Jinshajiang Zhang Yutong: Redefine the whole industry with data and intelligence

"We judged 5 or 6 years ago that the infrastructure of cloud computing is a giant track. We have seen the first batch of entrepreneurial teams in this field, but based on the judgment of the entire track, we did not take action." Asked about his views on IaaS, Zhang Yutong directly gave his judgment on the entire track that year. This judgment is obviously verified by the market and time.

Zhang Yutong, Partner of Sands River Ventures

Some tracks are not born as a startup

In the history of the domestic enterprise service industry, waves of companies have been born, and the most well-known one is Alibaba Cloud. Today, in the field of IaaS, it seems that the dust has settled. Regarding the weekly investment and financing reports released by Titanium Media · "Potential Investment" and the 2019 primary market semi-annual report statistics, few IaaS new companies have received financing.

Reviewing the development process and pattern evolution of the underlying infrastructure track of cloud computing, Zhang Yutong said: “This is something that giants will definitely do and can do well.” Cloud computing originated from the internal rigid needs of e-commerce or giant companies. It met the needs of cluster computing earlier than small and medium-sized enterprises and accumulated the ability to handle peak computing, but the demand for such computing is not stable and predictable. So from the giants themselves, they naturally have the need for flexible management and external export capabilities.

From the demand side, “the more the lower the customer’s needs are, the more unified” the upper-layer applications can support transactions or data analysis, with different business forms, while the lower-layer needs almost the same basic core capability of computing , Low latency, low power consumption and so on. In the final analysis, this type of enterprise service company is "exporting computing power," and this computing power is undoubtedly a scale effect.

Based on this logic, as early as 2013, Zhang Yutong also looked at many cloud computing start-up teams and made a judgment: even if they have a certain first-mover advantage, their business is growing rapidly with the birth of mobile Internet emerging companies, but this competition Dao himself is not friendly to startups with difficulties.

In Zhang Yutong’s view, another opportunity that does not belong to a startup company is about "time": the opportunity that belongs to a startup company should be an opportunity in the next 1-3 years, and those things that take 10 years to see are more suitable. Universities or research institutes of large companies do it. For example, strong artificial intelligence, such as virtual reality, have huge imagination, but the technology is far from mature. That is not an opportunity for entrepreneurs. Startup companies should focus on the application of a certain core technology in a specific scenario.

For early investors, this seems to herald a paradox about time. Ending prematurely will inevitably become the "historical promoter" of the education market, and missing the spotlight is a pain that investors cannot bear. Whether the trigger can be pulled at the right time is a question that every investor must face.

In this regard, Zhang Yutong is not anxious. On the one hand, we must always be concerned about and sensitive to emerging technologies. On the other hand, in each specific case, we must think independently and do sufficient homework to verify whether the technology can be applied to a real scene and judge whether it is Is it the most efficient and executable solution in the actual business scenario, rather than paying for the technical vision or story?

Enterprise Service 2.0: Redefining an Industry with AI

In the short history of China's science and technology Internet, ERP has been around. After more than 20 years, its model has been well-known and accepted, but on the other hand, with the complexity of enterprise business and the escalating demand for systems, the former enterprise-level service "king" also seems to fall into the altar.

Why can't ERP carry the banner of enterprise service? Zhang Yutong admitted to Titanium Media that ERP is indeed the most rigid demand for enterprises, but with business changes, a set of ERP can no longer carry all the business of the enterprise. This is why companies need a strategy of large, medium and small front office. ERP cannot do CRM, ERP cannot do OA, and ERP cannot do HR. These requirements have always existed, but in the past, some of them were solved by customized software, and some were put on the next level of priority. However, with the development of technology and business With the evolution, to this day, these needs urgently need to be solved better.

Before, the biggest role of enterprise service software was "recording." All production and operation data can be retained, inquired, analyzed, and audited. Nowadays, companies have long been unable to meet the original requirement of "recording". With the enrichment and complexity of business, companies have gradually created new requirements for data analysis, data mining and even data-based forecasting and assisting business decisions.

When data becomes a new means of production, various industries will be redefined, which undoubtedly heralds huge opportunities for entrepreneurship and investment.

In Zhang Yutong's investment case, Moka HR is a typical representative. This company is committed to providing a set of SaaS-based intelligent recruitment system for medium and large enterprise users to help companies manage all recruitment processes in a unified manner. Due to the automatic data collection and real-time update, the work flow is interconnected and easy to collaborate, which makes the enterprise HR department Work efficiency has been greatly improved.

In Yutong's view, Moka is an example of the combination of AI and HR industries, and their business logic and strategic path are equally applicable to other general-purpose SaaS.

One is to let data really drive production. The so-called intelligent system refers to the predictive analysis of the operation of the enterprise and the auxiliary judgment in important decision-making. Only on the basis of a database with automatic collection and real-time update can an intelligent system be constructed. Automatic collection and real-time update are what the machine is good at and relatively mature in technology. Finding the most suitable point for leveraging technology, and achieving the ultimate at this point, is the first step for startups to use technology to drive industrial change.

