Image source @ Visual China
Text | The author | Lu Cun, typography | Yi Xin, Executive Producer | Yoda
The "AI Four Little Dragons" cloud has made new progress from the listing of science and technology.
On the evening of April 6, the CSRC has agreed to yuncong technology group co., ltd. (hereinafter referred to as "yuncong technology") science and technology innovation board registration application, just one step away from landing on the science and technology innovation board. This means that after experiencing questioning by the Shanghai Stock Exchange, the application being suspended and reviewed, and a long wait, Yuncong Technology will become the first member of the "AI Four Little Dragons" to land on A shares.
However, the barbaric growth period of the AI industry is over. On the back of the rapid growth of revenue, there is a huge loss quagmire. According to the prospectus, from 2019 to 2021, Yuncong's technology revenue was 807 million, 755 million and 1.076 billion yuan, respectively, of which 2021 increased by 42.6% year-on-year; the net profit attributable to the mother was -640 million, -813 million and -632 million yuan, respectively, and the total loss in three years was as high as 2.084 billion yuan.
Less than 7 years after its establishment, the cloud has shown fatigue from technology. Under the huge loss, there is a lot of resistance to turning losses into profits; the hardware business is getting narrower and narrower, dragging down the overall gross profit margin; too dependent on large customers, the ability to continue to operate is to be examined; research and development is more and more involuted, and the performance is overwhelmed; the business is highly overlapping, and the moat is difficult to say consolidated. How did the cloud successfully complete the road to IPO from technology?
01 Revenue is difficult to compare with giants, profitability to be examined
Cloud From Technology is the last unicorn in the "AI Four Little Dragons".
In 2015, Cloud was founded from technology, and Zhou Xi, who was 34 years old at the time, officially transformed from a scientist to an entrepreneur. In 6 years, Yun has raised more than 3 billion yuan from science and technology; many state-owned and government funds such as China Guoxin, Guangzhou Industrial Investment Fund, and Bohai Industrial Investment Fund are its shareholders, which is a veritable "AI national team".
In the early stage of development, Yun laid out the field of smart finance from science and technology, and set up a joint laboratory with the four major banks to explore the intelligent scene of the bank. At present, it has provided products and technical services to more than 400 financial institutions, including China Commercial Bank and China Construction Bank.
According to the prospectus, from 2019 to 2021, cloud revenue from technology was 807 million, 755 million and 1.076 billion yuan, respectively. Due to the small revenue base in 2020, the year-on-year increase in 2021 is as high as 42.6%.
According to the preliminary calculation of the company's management, the revenue in the first quarter of 2022 is about 177 million to 202 million yuan, an increase of 38.72% to 58.31% year-on-year, mainly due to the continuous implementation of new infrastructure projects and the increase in artificial intelligence solution projects.
Compared with the giants, the cloud is still large from technology. From 2019 to 2021, SenseTime's total revenue was 3.027 billion yuan, 3.446 billion yuan and 4.700 billion yuan, respectively, and the volume of Cloudcong Technology was less than one-third of its volume in the same period. In 2020, affected by the epidemic, SenseTime achieved 13.84% growth, while Cloudcong Technology fell by 6.44% year-on-year. In this short period of time, the cloud is hopeless to catch up from science and technology.
The bigger dilemma is that the cloud has not stepped out of the quagmire of losses since technology. According to the prospectus, from 2019 to 2021, the net profit attributable to the mother of Cloud From Technology was -640 million, -813 million and -632 million yuan, respectively, and the total loss in the past three years was as high as 2.084 billion yuan. In the first quarter of 2022, the company forecasts a narrowing of loss margins.
The "No.2 Study" found that serious losses have become a common feature of the "FOUR LITTLE TIGERS OF AI". From 2018 to 2020, Megvii Technology lost a total of 12.73 billion yuan; SenseTime lost a total of 20.548 billion yuan; from 2018 to the first half of 2020, Yitu Technology lost a total of 6.102 billion yuan. The cloud is the least profitable from technology among them.
In the prospectus, Cloud From Technology said it plans to turn a profit into a profit in 2025. According to the forecast results, from 2022 to 2025, the total revenue of Yuncong Technology will reach 1.682 billion yuan, 2.501 billion yuan, 3.227 billion yuan and 4.011 billion yuan respectively, with a compound growth rate of 33.60%.
"No Two Research" believes that with the current analysis of the profitability of the cloud from technology, it is easier said than done to reverse the loss situation.
From the perspective of gross profit margin, cloud technology ranks first from the bottom in the "AI Four Little Dragons". According to the prospectus, from 2019 to 2021, the gross profit margin of Cloud From Technology was 38.89%, 43.21% and 36.76% respectively; from 2020 to 2021, the comprehensive average gross profit margin of comparable listed companies was 59.99% and 55.94%, respectively, which were significantly higher than the level of Cloud From Technology in the same period.
