? Zhang Feida, Associate Professor of Accounting at CEIBS, applies the Ohlson Residual Income Model to analyze Unitree Technology through four valuation lenses—ROE, sustainability, growth, and risk assessment—in an attempt to answer this question.
In my earlier article, "Is SpaceX Worth 2.5 Trillion Dollars?", I introduced the Ohlson Residual Income Model: enterprise value depends not only on today's book value (the physical body), but far more critically on whether the company can continue to generate residual income above its cost of capital in the future (the soul). I won't repeat the formula here; interested readers can refer back to that piece. Comparing these two newly listed companies, SpaceX needs to prove when massive capital expenditures and current losses will convert into sustainable cash flows. Unitree Technology faces a different challenge: it is already profitable, but still small in scale; with fresh capital from its listing, can it extend its stage-by-stage high returns into long-term high returns?
Technological leadership is an industry judgment; company excellence is an operational judgment; a cheap stock is a price judgment. These three can sometimes hold simultaneously, and sometimes diverge completely.
What Has Unitree Already Proven, and What Has the Market Bought in Advance?
Unitree Technology is not a robotics company that merely relies on viral videos to grab attention without generating revenue. In 2025, the company achieved operating revenue of approximately 1.699 billion yuan, up 332.6% year-on-year; net profit attributable to the parent company was about 278 million yuan, while net profit attributable to the parent company excluding non-recurring gains and losses was approximately 591 million yuan, with net operating cash flow of about 670 million yuan. The fact that non-GAAP net profit exceeded GAAP net profit in 2025 was mainly due to one-time share-based payment expenses being recorded as non-recurring items. For a general-purpose robotics company still in a phase of rapid technological iteration, it is uncommon to see scale growth, profitability, and positive cash flow emerge simultaneously.
What is even more noteworthy is that the first-half 2026 data disclosed in the listing announcement shows revenue of 1.152 billion yuan, up 48.54% year-on-year; net profit attributable to the parent company of 274 million yuan, but non-GAAP net profit of 244 million yuan, down 19.34% year-on-year, and net operating cash flow of 232 million yuan, down 32.53% year-on-year. Growth is still intact, but profit quality and cash conversion are already beginning to feel the pressure of rising R&D and sales expenditures. Unitree's revenue growth is also not solely dependent on high-priced scientific research prototypes. Through in-house development of core components, motion control, and supply chain integration, the company has gradually turned quadruped robots and humanoid robots into products that can be sold in volume. In 2025, the company shipped over 5,500 humanoid robots. It has already completed the most difficult step for many robotics startups: transitioning from being able to build to being able to sell.
But the price the capital market is offering is not merely rewarding this 2025 report card. The market capitalization at Unitree's IPO price was about 61 billion yuan, implying a price-to-earnings ratio of roughly 103 times and a price-to-sales ratio of about 36 times based on 2025 non-GAAP net profit; by the close on August 24, the market capitalization was approximately 243.9 billion yuan, implying a P/E ratio of about 413 times and a P/S ratio of about 144 times on the same basis. This means the market is not just buying the 600 million yuan in non-GAAP profit Unitree has already achieved today, but rather the expectation of much larger-scale commercialization: robots entering factories, warehouses, commercial services, and even homes; Unitree evolving from a hardware manufacturer into an embodied intelligence platform; and today's shipments of several thousand units eventually expanding to deployment scales of hundreds of thousands or even millions of units. This article aims to answer: within the roughly 240 billion yuan price tag, how much is reasonable expectation, and how much is anticipation that will only be realized if the company maintains high-quality execution over the long term?
Four Dimensions to Scrutinize Unitree Technology's Future Value
ROE: Can Pre-IPO High Returns Be Rebuilt After the IPO?