Second, there are more opportunities to serve medium and large customers. In the field of corporate services, the differences in the needs of large, medium and small customers in the same industry are often even greater than those across industries. Facing small and medium-sized customers, general-purpose SaaS products are often relatively simple and thin. Such pure tool products are not only “made by the giants”, but also face the natural disadvantages of small and medium-sized enterprises, such as natural loss and elimination and low willingness to pay. The dilemma of low customer acquisition efficiency and low monetization efficiency. To empower SMEs, perhaps more consideration should be how to serve more C-end users through SMEs, and how to monetize through data, finance and other business models. In Zhang Yutong's view, there are more opportunities to serve medium and large customers. When medium and large enterprises are really moving and embracing the new generation of technology, it is undoubtedly a bigger and more sustainable business opportunity to meet their needs.

Looking for corporate service benchmarks in China's business ecosystem

The continuous popularity of the corporate service track in the United States in both the primary and secondary markets has made investors and entrepreneurs extremely excited. In the consumer Internet, the investment and entrepreneurial strategies of benchmarking the United States have repeatedly worked. In the field of corporate services, will history repeat itself?

Zhang Yutong's past experience gave her a deeper understanding of this issue. Zhang Yutong graduated from Tsinghua University with an undergraduate degree and studied at Stanford as a graduate student. In her view, the development of the industry has its own inherent laws. The US corporate services are 20 years earlier than us. The development and evolution of these companies are worthy of our study and review. But more importantly, the understanding and practice of corporate services in China's business ecosystem.

After the baptism of the Internet and mobile Internet, China is forming its own unique business ecology. In China, ToC giants have spared no effort in investing in ToB, and the worries that "nothing will survive under the giants" are rampant. Someone once joked that a single "operation" position reflects the very different Internet ecosystems between China and the United States. In the United States, it is difficult for people to understand what this post is neither a product manager nor a programmer. In China, user operations are important, not only operations, but also "strong operations."

In Yutong’s view, China’s business ecology is gestating unique opportunities:

1. Chinese companies are more willing to pay for calculable value.

With the penetration of cloud computing and big data, the traditional long software delivery model has gradually been abandoned by the times. Zhang Yutong believes that the cycle of custom development is too long and the results are uncontrollable, which does not meet the customer's pursuit of ROI. She gave a vivid example to Titanium Media: "Imagine you are a professional manager. During your tenure, will you have a system that costs a lot, has a long cycle, and has clear needs but unknown effects?"

In Yutong's view, RPA can become one of the few outlets this year. It is not groundless, it is a matter of course. The emergence of RPA is a one-stop solution to corporate IT pain points.

RPA development efficiency is high, and the deployment cycle is short: RPA based on screen recording, graphical interface, and application ecology can minimize the amount of code, improve development efficiency, and greatly reduce the threshold for companies to launch robots. The usual RPA project deployment cycle will not exceed 3 to 6 months. More intuitively, in some cases, there is a clear one-to-three or even one-to-five relationship between the robot and the worker, and the labor cost saved is easy to calculate. Generally, breakeven can be achieved within 9 months of implementation. The value is clear and the investment return period is short. For Chinese corporate customers, decision-making costs have been greatly reduced and their willingness to pay has increased significantly.

2. Large vertical industries are experiencing great opportunities for change.

In addition to general-purpose SaaS, it is also promising to use technological changes to promote the transformation and upgrading of vertical industries. In terms of vertical industries, China and the United States are also different. In the United States, the largest vertical industry is healthcare, while in China, manufacturing is even more impressive.

Now, manufacturing companies are beginning to actively embrace automation technology. Currently, 90% of Chinese manufacturing companies are in the process of shifting from manual to semi-automated and automated. Goldman Sachs analysis data shows that the cost of some underlying sensors of the Internet of Things has dropped by 50% in the past decade, bandwidth costs have dropped by nearly 40 times, and data processing costs have dropped by nearly 60 times. If a new wave of innovative technologies represented by artificial intelligence can increase the productivity of China's manufacturing industry by 15%, it will create a value of 4 trillion yuan.

Seeing the huge value, Zhang Yutong invested in Heihu Intelligent Manufacturing, starting from solving the problem of "information transmission behind factories" and realizing data-driven manufacturing. In just two years, Heihu Intelligent Manufacturing has formed a cloud-based, algorithm-driven, flexible and configurable multi-platform real-time collaboration system. The data granularity of the production site can be refined to 10 times that of traditional industrial-grade software. Through real-time calculation and deep learning, and through open API interfaces, the data can be combined and applied to form a complete upstream and downstream intelligent manufacturing ecosystem.

Look up at the stars and keep your feet on the ground. In Zhang Yutong's investment philosophy, there is no such thing as a common sense in the market, which is sometimes enthusiastic and sometimes sorrowful and joyous, but is always rational and optimistic. She said that it is unreliable to talk about the direction of investment out of time; she firmly believes that a batch of tens of billions of US dollar corporate service companies will be born in China in the next 10 years. This is an inevitable result of technological and business changes. (This article was first published in Titanium Media, author/Yuan Jing, Guo Hongyu, editor/Cai Pengcheng)