2019 Global Enterprise Blockchain Benchmark Study Report (Full Text)

Translator's Foreword: Recently, the University of Cambridge Alternative Financial Center released the second phase of the global enterprise blockchain benchmark research report, based on 160 entities in 49 countries and 67 enterprise areas deployed in the production environment. According to the data of the blockchain network, most of the current enterprise blockchain plans belong to the category of “blockchain memetic”, while only 3% of the analysis networks meet the standards of the comprehensive multi-party consensus system.

The study also explores the main technical concepts of blockchain technology and outlines the key features typically implemented by blockchain networks. In addition, the report describes various types of participants in the ecosystem, describes their activities and related income models, and empirically demonstrates the industry's development with employee growth data. The report highlights key drivers of blockchain strategy, activity assessment and remaining key challenges for wider adoption, and explores potential future development trajectories.

The authors: Michel Rauchs, Apolline Blandin, Keith Bear, Stephen McKeon

Translation: Babbitt (free and easy)

(Note: The full text of the report is about 27,000 words, and the core ideas can be found in the summary section)

table of Contents

1 Introduction

2, the research team

3, thanks

4, the abstract (too long to read version)

5, the first section: blockchain 101 – clear common misunderstanding

  1. Introduce the core concept
  2. Mapping different interpretations

6. Section 2: Introducing the enterprise blockchain ecology

  1. Exploring ecological landscape
  2. Know what industry participants are
  3. Industry development

7, the third section: adoption status

  1. Transfer to production
  2. Project life cycle
  3. Project scope
  4. Network design

8. Section 4: Blockchain strategy and business model

  1. User motivation
  2. Strategic implementation
  3. Evaluation of blockchain activities within the company
  4. Supplier strategy
  5. Blockchain revenue model

9. Section 5: Looking into the future, what will the blockchain industry be like?

10. Appendix 1: Research Samples

11. Appendix 2: Technology Platform

12. Appendix 3: Challenges

Foreword

The terms "blockchain" and "distributed ledger technology (DLT)" have become more and more frequent in our everyday vocabulary. However, they rarely involve the same thing, and unfortunately, the convergence of terms has hampered the clear assessment of blockchain technology in different industries.

Our second global corporate blockchain benchmark study aims to reveal the different interpretations of these terms and the potential confusion they are creating.

The report provides an in-depth analysis of how the enterprise blockchain ecosystem has evolved since 2017, using data collected from 160 entities and 67 active blockchain networks.

We believe that this report differs from other studies in that it focuses on the state of network deployment and investigates the ongoing development phases of enterprise blockchain-based projects.

This approach reveals several development models that provide a common reference point for industry participants and the general public. In many industries, enterprise blockchains are seen as solutions that build common data standards across organizations, eliminate organizational isolation, and facilitate record coordination to help improve overall efficiency and create new services.

In 2018 and 2019, multiple networks moved from concept to production, and the private sector's adoption of enterprise blockchains continued to increase. However, the blockchain technology is not a panacea, and industry adopters need to weigh it.

As the experience mentioned in this report proves, the current adoption of blockchain is indeed a slow and challenging process that requires in-depth cross-organizational coordination and careful legal and organizational design choices.

This explains, at least in part, that in most deployed networks, we observe that centralized design approaches dominate, although they expect to gradually distribute control over time .

We hope that this analysis will make an important contribution to the global deconstruction of the enterprise blockchain ecosystem.

As always, we recognize that the ability to conduct high-quality research depends to a large extent on industry initiatives and the cooperation of entities, and we thank those who provide information and data to the co-authors of this report.

Dr. Robert Wardrop, Director, Cambridge Alternative Finance Center

The second global enterprise blockchain benchmark study is a direct and comprehensive study of the use of blockchain conditions across the financial services industry.

It provides insight into how the blockchain program is built, supported, and moved to production, with a particular focus on proprietary, licensed distributed ledbook technology (DLT).

The study selected more than 160 stakeholders from a sample of 49 different countries around the world and then conducted research around blockchain projects.

The study will delve into the terminology surrounding the blockchain program to distinguish between true DLT projects and "blockchain memes."

There are several prominent points in the report: First, the success of the blockchain cannot and does not happen as isolated in the network. The real transformation of the ecosystem takes time. The new technology must prove that it can be established in the new paradigm. trust.