The Ohlson model places ROE at the core of value creation because only when ROE consistently exceeds shareholders' required rate of return can a company continuously generate residual income. And only when new investments continue to yield returns above the cost of capital does growth truly add value. For Unitree Technology, ROE should not be viewed as just a single impressive ratio; it is better analyzed using DuPont analysis, breaking it down into three components: profitability (product strength), operational capability, and capital deployment capability. The formula ROE = Net Profit Margin × Asset Turnover × Equity Multiplier, when applied to a robotics company, can be understood respectively as: can products generate sufficient profit, can assets turn over efficiently, and can management effectively utilize leverage—corresponding to profitability (product strength), operational capability, and capital deployment capability.
Unitree's profitability first comes from product strength. In 2025, the company's revenue was approximately 1.699 billion yuan, with non-GAAP net profit attributable to the parent of about 591 million yuan and net operating cash flow of approximately 670 million yuan. More importantly, through in-house core component development, motion control, whole-machine design, and cost engineering, the company has transformed robots from high-priced research prototypes into products that can be sold in volume. For a robotics company, product strength does not mean selling at the highest possible price; rather, it lies in whether performance improvement and cost reduction can happen simultaneously. Price reductions help expand installed base, but to avoid compressing gross margins, unit costs need to fall even faster, or revenue from software, services, and solutions needs to gradually make up for the gross margin pressure from hardware price cuts. Key metrics to track going forward include gross margin, unit manufacturing costs, R&D conversion efficiency, and whether new products can maintain a competitive price-performance ratio.
The actual first-half 2026 data has put this stress test firmly on the table: revenue still grew 48.54% year-on-year, but this is a clear deceleration from the 332.6% hyper-growth of 2025; over the same period, non-GAAP net profit fell 19.34% year-on-year, and net operating cash flow declined 32.53%. The company explains that the primary reasons are expansion of R&D teams, new product development, and increased sales investment. For a high-growth company that has just gone public, the most critical next step is not just continuing to grow, but proving that increased R&D and sales expenditures can be converted back into profit margins and cash flow.
Before its IPO, Unitree benefited from a relatively asset-light structure, rapid product iteration, and solid operating cash flow. The robotics industry changes quickly, and operational capability is not just about factory production efficiency; it also depends on whether the entire chain—from R&D project initiation, product finalization, supply chain procurement, production delivery, to customer payment collection—can maintain a sufficiently short cycle. After the IPO, this advantage will face new tests. The raised funds will be directed toward robot models, core R&D, new product development, and manufacturing base construction, which could all increase fixed assets, inventory, and R&D spending. If capacity construction outpaces real demand growth, or if new scenario deployments rely heavily on manual customization, asset turnover may decline. Going forward, key metrics to watch include inventory turnover, accounts receivable, operating cash flow conversion, capacity utilization, and deployment cycle times for individual scenarios.
Unitree's high pre-IPO returns were not primarily amplified by high financial leverage. The company previously had a low debt-to-asset ratio and minimal short-term borrowings; as of June 30, 2026, net assets attributable to the parent were approximately 2.88 billion yuan, while net IPO proceeds were about 5.917 billion yuan—roughly 2.05 times its then-current net assets. In other words, the new capital will raise the equity "denominator" in one large step without simultaneously elevating net profit to the same degree. Therefore, a staged decline in ROE after listing is not just unsurprising—it is almost a mathematical certainty. This is not inherently a bad thing; the key lies in what returns the new capital ultimately generates. Capital deployment capability is not just about whether to use leverage; it is more critically about how to prioritize R&D, manufacturing bases, new products, and ecosystem investments—which projects should be scaled up, and which should be cut in a timely manner. Only when the marginal returns on new capital consistently exceed the cost of capital does financing create value, rather than merely inflating the balance sheet.
Financial conclusion: Unitree's next-phase ROE will need to be rebuilt more through product margins, asset turnover, and capital allocation discipline, rather than relying on leverage alone. The IPO is not the end of high ROE; it is a reset of the capital denominator.
Sustainability: Can the Moat Transition from Performing to Working?
