I have heard from many investors that the longer you look at the "industrial Internet", the more confused it becomes. However, what really puzzles and awes us is actually "industry", not "Internet".
From the end of 2012 when GE first proposed "Industrial Internet", to the German exhibition in Hanover in 2013 when "Industry 4.0" was put forward, and to the release of "Made in China 2025" in the State Council, China in 2015, industrial Internet and advanced manufacturing have been promoted to the strategic level in many countries.
However, Gartner’s report "Magic Quadrant of Industrial Internet Platform" released for the first time in 2018 was very restrained. In this paper, it was suggested that CIOs "must be cautious" when investigating suppliers, mainly because:
The application scope of the current platform is limited to very "limited" scenarios, and device connection management is the most basic requirement;
Well-known manufacturers such as GE and Siemens are excluded from the quadrant diagram because they do not support privatization deployment.
The former almost poured cold water on the value of industrial Internet, while the latter also questioned the feasibility of SaaS in this field.
Even in the quadrant map updated in 2019, no supplier has entered the "leader" quadrant, and the market is still scattered.
(Source: Gartner 2019)
Then, in 2018, GE broke out and planned to sell its industrial Internet platform Predix. In 2019, industries such as automobiles and consumer electronics also entered cyclical adjustments. Now, in the face of frequent "black swan" incidents and the "de-globalization" situation, the demand side has shrunk, which further aggravated the shock at the manufacturing end.
The industry is in dire straits, innovation enters the deep water area, and investment is even worse.
Shortly after the publication of the work "Hub", Sino-US trade friction started. Various media have analyzed and compared the impact of friction on manufacturing industry, and think that the "8" network of global supply chain proposed in Hub has failed.
However, at this time, it launched a survey that was completely opposite to the form of Hub-devoting to the fieldwork. He went deep into Viet Nam, Pearl River Delta and other manufacturing centers, and from the micro level of a factory, a product and even a "China businessman", he restored the real reasons behind the industrial transfer and reproduced the hidden information behind the relocation of factories and businessmen. Spillover: The Future History of Made in China was born under such a background.
Therefore, to understand the industrial Internet and IoT, we must first start from the first principle, based on the industry itself, and dismantle the factory value chain, so as to find the fulcrum for technology and the Internet to incite the industry.
Secondly, taking history as a mirror, we can review the transformation path of giants such as traditional manufacturing and industrial software, and analyze the pain they experienced and the way forward, so as to get rid of the false and retain the true under the slogan.
01 Redefining Industrial Internet
The following is the departmental structure, functions and corresponding value chain of a typical factory.
The figure reflects the complete process of a product from the OT (operation technology) level, and the factory must ensure the efficient cooperation of all links, form the unified management of "people, machines, materials, methods, environment and measurement" in production, and ensure the quality and consistency of finished products, and must not lack the support of IT (information technology).
More importantly, the value of Product Life-cycle Management (PLM) system can be truly reflected by connecting the basic IT systems, breaking the data islands and effectively corresponding to each link in the value chain. PLM has laid the foundation for the digital transformation of enterprises, and it is also the foundation of the "integration of industrialization and modernization" that we often hear.
PTC, an American parametric technology company, which is in the leading position in Gartner’s magic quadrant, pointed out in its White Paper on Enterprise Digitalization Transformation that enterprise digitalization is not an end, but a means.
This report puts forward a very valuable concept "Digital thread":
"Using digital technology, covering the whole product life cycle and the whole value chain, we will build a digital data stream that integrates data and things, runs through product research and development, manufacturing, marketing, operation and service, and provide real-time data analysis and decision support for all levels of enterprises."
It is not difficult to see that the "digital mainline" is essentially a further extension of the product life cycle management system. The factory establishes a network from the production side through more advanced sensing, calculation, control and transmission equipment, so that employees or managers of each department can trace back the products and services of the enterprise at the right time and in the right way, and obtain the most reliable, effective and real-time data.
At the same time, the "digital mainline" is also the basis of the "Digital twin". Engineers can analyze and compare the real running state of products or equipment in a specific scene with the digital model through mainline data, and provide real-time monitoring, simulation verification and prediction analysis for end users. The factory itself can also conduct product simulation and quality analysis in the research and development stage to improve the design of the next generation of products.
Simply put, the first step in the digital transformation of modern enterprises is to establish a digital main line.