It has been noted in some places that the blockchain has entered the "disillusionment" phase of the technical speculation curve defined by Gartner. The study shows that trials, adoption, and implementation are still strong, but most blockchain programs fall into the category of “blockchain memes,” and “only 3% of the analysis networks meet the criteria of a comprehensive multi-party consensus system. One-of-a-kind projects can be viewed as potential distributed ledger technology systems that gradually eliminate single points of failure and control."

There is still a strong misunderstanding about trust in the blockchain. Potential participants are wary of whether the blockchain is trustworthy;

The blockchain originates from Bitcoin, a digital currency and transaction accounting book that is tailored to specific purposes and that builds trust between untrustworthy participants.

Observers have found that the concept of blockchain can be applied elsewhere, so the blockchain is beginning to be a utopian solution to the problem, which led to the adoption of the “blockchain meme” program. As of now, the blockchain is still in the early stages of providing real solutions to practical problems.

The financial services industry is a complex ecosystem, and we all experience the transformation of the blockchain/DLT. The true subversive potential means that the entire paradigm and operating model can be subverted. Many participants are very cautious when entering the unknown, because those who are in power are severely disrupted by the success of the blockchain, and these people have a vested interest, they will try to make the blockchain technology unsuccessful.

We would like to thank all individuals and teams who are committed to creating this global study. As Invesco continues to evolve in the pursuit of alpha generation performance and elite customer experience, we are implementing multiple blockchain initiatives at various intersections in the value chain.

Our relationship with CCAF and our access to global benchmark research reports help us understand the general trends in financial services that help us adjust our strategy and apply the right blockchain/distributed ledger technology Make an informed decision at the pace.

Dave Dowsett

Invesco Global Technology Strategy Director, Emerging Technologies

research team

Author

Michel Rauchs : Michel is the head of the Cambridge Alternative Finance Centre in the area of cryptocurrency and blockchain .

M.rauchs@jbs.cam.ac.uk @mrauchs

Apolline Blandin : Apolline is the Research Manager for Cryptographic and Blockchain at the Cambridge Alternative Financial Center.

A.blandin@jbs.cam.ac.uk @ApollineBlandin

Keith Bear : Keith is a researcher at the University of Cambridge Business School's Alternative Finance Center.

K.bear@jbs.cam.ac.uk

Stephen McKeon : Stephen is an associate professor of finance at the University of Oregon and a visiting assistant at the Cambridge Alternative Finance Center.

Smckeon@uoregon.edu @sbmckeon

Acknowledgement

We would like to thank Invesco for sponsoring this research, especially Dave Dowsett, Kevin Lyman, Henning Stein, Bradley Bell and Heather Wied to provide unremitting support throughout the research process.

We would like to thank the following individuals for their generous support and help during the research: Ashwini Anbujaran (Hyperledger), Richard Bloch (Hyperledger), David Boswell (Hyperledger), Gideon Greenspan (Multichain), Antony Lewis (R3), Dominic de Lisle (Calastone ), Marta Piekarski (Hyperledger), William Lovell (Hyperledger), Kevin Rutter (R3), Todd Henshaw (British Blockchain Association), and Jed Talvacchia (R3).

Special thanks to our interns, Larisa Barbu, Thomas Eisermann, Jaya Lalwani, Jinjun Liu, Jordan Matthews, Sebastian Resano, and our visiting students Kristina Klein and Martino Recanatini for their excellent work in survey communication and data collection.

We would also like to thank Xiao Ling of the University of North Carolina for his contribution to the draft investigation and a thorough review of the early draft by Hatim Husssain and Robert Pulser.

We would also like to thank the entire CCAF team, especially the center leaders Dr. Robert Wardrop, Dr. Raghavendra Rau and Bryan Zhang for their continued support and help. Special thanks to our interior designer Louise Smith for the gorgeousness of our report. design.

In addition, we hope that Philippa Coney and Charles Goldsmith of the Cambridge Business School will continue to support the production and publication of this report.

Finally, this study is impossible without the generous support of all the organizations involved in our investigation:

Note: Some respondents are reluctant to publicly disclose their participation.

123

Report summary

Since we released our first global blockchain benchmark research report in September 2017, the enterprise blockchain ecosystem has undergone tremendous changes: although two years ago, the industry landscape was dominated by inexhaustible experiments and short proof of concept. Leading (usually publicized and publicized), but hype has gradually given way to a truly sustainable blockchain network that is increasingly being deployed in production environments.