A single year of high ROE cannot support a high valuation. The Ohlson model is more concerned with how long this excess return can persist. Sustainability, in operational terms, is the moat question: can today's technological advantages continue to convert into profits and cash flow amid competition, price cuts, and industry iteration? Unitree's most obvious moat lies in motion control and full-stack engineering capability. From quadruped robots to humanoid robots, the company has built strong synergistic capabilities across motors, reducers, joints, motion control, reinforcement learning, whole-machine design, and cost engineering. The value of these capabilities is not just reflected in impressive stunts on video, but in the ability to turn new movements, new models, and core components into products quickly and at lower cost.
Investors need to distinguish between two types of value. The first is demonstration value: the ability to run, jump, flip, and box, proving motion control and whole-machine engineering capabilities. The second is production value: working continuously for hours or even thousands of hours in unfamiliar environments, maintaining stable task success rates, keeping maintenance costs manageable, recovering quickly from failures, and delivering a lower total cost to customers than manual labor or traditional automation. The former is more likely to generate traffic and early orders; the latter is more likely to generate repeat purchases, long-term contracts, and scaled capital expenditures.
Two pieces of news from August 20 perfectly illustrate the difference between "demonstration value" and "production value." On one hand, the day after its listing, Unitree launched the R1 bionic 7-axis dexterous robotic arm at a starting price of 9,900 yuan, targeting scientific research, education, material sorting, assembly, and service robot exploration—demonstrating that hardware productization and cost reduction are still advancing rapidly. On the other hand, Wang Xingxing admitted at the 2026 World Robot Conference that the biggest bottleneck to robots entering factories and homes on a large scale remains insufficient efficiency and embodied intelligence generalization—new tasks often require retraining. The former shows that the ability to "build it and sell it cheaper" continues to strengthen; the latter reminds investors that "stably performing labor for humans" remains an uncrossed threshold.
A commercial moat requires going one step further than a technical moat. Factory customers care more about whether a robot can work steadily for eight consecutive hours, whether deployment takes weeks or months, who maintains it when it fails, whether it needs re-customization for a new scenario, and ultimately how much cost it saves for the customer. When reliability, delivery, after-sales, data, and customer economics collectively form a barrier, technological leadership is more likely to translate into sustainable ROE. Technical barriers themselves also require continuous maintenance. The prospectus materials note that the company's relatively small number of patents may, to some extent, increase the difficulty of protecting core technology and preventing imitation. For a robotics industry with rapid iteration, the moat is less likely to be a single static patent and more likely to be the dynamic capability formed by continuous R&D and iteration, cost engineering, supply chain, real-world scenario data, and organizational learning.
Financial conclusion: What is easiest to price on listing day is the performance; what is hardest to deliver over the next decade is the labor. Unitree's sustainability depends on whether it can convert its motion control advantage into stable labor capability, generating repeat purchases and sustained cash flow.
Growth: Is Future Growth About Selling More Robots or Selling More Productivity?
In the residual income framework, growth itself does not inherently create value. Only when new investments continue to generate returns above the cost of capital does growth amplify what I referred to earlier as "soul value." For Unitree, what matters is not ultimately how large the robotics industry becomes, but whether it can maintain high returns during expansion. Unitree's current revenue core remains hardware. The hardware business can grow very large, but the valuation ceiling for hardware companies is typically constrained by price declines, manufacturing costs, inventory, after-sales, and competition. How high the long-term valuation can go depends more on whether Unitree can complete a three-stage upgrade.
The first stage is hardware productization: turning robots from research prototypes into standard products, expanding the market through volume, cost reduction, and rapid iteration. Unitree has already crossed this stage successfully. The second stage is scenario solution-ization: customers are no longer buying just a robot, but a combination of the robot body, end effectors, model training, on-site deployment, software updates, and maintenance services. The revenue structure shifts from one-time delivery toward a mix of hardware, projects, and services. The third stage is labor platform-ization: different developers and industry customers can develop tasks on the same set of bodies, interfaces, models, and data ecosystems; adding new scenarios no longer requires starting from scratch each time; more installed base generates more real-world data, which is then used to improve model generalization and task success rates. If this positive cycle forms, Unitree's growth would no longer just be about how many units it sells this year, but would become: more installed base → more data → stronger models → higher success rates → more scenarios and repeat purchases.