"Industrial Internet" is a series of digital innovations around the value chain on the basis of digital mainline, so as to enhance the competitiveness of products and services and ultimately improve economic benefits.
Among the earliest representative cases of industrial Internet, GE is most proud of its successful transformation in aero-engine business.
(source: GE)
Relying on advanced sensors, automation, internet of things and other technologies, the digital main line of engine production is established, and then a real-time dynamic digital twin model is built for each engine through big data and AI technologies to realize the integration of mechanism and mathematical model. GE finally provided airlines with additional value-added services such as real-time engine condition monitoring, operation and maintenance management and operation optimization to help customers further avoid losses caused by engine failures and even safety accidents, and reduce the overall fuel cost by tens of millions of dollars every year.
GE not only upgraded its own business from product sales to continuous service, but also reconstructed the value chain of engine business from the beginning to the end in conjunction with partners such as ground service and engine parts factory.
This case also reveals the three core values of the industrial Internet. I believe that all innovations will return to these three points in the future:
Create additional and quantifiable business value for customers;
Reconstruct the value chain to create additional economic benefits for enterprises themselves;
Establish upstream and downstream networks to promote the redistribution of profits in the industrial chain.
On this basis, we further obtained the relationship between Internet of Things, industrial Internet, industrial Internet and even consumer Internet.
Just like the consumer Internet, the industrial Internet must be based on connectivity, digitizing, networking and intelligentizing all links in the industrial chain, and cooperating with the highest quality, the highest efficiency and the most reasonable cost through data driving.
Industrial Internet is an example of industrial Internet thinking in industry. Its ultimate value is to make all users and partners more intelligent and flexible in this network, and finally get greater economic benefits.
02 when the industrial internet gets an electric shock SaaS
A company that cannot serve large enterprise customers is not a mature SaaS company.
This sentence contains two meanings:
First of all, maturity is not linked to the quality of the company, because good and bad are a relative evaluation. And "maturity" represents a sign that the company is or has established a certain moat and is developing from "big" to "strong";
Secondly, this is not to deny the value of SaaS to small and medium-sized customers. It is now standard for SaaS companies to choose a niche market with vertical scenes, but this can never explain the company’s core competitiveness.
If Salesforce was born that year and only made traditional CRM vendors sneeze, then the big account strategy after listing caught all traditional software and IT companies off guard, and the "post-wave" led by Oracle, SAP and Microsoft announced "cloud transformation" one after another, and then a tortuous transformation route gradually surfaced.
When we say that industry is testing IoT and SaaS, two things are actually happening behind it:
First of all, industrial or manufacturing giants began to explore the paradigm shift of business from one-time sales to providing continuous services. The underlying logic is the upgrade of business model, which drives products and services to upgrade intelligently, just like the GE engine case mentioned earlier.
Secondly, traditional industrial software starts the iteration of products from the traditional architecture to the underlying architecture of SaaS. One sentence in the book Software Defined Manufacturing is impressive: industrial software is used, not developed. Decomposing industrial software with traditional architecture into flexible and configurable applications with unified interface is industrial APP.
APP has the characteristics of vertical scene cutting, focusing on a single pain point, flexible and configurable, which is completely consistent with the design idea of SaaS products.
In my opinion, this is probably the biggest opportunity for SaaS companies in the industrial Internet.
However, due to the complexity and professionalism of the industry itself, the opportunities seem numerous, and the actual interlacing is like a mountain. Without understanding the pains and solutions of traditional industrial giants in the process of cloud transformation, it is impossible to comprehensively and objectively evaluate new opportunities.
Ptc: the way to transform the industrial software giant
There is nothing new in industry. As an industrial software company established in 1985, PTC has established a product matrix with CAD, PLM and SLM (post-service life cycle management) as its core by 2013. In 2013, the annual revenue has exceeded 1 billion US dollars, but the annual growth rate is only 3%. The bottleneck of development is that the revenue of software business sold by one-time License is only 340 million US dollars, down 1% year-on-year, and the bulk of the revenue has been transferred to the follow-up Support service.
Therefore, at the end of that year, it was natural for the market to acquire ThingWorx, an IoT application development platform, for US$ 120 million, which not only made up for the shortcomings of SLM software in equipment connection and data management, but also laid the foundation for building a Thing-centered IoT cloud development platform.