From July to November 2018, the research team collected survey data from more than 160 entities (including startups, established companies, central banks, and other public sector agencies) in 49 different countries around the world. Between April and June 2019, we also collected additional data from 67 enterprise blockchain networks to provide new insights into the state of the ecosystem and capture the major trends in the fast-growing sector of the blockchain.

The report reveals the following main findings:

1. Banks, financial markets, and the insurance industry account for the largest share of active blockchain networks :

The 2017 trend continues: 43% of enterprise blockchain network deployments fall into the financial services arena, far ahead of any other sector and industry. Specific use cases for the network are sometimes difficult to identify, but supply chain tracking, transaction infrastructure, and document authentication currently seem to dominate.

2. Successful projects require a long-term perspective and commitment :

Transforming critical market infrastructure takes longer than simple application development: the medium-sized enterprise blockchain project takes about 25 months from proof of concept to production, and the full launch of some large networks It takes more than 4 years and a half.

3. There is a common network of founders between the cooperative organizations :

While there are many reports of large alliances developing new blockchain networks, 71% of networks are initiated by individual founders of the Alliance Chain Program. 88% of deployed blockchains are designed for shared use among multiple independent entities, but most limit membership to partners. Only 19% of the network is operated jointly by direct competitors.

4. Cost reduction is the main value proposition of active networks, but it is expected that the second phase will increase revenue :

Currently, 72% of active networks are primarily used to reduce participants' costs by reducing reconciliation efforts. However, 69% of network participants indicated that the main motivation for joining a coalition chain project is the potential to generate incremental revenue by offering new products and services.

5. Hyperledger Fabric seems to be the platform of choice for all industries :

48% of the alliance chain projects chose Hyperledger Fabric as the core protocol framework for the network, followed by R3's Corda platform (15%) and Coin Sciences' MultiChain framework (10%).

6. Most active alliance chain networks are highly centralized, but according to the plan, they will gradually decentralize control over time :

81% of the alliance chain networks have leading entities that lead the governance process (centralized social consensus), and many networks (at least in their current form) use third-party service providers to host and operate nodes on behalf of network participants (centralized network consensus) .

7. Unclear terminology and marketing hype contributed to the “blockchain model” :

77% of active enterprise blockchain networks have little in common with multi-party consensus systems except that they contain some of the same technical components (eg, encryption, peer-to-peer) and similar terminology. Nonetheless, “blockchain memes” are still a powerful catalyst for overcoming corporate inertia and driving widespread organizational change, both within organizational boundaries and across organizational boundaries.

The study also explores the main technical concepts of blockchain technology and outlines the key features typically implemented by blockchain networks. In addition, the report describes the various types of participants in the ecosystem and describes their activities and related income models, while demonstrating the industry's growth using employee growth data. The report highlights key drivers of blockchain strategy, activity assessment and remaining key challenges for wider adoption, and explores potential future development trajectories.

First, blockchain 101: clear common misunderstandings

1, 1 Introduction to the core concept: the term "blockchain" or "distributed ledger technology (DLT)"?

Depending on the discussant, there are many terms in the industry that may have the same meaning or may involve completely different concepts. Similarly, there are various contradictory definitions of what constitutes a “blockchain” or a “distributed ledger”. This situation often leads to misunderstandings, unnecessary confusion and hinder further development.

Although we have repeatedly proposed a more general alternative term, the Distributed Netbook Technology (DLT) system, we recognize that both “blockchain” and “DLT” have been established as covered terms that are often used interchangeably. Therefore, we will use them in the rest of this report.

Distributed ledger technology (DLT) is a subset of distributed systems

Distributed ledger technology (DLT) can be thought of as a subset of a distributed system, a system of multiple independent components (such as computers) that communicate with each other. They are typically based on a peer-to-peer (P2P) architecture in which computers (nodes) exchange messages directly between each other without having to go through a central server.

"The design of the DLT system for operation in hostile environments"

Most distributed systems consist of multiple nodes that collectively store and process data, but are actually owned or controlled by a single entity. The difference between DLT and these systems lies in the lack of a central mechanism for how the coordination nodes agree on the state of the system:

The DLT system is specifically designed to operate under hostile conditions, so they can remain operational even when there are unreliable components (such as hardware failures, connection problems) and malicious behaviors that attempt to compromise the system.

Key features of the DLT system

The DLT system is a multi-party consensus system that enables multiple untrusted entities to agree on a transaction order in a confrontational environment without relying on a central trusted party.