If entering each new scenario requires substantial customization, field engineers, and after-sales personnel, then the larger the scale, the more complex it becomes—revenue may grow, but profit margins, asset turnover, and cash flow may not improve in tandem. This model can boost revenue, but it may not create more residual income. Metrics to watch in the future may no longer be limited to shipment volume, but also include whether the same customer makes repeat purchases, whether the proportion of software and services revenue is rising, whether deployment cycles per scenario are shortening, whether actual robot operating hours are increasing, and how much total robot cost customers must bear to save one yuan of labor cost.
Financial conclusion: What determines the quality of Unitree's growth is not how many robots are sold, but whether each delivery can create repeatable, scalable, continuously chargeable productivity without diluting capital returns.
Risk Assessment: Why Should Some Future Returns Be Discounted at 80% While Others Only at 50%?
Risk in the Ohlson model ultimately feeds into the shareholders' required rate of return, i.e., the discount rate. The more uncertain future returns are, the higher the return investors demand (risk compensation), and the lower the valuation they are willing to offer today. For a founder-led hard-tech company like Unitree, risk assessment can be viewed from at least three angles: management quality, corporate governance, and ESG.
Regarding management quality: Wang Xingxing simultaneously serves as Chairman, General Manager, and Chief Technology Officer, is also the company's core technical figure, and has over 15 years of robotics R&D experience. The trinity of founder, technical leader, and operational head has been an important organizational asset for Unitree's rapid decision-making and continuous iteration. But after listing, the tasks facing management will change. The company no longer only needs to build and sell robots; it must also manage several billion yuan in new capital, manufacturing bases, talent pipelines, investor expectations, and more complex global operational risks. For investors, management quality is better assessed through long-term commitment willingness, capital allocation discipline, the ability to cut losses on failing projects, whether information disclosure is sufficient and candid, and whether a second-tier leadership team independent of the sole founder can be established.
Regarding corporate governance: Unitree has adopted a special voting rights arrangement: Class A special voting shares held by Wang Xingxing carry 10 votes per share, while ordinary Class B shares carry one vote each. At the same time, he holds the roles of Chairman, General Manager, and CTO. The advantage of this structure is a shorter decision-making chain, allowing the founder to adhere to a long-term technology roadmap without being swayed by short-term capital market sentiment. But this also means that independent board oversight, related-party transaction constraints, major capital allocation decisions, minority shareholder protection, and succession mechanisms become more important. The company has established mid-to-long-term incentives through an employee stock ownership platform, which helps retain core talent. What is more worth observing in the future is whether, as the organization grows, decision-making can rely more on institutional processes and whether major decisions have sufficient professional checks and balances.
Regarding ESG: As robots enter society, externalities will shift from soft issues to hard costs. From a traditional environmental risk perspective, Unitree primarily uses assembly-based production and does not belong to a heavily polluting industry; the prospectus materials disclose that there were no environmental violations or workplace safety accidents during the reporting period. This suggests that traditional manufacturing-related environmental and safety risks are currently relatively controllable. However, for general-purpose robots, the more important future ESG issues may not be about smokestacks, but about real-world usage scenarios: physical safety when robots work alongside humans, data security and privacy issues from cameras and sensors, cybersecurity, supply chain and overseas compliance, and the impact of automation on employment and liability boundaries. Once robots enter factories, commercial spaces, and even homes, these issues can translate into recalls, insurance, litigation, compliance, reputation, and barriers to customer adoption.
Overseas compliance risk has also shifted from an abstract "future possibility" to a real policy variable. In July 2026, the US FCC included newly manufactured advanced humanoid and quadruped robots from overseas in stricter equipment certification restrictions. Unitree has disclosed that its main models currently on sale and already FCC-certified are not affected for now, but future new models may face the risk of being unable to sell in the US. The company's overseas revenue has accounted for more than 40% of total revenue in each of the past three years, with US market revenue accounting for approximately 13.30% of total revenue in 2025. Therefore, overseas regulatory and geopolitical risks can directly affect the growth trajectory and ultimately factor into the risk premium demanded by valuation.