(the way of PTC transformation)
CEO Jim Hepplemann once talked about his views on the future from the demand and manufacturing sides in an interview:
The deeper change comes from the downstream customers of manufacturing industry, who care about how to get more "functions" and "services" from products, but not necessarily about "owning" products;
From the perspective of PTC customers, they are concerned about how to digitally manage the whole process from demand research, design research and development to production and delivery, and finally improve R&D design and develop value-added services.
As Jim said, two boundaries are gradually blurring:
At the value chain level, the boundary between products and services is blurred; On the IT level, the boundary between ERP and PLM is blurring.
These two changes also correspond to two important values of SaaS: continuous service will replace products as the main source of income; The interconnection of life cycle data on the cloud has become the basis of value release.
Returning to the above transformation roadmap, PTC has made two major changes since 2014:
Change the sales mode of one-time authorization to subscription; Continue mergers and acquisitions and strategic cooperation to improve the layout of IoT and create an application development ecosystem with ThingWorx as the core.
The first change brings the pain of growth. Due to the change to annual installment payment, the initial investment of customers and the fees paid for follow-up support will be reduced accordingly, and the conversion of old customers cannot be achieved overnight.
According to the internal calculation of PTC at that time, when the customer continues to pay for more than 4.5 years, the total investment will be the same as that of the previous one-time payment, that is, PTC will face the pressure of at least 4 years in revenue. This is basically consistent with the assumption mentioned in the "Risk Factors" in the company’s 2015 annual report:
“Our plan through 2021 assumes that our license revenue and earnings will decrease in 2016, 2017 and 2018 due to lower up-front revenue recognized for a subscription license compared to a perpetual license, And assumptions increases in revenue from a recurring subscription revenue stream beginning in 2019. "(By 2021, the plan assumes that our revenue will continue to decline from 2016 to 2018 until it recovers through continuous repurchase in 2019).
(Source: PTC Financial Report)
I tracked the revenue in recent five years from the income type and product structure, and found two points:
1. The effect of the transition to SaaS charging mode is beginning to show: the proportion of continuous income driven by subscription is finally close to 50% after five years of launch. It should be noted that the part of continuous income brought by services is excluded here, because it has nothing to do with the transition of subscription;
2. There is a long way to go to develop the IoT platform: the IoT business driven by ThingWorx is still in the early stage of development, accounting for only 12% of the total business, that is, the total global revenue is only about 150 million US dollars, indicating that most of the subscription revenue comes from the change of the charging model of traditional industrial software.
Secondly, what role does ThingWorx play in the IoT sector?
Generally speaking, ThingWorx empowers customers on its platform to better realize cash through innovative services, and the platform itself benefits from this continuous service.
For example, Colfax is a manufacturer of industrial products and mechanical equipment, whose products include precision ventilation systems and Welding & Cutting machines, and provide equipment maintenance services and sales of related consumables. Colfax aims to provide more quantifiable value-added services on maintenance services, and guide the design and development of new products through after-service.
With the cooperation with ThingWorx and Microsoft Azure, Colfax has achieved three goals:
Uptime: By collecting and analyzing the key data of the air compressor in the ventilation system and establishing the machine failure model, it can help customers avoid additional losses caused by temporary failures, and formulate maintenance countermeasures in advance to improve the success rate of the first maintenance;
WeldCloud & CutCloud: By connecting welding and cutting equipment, the transparency of equipment operation efficiency is improved, and the standardization of operation documents and asset management efficiency are promoted; Finally, because the business and data are connected in the cloud, we can give effective feedback to the new products being developed, and at the same time, we can synchronize with the suppliers in real time in the design of related consumables to avoid the suppliers’ invalid investment in the early stage.
This typical cooperation case basically shows the advantages and disadvantages of ThingWorx and PTC:
Although the financial report only shows that ThingWorx is charged on a subscription basis, I learned through recent financial report calls and interviews with insiders that:
First, the key to growth lies in the depth of customers’ use of the platform. When customers connect more devices through the platform, or increase the number of measuring points on each device, it means that more data will be mined and applied, and the unit contract value, namely ACV, will increase, thus increasing ARR.
Secondly, the exposed problem lies in PTC’s lack of in-depth understanding of the customer’s industry, that is, the combination of equipment and usage scenarios. At present, the company chooses to directly face the end customers together with its partners, aiming at understanding how customers go from data mining to diagnostic analysis through joint services and solving practical problems.