The DLT system needs to ensure the following five attributes:

  1. Shared record keeping : enables multiple independent entities to provide data entry and participate in the creation of new records;
  2. Multi-party consensus : requires multiple independent entities to collectively agree on the order of transactions without a central agency;
  3. Independent verification : Enables each participant to independently verify the status of their transactions and the integrity of the system. This also includes detecting unauthorized changes applied to the record in a simple manner;
  4. Tamper proof : Allows each participant to detect consistent changes applied to trivial records;
  5. Tamper-proof : It is difficult for a single party to unilaterally change past records (ie, transaction history).

It is worth noting that the distributed ledger technology system is dynamic and constantly evolving. This continuous conversion may affect system characteristics and may therefore affect the core features highlighted above. As we will see, for a variety of reasons, many of today's "distributed ledger technology systems" are operating in a closed, non-powered, secure environment. More information can be found in the Distributed Book Technology System: 2018 Conceptual Framework Report.

License vs. no license required

The DLT system can be public and unlicensed: this means that anyone can freely join, use, and leave the network, and participate in the network consensus process without requiring permission. Because these systems have dynamic membership (ie, the number of peers on the network is unknown and may change), they rely on a combination of economic incentives through their native tokens and game theory to work properly and remain secure.

The DLT system can also be private and licensed: access to the network is limited to a limited number of network gatekeepers. Different levels of permissions (such as the right to participate in the "Network Consensus") are assigned to network participants. These systems have fixed members (ie, the number of peer nodes on the network is known): since all participants have been identified, a contractual agreement can be established between them to punish misconduct.

The remainder of this report will focus on private and licensed DLT systems operating in an enterprise environment. Decompose "consensus"

Common blockchain arguments usually involve the concept of “consensus,” but there is usually a lack of clear description of the nature of the agreement, ie, what is the consensus. We introduced two related but different types of consensus concepts in Figure 1.

Figure 1: There are two types of consensus for each DLT system

P1

First, network participants and other stakeholders need to agree on the rules set of the management system (represented in the agreement by consensus rules) and the appropriate process for applying changes to the rules. Social consensus is not only about the governance of the system, it also involves implicit agreements between stakeholders on the nature of the system (and related features). Any party that is remotely involved in the system—either directly or indirectly—can play a role in social consensus, albeit with varying degrees of impact on the process.

"Social consensus" rules, "network consensus" implementation rules

Once stakeholders agree on the nature of the system, including its key attributes, network participants need to agree on the system-generated records (the content itself).

Network consensus refers to the process of resolving potential conflicts within a P2P network boundary that may be caused by multiple valid but conflicting ledger entries.

This consensus is enforced by block producers, limited to the order of transactions, and operates within the community-defined set of rules.

Social consensus can always override the network consensus

Although different, the two types of consensus are closely linked. Social consensus implies a network consensus that generates the latest system state that is universally accepted by all network participants and external stakeholders. Therefore, it must be pointed out that social consensus can always override the network consensus, because changes in rules may invalidate the transaction ordering decisions made by miners and other block producers.

A functional DLT system should have an appropriate system of checks and balances to prevent a single entity or a group of participants colluding with each other to dominate social or network consensus, as this will create a single point of trust and thus failure. For example, a block producer can be inspected by fully verifying the threat to reject a node (auditor) that digs a block that does not comply with the protocol rules.

1, 2 mapping different interpretations

Unclear terms, blurred boundaries, and lack of reliable definitions have caused the terms "blockchain" and "DLT" to actually lose their meaning. This development not only affects effective communication and understanding, but it is also possible to confuse different concepts under a single general term, making it difficult to objectively evaluate and compare different projects.

Introduce two main explanations

In order to make these terms meaningful again and easy to discuss, we recommend dividing all uses of the tagged “blockchain” or “DLT” project into two categories:

1. Use blockchain as a multi-party consensus system :

A joint service network that is not controlled by a single entity; operates in a hostile environment and occasionally resolves conflicts around the order of transactions;

2. Use the blockchain as a "mechanical" :

There are no or limited multi-party consensus concepts, and there are no potential differences in transaction sequencing; often as a catalyst to change various systems and processes.

The first category: multi-party consensus system

This category refers to the DLT system defined on page 12. The key distinguishing feature is the concept of multi-party consensus: several independent entities need to agree on a set of shared data without relying on the central coordinator, and occasionally the differences in the order of transactions are resolved through network consensus (see page 14). .