Financial conclusion: Wang Xingxing may be both Unitree's most important organizational asset and a key-person risk that needs to be incorporated into the discount rate; concentrated control enhances execution efficiency but also requires stronger institutional checks and balances. Risk assessment is not about labeling the company; it is about determining at what discount future returns should be valued.
How Much Future Has the Approximate 240 Billion Yuan Already Pre-Paid For?
The four dimensions above answer how good the company's intrinsic quality is; the price test answers whether the stock is cheap right now and whether there is sufficient margin of safety. The two must be considered separately. A good company can be bought at a high price, and a bad company can be cheap, but value investing is more concerned with this question: how much future has the current price already bought in advance?
Based on the approximate 243.9 billion yuan market capitalization on August 24, Unitree corresponds to roughly 413 times 2025 non-GAAP net profit and about 144 times sales. Compared to the 277.9 billion yuan on August 20, the market has already proactively adjusted expectations downward, but such a valuation still implies that the company must not only maintain high growth, but also see ROE recover after IPO dilution, sustain its moat, scale its growth in a replicable way, while avoiding major mistakes in management and governance. We can do a scenario calculation solely for understanding how high market expectations are. Assuming investors require a 12% annualized return and the market is still willing to give Unitree a 30x P/E ratio in ten years: at the approximately 61 billion yuan IPO market capitalization, net profit ten years later would need to reach about 6.3 billion yuan, implying an average annual growth rate of roughly 26.7% from the 2025 non-GAAP profit of approximately 591 million yuan. At the approximately 243.9 billion yuan market capitalization as of August 24, net profit ten years later would need to reach about 25.3 billion yuan, implying an average annual growth rate of about 45.6% over the next decade.
This is not a forecast of Unitree's profit, nor is it a target price. It simply converts price into an operational task: the IPO price corresponds to a requirement to maintain a decent operational performance over the long term, while the August 24 market price—even after a significant pullback from the first-day high—still requires Unitree to sustain high ROE, preserve its moat, maintain rapid growth, and keep risk at low levels for a very long time. It is also important to note that Unitree's freely tradable shares at the time of listing accounted for only about 7.44% of the total post-issuance share capital. In a situation of high market attention and limited tradeable supply, prices in the first few trading days will be simultaneously influenced by fundamentals, scarcity, and trading sentiment. Short-term market capitalization can serve as an observation of market enthusiasm, but it should not be directly regarded as a stable long-term fair value.
The high valuation risk does not negate that Unitree is an excellent company. It means that the higher the price, the fewer mistakes investors can tolerate. A one-year product delay, a few percentage points lower gross margin, declining asset turnover, capital allocation errors, or rising governance risks—all have amplified effects on long-term value.
Financial conclusion: Four-dimensional analysis helps determine how far the company can go; the price determines how much margin of safety investors have left. The question of the approximately 240 billion yuan market capitalization is not whether Unitree is excellent enough, but whether the market has already priced too much future excellence into today's price.
My Assessment: A Good Company and a Good Price Should Be Answered Separately
Is Unitree Technology a good company? From an ROE perspective, Unitree has demonstrated initial capabilities in productization and capital efficiency, but it will need to rebuild high returns after the IPO. From a sustainability perspective, its motion control advantage is already strong, but the commercial moat still needs to be validated through stable work, repeat purchases, and cash flow. From a growth perspective, the long-term ceiling lies in moving from hardware to scenarios, and then to a labor platform. From a risk perspective, founder-led management is both a source of efficiency and requires more mature governance and ESG systems to lower the discount rate.
Is Unitree in a good industry? Most likely yes. Demographic trends, rising labor costs, replacement of dangerous jobs, advances in AI models, and declining robotics hardware costs collectively support long-term demand. Reuters, citing industry data, reported that humanoid robot deliveries in China exceeded 40,000 units in the first half of 2026, accounting for the vast majority globally; but at the same World Robot Conference, Wang Xingxing also clearly stated that the current efficiency and generalization capabilities of robots in real-world scenarios remain below what is needed for large-scale adoption. In other words, rapid industry growth and not-yet-fully-mature commercialization can coexist. A large industry does not guarantee every company succeeds, nor does it mean the commercialization path will be as linear as the market imagines.