High accounts receivable can also have a negative impact on profitability. In the past three years, cloud accounts receivable from technology have remained high, with corresponding balances of 308 million, 523 million and 420 million yuan respectively. Especially affected by the epidemic in 2020, accounts receivable accounted for nearly 70% of the current year's revenue; in other years, accounts receivable have also approached 40% of the current year's revenue.
If the gross profit margin continues to decline and accounts receivable continue to be high, the cloud's turnaround plan from technology will most likely become a mirror flower. AI track research and development cycle is long, high investment, how to break the curse of loss, to achieve sustained profitability, has become the biggest challenge of cloud technology.
02 Hardware gross margin is low, and the business is getting narrower and narrower
As one of the "Four Little Dragons of AI", the cloud is both repaired from the technology software and hardware business.
According to the prospectus, the main products of Cloud From Technology include human-machine collaborative operating system and artificial intelligence solutions, of which artificial intelligence solutions dominate. From the perspective of industry attributes, the human-machine collaborative operating system belongs to the platform field and can be understood as a software business; artificial intelligence solutions belong to the application layer, that is, hardware business.
According to the prospectus, from 2019 to 2020, the revenue of the human-machine collaborative operating system was 183 million, 237 million and 136 million yuan, accounting for 23.48%, 31.50% and 12.72% of the total revenue, respectively; the revenue of artificial intelligence solutions was 597 million, 515 million and 934 million yuan, accounting for 76.52%, 68.50% and 87.28% of the total revenue, respectively.
Due to the characteristics of the AI supply chain, the gross profit margin of the cloud from the two businesses of technology is very different. According to the prospectus, from 2019 to 2021, the gross profit margin of the human-machine collaborative operating system was 89.30%, 75.86% and 73.99%, respectively; the gross profit margin of artificial intelligence solutions was 23.43%, 28.19% and 31.34%.
Although the gross profit margin of the hardware business is generally low, the gross profit margin of cloud from technology's artificial intelligence solutions is also at a low level compared with peer competition. According to the 2021 financial report data, the gross profit margin of artificial intelligence solutions of Yitu Technology, Cambrian and Megvii Technology is generally above 50%. The low gross profit margin of the hardware business has become the "culprit" of the cloud ranking in the lower gross profit margin from the main business of technology.
"No Two Research" believes that in addition, the hardware business of cloud technology is still facing a narrower and narrower dilemma.
As the JDM model (joint development) adopted by cloud technology is subject to the cooperation of hardware manufacturers, the gross profit margin of artificial intelligence solutions is pulled down, which drags down the overall gross profit margin level; at the same time, third-party software and hardware products are not core technologies, and excessive use will also cause the risk of core technology instability. According to the prospectus, the revenue of third-party software and hardware products accounted for 57.71%, 42.32% and 36.17% respectively, and the degree of dependence was high.
In this case, continuous acquisition of new customers becomes a way to improve profits. From 2018 to 2020, the number of customers who purchased AI solutions was 859, 965 and 741, respectively, and the overall decline was declining.
Perhaps Cloud From Technology also realized that the hardware business has limited improvement in the overall operating conditions, so it turned to supporting the software business with higher gross profit margins. However, the software business is mostly a consumption, and the growth of the software business will also drive the hardware business accordingly, and it is not easy to get rid of the dilemma.
Today, cloud technology has been gradually implemented in the four major fields of smart finance, smart governance, smart travel, and smart business, of which smart finance and smart governance are the main components of cloud business landing. Because smart gates, AI cameras, financial TELLErs and other equipment do not have too many differentiated requirements, the popularization of solutions is relatively easy, so Cloudcong Technology has considerable advantages in business landing.
However, CloudCong Technology is facing an increase in customer concentration. According to the prospectus, from 2019 to 2021, the concentration of the top five customers was 51.83%, 27.92% and 69.58%, respectively. Relying too much on a small number of large customers, the future sustainable operation of the cloud from technology may face a test.
There are also many problems with G-side customers. On the one hand, it may lead to the cloud from the technology bargaining power is relatively weak, the payment collection cycle is long, the application scenario is more limited; on the other hand, the huge commercial consumer market may be under the control of competitors.
The business environment has many challenges and the capital market is changing, and if you can find a breakthrough in the application field, you can cross the valley floor. As Yang Hua, vice president of Cloud Conghua Technology, said in an interview in December 2021, landing from the "cloud" is the destination of AI. Better seeking the balance between R&D and commercialization, hardware and software, becomes a compulsory course in the future of cloud technology.