The CEO admitted on the earnings conference call in Q1 this year that ThingWorx is still in the stage of "Land and expand", which is the promotion method adopted after Salesforce announced its key account strategy after listing. It is said that at present, 60% of new orders come from customers’ expanding demand.
Finally, PTC hopes to establish such an ecology: the platform gradually retreats behind customers, just like Microsoft’s role in PC and software, and Apple’s role in personal mobile phones and apps, helping platform users to realize cash effectively and continuously.
(ThingWorx ecological concept)
From this point of view, the transformation of PTC can only be said to be half successful. Combined with the previous engine case of GE, we can draw the following conclusions:
1. Industrial software and data interoperability are the basis of building industrial APP, but the development of individual software has reached the bottleneck;
2. The best practices of industrial Internet are usually born within large customers, and first landed in high-end and advanced manufacturing industries;
3. In order to realize continuous charging, the industrial Internet platform must first help customers to upgrade to the service mode.
Therefore, the transition from PTC to subscription or so-called "SaaS" is not the ultimate goal, but IoT is.
The transformation of PTC also explains from the side why the number of Internet of Things and industrial Internet platforms is still expanding at home and abroad, and has not begun to converge.
(Source: IoT Analytics)
The leader of each sub-industry has the opportunity and hope to open the best practices in the whole industry, so as to connect the upstream and downstream of the industry and establish an efficient collaborative network.
What does SaaS mean to industry?
In the future, whether in the factory or in the supply chain and its outsourcing network, it will certainly become more flexible.
The foundation of flexibility is to realize data coupling and interaction. In the IoT world, it is software that transforms multiple heterogeneous data in the physical world into numbers and information that can be uniformly processed. The data source not only includes human operation, but also a large number of complex devices driven by the principles of mathematics, physics and machinery, so the word "software" has two meanings here:
The first layer is industrial software that assists people in manufacturing; The second layer is the micro-control and drive system embedded in precision equipment.
When the software meets SaaS, the former goes to the cloud in business mode, which greatly accelerates the high-speed calculation and real-time interaction of the underlying data, and is the key to establishing the digital mainline; The latter laid the foundation for the transformation of manufacturing industry from product sales to continuous service from product structure and design concept.
At the same time, the value of SaaS has been further amplified in this crisis. At the end of March, the CEO of PTC talked about a customer from an automobile OEM. Usually, their engineers use locally deployed CAD and PLM software to design and simulate on a large Workstation in the office. After telecommuting, these softwares can’t run on the ordinary PC at home, so I turned to PTC for help, hoping to find a set of SaaS solutions that can run and interact on the cloud as soon as possible and be compatible forward.
Looking back at the opportunities in China, according to the PTC Annual Report in 2014, China’s business accounts for 5-7% of the total revenue, corresponding to about 80 million US dollars, which is only the industrial software part. CAD and PLM are mainly used in aerospace, automobile, construction machinery and communication in China. These customers are deep users of industrial software and will never easily replace suppliers.
As the IoT business involves the implementation and landing of the solution, as mentioned in the analysis of advantages and disadvantages, PTC insisted on facing the end customers together with integrators in the early stage, and tried to go deep into the market in the form of establishing a subsidiary in China, but it was abandoned in the later stage due to various force majeure factors, so the progress of IoT business in China has not been smooth.
On the other hand, China’s manufacturing industry is known as "big but not strong". Uneven information foundation, special business environment and traditional payment habits have all become obstacles to the landing of IoT platform and innovative applications. I will discuss it when I have the opportunity in the future.
Therefore, when evaluating the feasibility of SaaS in industrial internet, we should first understand whether we are thinking about this issue from the perspective of business model or product positioning.
It takes PTC five years or even longer to prove that the latter is the real ambition and driving force for growth, but there is no doubt that PTC has a strong software foundation and customer base that cannot be copied by domestic related enterprises.
Finally, for start-ups, instead of sticking to the limitations of domestic business environment or mode, it is better to think about how to jointly tap customers who really need or are trying to upgrade their services with the help of industry partners and IoT platforms.
SaaS is highly flexible and open, which is the basis for the industrial Internet to incite the industry with continuous services. However, it is unrealistic and unreasonable to make a final judgment on the future, and even SaaS itself is constantly being iteratively upgraded.
At this moment, it has more important reference significance to resume the transformation of traditional manufacturing and software giants. After all, as "Hub" said:
History is futurology.
-End-