Is Unitree currently a good price? This cannot be answered by industry enthusiasm alone. The approximately 240 billion yuan market capitalization is no longer just rewarding what Unitree has done right in the past; it is also paying in advance for what it might achieve over the next decade. Compared to the extreme first-day high, the price has clearly cooled, but the current valuation still reflects a very high degree of recognition, leaving an extremely thin margin of safety for execution missteps. I am more inclined to view Unitree as an excellent hard-tech company with genuine technology, genuine products, genuine revenue, and genuine cash flow. It is at a critical stage of converting product strength into high-quality capital returns, but the current market price already demands a great deal from the future.
The previous article on SpaceX discussed how to anchor dreams to cash flow. This article on Unitree centers on how to anchor product strength to ROE, how to turn movement advantages into a sustainable moat, how to turn shipment volumes into replicable growth, and how to gradually convert founder-driven speed into governance capabilities that are risk-controlled for a listed company. What is more worth tracking in the future is not daily stock price fluctuations, but the following five things:
ROE: After the IPO, can gross margin, asset turnover, and capital allocation collectively drive ROE to recover after dilution? Sustainability: Can humanoid robots consistently enter industrial and commercial production scenarios—beyond research, demonstration, and data collection—and achieve stable operation and repeat purchases? Growth: Can software, services, solutions, and a developer ecosystem increase revenue repeatability, shifting growth from selling hardware to selling productivity? Risk assessment: Will management's capital allocation, governance checks under special voting rights, product safety, and ESG risks raise the discount rate on future returns? Valuation: Can profit growth catch up quickly enough to the high expectations already embedded in the current market capitalization?
Understanding these dimensions, and then placing them back into the price, provides a more complete understanding of Unitree Technology. Investing in Unitree is not buying a single backflip, nor just the concept of embodied intelligence; it is buying whether this company can consistently create high ROE through product strength, consolidate technological advantages into a sustainable moat, convert scenario expansion into high-quality growth, and—with management, governance, and ESG risks under control—ultimately deliver future residual income to shareholders' accounts.
Professor Zhang Feida is an Associate Professor of Accounting at CEIBS. Prior to joining CEIBS, he served as tenured Associate Professor at the University of Queensland Business School, tenured Associate Professor and MBA Centre Director at Murdoch University Business School, Visiting Professor at Xiamen University, and Assistant Professor at Sun Yat-sen University. He received his PhD in Accounting from Hong Kong Baptist University and his Master's in Accounting from Xiamen University. His research focuses on financial management and decision-making, corporate governance, capital markets, corporate social responsibility, and corporate finance. His extensive research has been published or is forthcoming in top international academic journals, including the Financial Times' top 50 global business journals such as the Journal of International Business Studies (two papers), the Journal of Financial and Quantitative Analysis, the Journal of Business Ethics (three papers), the Journal of Corporate Finance (three papers), the European Accounting Review (two papers), as well as leading domestic journals such as Management World and Accounting Research.
Note: This analysis does not use the regulatory-mandated lower of GAAP/non-GAAP statutory diluted P/E ratio; it consistently uses the post-issuance diluted P/E ratio based on non-GAAP net profit. The prospectus discloses that the diluted P/E ratio calculated on the lower of GAAP/non-GAAP basis is 219.23 times. Data sources and basis: This article relies on Unitree Technology's prospectus, listing announcement (including unaudited first-half 2026 financial data), Shanghai Stock Exchange trading information, issuance review materials, public information from the 2026 World Robot Conference, and the company's disclosure of the impact of US FCC policy; valuation and scenario growth rates are the author's estimates. The 12% annualized return and 30x P/E ratio after ten years are solely reverse-valuation assumptions. Market and financial data are as of the close of August 24, 2026, and do not constitute earnings forecasts, target prices, or securities investment advice. Cover and in-text images were AI-generated.