In the context of a licensed network, these systems are best described as a federated service network that distributes control among network participants so that social consensus and network consensus are not easily controlled by a single entity. However, as explained in the following sections, most networks initially adopted a more centralized approach when they were released, for a number of reasons. Therefore, we recommend that these networks be referred to as potential DLT or enterprise blockchain systems because they are not compliant when analyzing them, but they have a clear and firm roadmap to gradually distribute control. To effectively become a "real" multi-party consensus system.

The second category: "blockchain model"

This category contains all projects and networks called “blockchains” or “DLTs” that do not form the previously defined multi-party consensus system.

Instead, they typically use several component technologies used by the DLT system to meet the needs of a given business case. Figure 2 outlines the key component technologies that are often used separately (encryption primitives and concepts from distributed systems.)

Figure 2: Component technology used in the "blockchain"

P2

The “Block Chain Memetic” project is designed to serve the following business cases:

a) the concept of a multi-party consensus that is not required or available; and

b) It is impossible to disagree on the order of transactions (such as events, facts).

This means that by definition, these items cannot display key features defined as blockchain or DLT systems.

What is the technical principle behind the “blockchain model”?

Projects and networks may use concepts and tools related to blockchain, but they do not intend to support multi-party consensus for a variety of reasons:

  1. A single “trusted” source: an incident-only event log that produces a single, consistent “truth” source for all participants;
  2. Advantages of component technology: The application of some cryptographic primitives and the borrowing of specific concepts from distributed systems can provide significant benefits to existing products (eg, through independent verification, integrity checking, and better authentication to improve security, Enhanced resistance and additional password guarantee);
  3. Off-the-shelf blockchain protocol framework: off-the-shelf blockchain software libraries and platforms that are readily available, usually free and open source, to quickly build new systems and networks with different privilege configurations;
  4. Architectures that are suitable for some purposes: Solving specific use cases from a practical perspective may mean choosing elements that are distributed in some ways but concentrated in other areas. Enterprise blockchain deployments often force pragmatic rather than idealistic decisions.

A common feature of all "blockchain memes" projects is that they use the "blockchain" or "DLT" brand when promoting their business products. In addition, these projects seek to generate a single, consistent source of records that stakeholders can access, even if they can only be provided by a single control entity through a trusted application programming interface (API).

The existence of a single, shared authoritative record view provides a broad opportunity for better data management and sharing, and automation of cross-entity business processes and workflows.

However, there are also huge differences between the projects belonging to this category: as shown in Figure 3, the range of project types with different objectives, design requirements and implementations makes the difference in this category quite large.

Figure 3: "Blockchain Memetic" spectrum

P3

“Blockchain Memes” may be understood as an influential driver of organizational change to promote the development of common data standards between entities, improve the availability of general data, and reduce the number of individual books between parties The steps required for the system.

Therefore, it can be considered as an effective coordination mechanism to coordinate the interests of different stakeholders who would not have cooperated on such undiscussable matters.

“Blockchain Memes” act as a cause of behavioral change at the internal and industry levels

Therefore, “blockchain memes” can serve as a powerful catalyst, first encouraging entities to rethink existing infrastructure and business processes, and ultimately to a new core industry infrastructure that is shared by different parties. , sharing and operations.

The “blockchain meme” helps resolve political barriers to creating industry utilities and drives behavioral change.

These systems also provide additional password guarantees (independently verified by network participants) based on design decisions and network configuration.

Why is the distinction important?

An analysis of more than 60 active enterprise blockchain networks deployed in a production environment shows that more than three-quarters of the projects fall into the “blockchain meme” category (Figure 4). Only 3% of the networks meet the standards of a multi-party consensus system, and one-fifth of the projects can be considered as potential distributed ledger technology systems, which are gradually moving toward eliminating single points of failure and control.

Figure 4: Most active enterprise blockchain networks can be classified as "blockchain memes"

P4

(Note: The data is based on CCAF's 67 active enterprise blockchain network datasets.)

“ Introduction to research samples

All empirical data provided in this report is based on one of the following data samples:

1. Active network example: CCAF provides a list of 67 enterprise blockchain networks in 25 countries around the world. These networks have been deployed in production and are currently active. 2. Survey sample: In addition to active network samples, this The analysis also combines data from three online surveys conducted by CCAF in 2018: a. Suppliers: 60 respondents, including start-ups and large companies from 25 countries around the world; b. Network operators, participants and End users: 56 respondents, including start-ups and large companies from 22 countries around the world; c. Public sector: 45 respondents from 33 countries around the world, including government agencies, public agencies and central banks;

See Appendix I for more information on the sample. The data source will be properly identified below the relevant charts and data points.