03 Ai industry cooling down, where does the cloud go from?
The investment boom in the AI industry has gradually shifted from fanaticism to sanity.
With the popularization of technology and the intensification of competition, the bubble of the AI industry has long spread to the technical layer. However, due to the poor short-term rate of return and the suppression of the "entity list", the funds of hard technology are accelerating to the rapid development of autonomous driving, robotics and other areas, and AI has experienced cliffs and stepped off the altar. That's not good for unicorns that rely on financing.
Under the internal and external difficulties, the industry's internal involvement has become more and more intense. Internet giants and traditional software and hardware manufacturers have long been in the game, crushing unicorns in terms of talent, research and development, funds and customers; the "AI Four Dragons" may only rely on landing experience and application accumulation.
However, at present, the homogenization competition between AI enterprises is serious, especially the intertwining and overlapping of businesses between the "FOUR AI Tigers". Essentially, AI is an underlying technology, and the opportunity lies in how to deeply integrate with various fields. "Fuji Research" believes that for the cloud from technology, finding differentiated areas and deepening them will be the basis for it to achieve a turnaround and build a moat.
In fact, SenseTime, Megvii and Yitu have gone further in the dislocation competition. SenseTime tilts towards the direction of M&A investment and strives to expand the circle of friends; Megvii Technology focuses on the supply chain Internet of Things, hoping to make efforts in the subdivision track; Yitu Technology positions AI chips with higher requirements and greater challenges, and takes the ecological closed-loop route of combining software and hardware; and Cloud From Technology is still in the direction of security and finance, which has long been a must for AI applications to land, and the uniqueness is insufficient.
In addition, the growing trend in R&D expenses has overwhelmed the cloud from technology. According to the prospectus, from 2019 to 2021, yuncong technology's research and development expenses were 454 million, 578 million and 534 million yuan, accounting for 56.25%, 76.59% and 49.67% of the total revenue, respectively. Under the general trend of intensifying competition and R&D involution, R&D expenses may continue to grow.
By increasing investment in R&D, it is still the general consensus of the AI industry to widen the moat as much as possible. Under the premise that commercialization has not yet been fully implemented, and increasing investment in research and development, the cloud is doomed to be difficult to change the loss status quo in the short term from technology.
High equity incentives also put a lot of pressure on performance. From 2019 to 2021, Cloud Recognized Share Payment Expenses of RMB231 Million, RMB190 Million and RMB177 Million respectively.
In addition, the selling expense ratio is also diluting net profit. From 2019 to 2021, the sales expense ratio of Yuncong Technology was 28.29%, 36.28% and 26.05%, respectively, which was higher than the average level of comparable listed companies.
In the case of increasingly difficult to obtain the favor of capital and its own inability to make blood, the time left for cloud technology is running out, and cost reduction and efficiency increase have become more and more the embodiment of competitiveness. In addition to security and finance, automotive, medical, education and other scenarios are still the "deep water area" of AI commercialization, and the maturity of algorithms or the scale of application landing are still low, which is a new direction worthy of deep cultivation in the future, and it is also the focus of differentiated play.
After the investment tide recedes, the cloud can only return to itself from the improvement of profitability in technology. This lies not only in the breakthrough of its own technology, but also in the improvement of the data collection environment and the large-scale expansion and maturity of application scenarios. All this can not be solved in the short term, but only by virtue of this can we occupy a place in the development of the market.
04 AI Unicorn Blood IPO: Easy to run and hard to sprint
For AI, this is the best of times and the worst of times. In the midst of internal and external troubles, AI companies that are deeply involved in losses have come to the crossroads of life and death.
In the context of megvii technology and Yitu technology IPO encountered obstacles, the cloud from technology that is only close to ringing the bell to be listed has undoubtedly gained a pioneering advantage. However, the listing is far from going ashore, and it is easy to rush and difficult to sprint. Under the rapid growth of revenue, it is a gap that cannot be ignored with the giants; it is difficult to turn losses into profits; the hardware business is getting narrower and narrower, dragging down the overall gross profit margin; too dependent on large customers, the ability to continue to operate is to be examined; research and development is more and more internal, and the performance is overwhelmed...
Every step forward from technology to the cloud still faces a siege of competitors. Only by looking for a differentiated competitive path can we form a competitive advantage, obtain continuous hematopoietic ability, and completely solve the curse of loss. The dividend of the AI industry has faded, and only this is the key to breaking through. (This article was first published on the Titanium Media APP)
Some references in this article:
How long can the cloud be rolled from technology? BT Finance
"Cloud bleeding from technology to listing, AI industry "money" is uncertain? Titanium Media
"Cloud from Science and Technology Yang Hua: Landing from the "Cloud" is the Destination of AI", Fudan Management Vision