The “blockchain memetic” project often has little in common with previously defined DLT systems, but they are often confused with actual multi-party consensus systems. This development has led to the widespread belief that the “blockchain” is a magic bullet that overestimates the actual capabilities of the technology itself. If an internal accounting system operated by a single company can be titled “blockchain”, then the “real” DLT system is correctly evaluated and differentiated for industry peers, regulators, policy makers, and other stakeholders. And the attributes of the "blockchain memes" project (and related trade-offs) become more and more difficult. This, in turn, presents challenges for performance comparisons and assessments of potential applicability and impact.

This is not to say which category is better. In fact, the global impact of the “blockchain meme” project is likely to be greater than the multi-party consensus system. It is only because of the emergence of common data standards for the entire industry that it is likely to bring huge Increase efficiency and create new services and business models.

The application of the "real" multi-party consensus system may still be niche, with little impact except for its narrow use cases.

“Blockchain Memetic” has the potential to significantly shape industrial transformation

However, it should be noted that the classification is not static. We need to understand the business operations and overall strategic constraints that will force the network to start in the form of a “meme” or a potential distributed ledger technology system. Because these systems are dynamic and evolving, their respective analyses should follow.

Section 2: Introduction to the enterprise blockchain ecosystem

2, 1 Explore the ecological landscape

In recent years, a large number of new entrants have emerged in the blockchain ecosystem, and they have launched a number of new projects, products and services at an alarming rate. This may make it difficult for outsiders – and even insiders – to track the proliferation of new products across multiple market segments.

Figure 5 shows the reader a useful lens for exploring and navigating the ever-changing blockchain environment: the framework divides the environment into three interrelated “layers” of creation value (protocol, network, and application layer). It is worth noting that this mental model is not limited to enterprise DLT projects, but can also be applied to open and unlicensed cryptographic asset DLT systems.

Figure 5: Three-layer thinking model for exploring blockchain landscapes

P5

Note: This model is based on the framework originally developed by Colin Platt (2016).

1) Protocol layer: technical components that constitute the "back end"

The protocol layer is a collection of core protocol frameworks, that is, a collection of core technology building blocks that form the backbone of any blockchain system technology. The protocol layer includes infrastructure components on which to build networks and applications. Since the protocol is just code (open source or proprietary), the protocol layer itself does not provide much value if there is no corresponding network.

2) Network layer: a shared service network that produces a single "truth" source

The network layer consists of the actual P2P network, which makes data sharing and verification a blockchain system by connecting participants. The network layer includes all of the business networks that are operated and maintained by multiple participants, with the goal of generating a single consistent "truth" source of shared recordsets. Networks can be built using standard core protocol frameworks such as Hyperledger Fabric, or a combination of modular core building blocks borrowed from multiple core protocol frameworks.

3) Application layer: products and services that create real business value

The application layer can be thought of as the primary user interface for a blockchain network: it includes business applications that connect to existing networks to take advantage of the underlying data record set ("unique source of truth") that is jointly generated and maintained by network members. The application layer is the main driver of business value because it provides tangible products and services to end customers and users.

A "licensed" application that restricts the use of a particular party may also exist on an open, unlicensed blockchain network.

Applications can also be "blockchain agnostic", which means they can be plugged into several separate networks as needed.

Figure 6 illustrates the JPMorgan Interbank Information Network (IIN) as an example of how a three-tier framework can be used as a lens to evaluate different components of a given project.

The first active application, IIN Resolve, uses the network to simplify compliance checks on proxy banking to accelerate cross-border payments. Several other applications are under development, and they all utilize the underlying IIN network.

Figure 6: Using the IIN of JPMorgan Chase to illustrate the landscape framework

P6

2, 2 What are the participants in the industry?

Participant and activity type

The blockchain enterprise ecosystem consists of a set of different actors who perform different activities. Participants and activities can be divided into six categories:

  1. Suppliers design, develop, implement and maintain a technical infrastructure that powers blockchain networks and applications;
  2. The network operator is responsible for managing the blockchain network that is operated and maintained by the network participants ;
  3. Application providers provide end users with an interface to access the blockchain network to achieve business goals that leverage the utility of a single record source.
  4. Many other stakeholders provide tangible services and insights that benefit the entire ecosystem;

Table 1: Outlines the main types of participants in the ecosystem and their associated activities.

P7P8

It should be noted that an entity can play multiple roles and participate in multiple activities. For example, survey data suggests that most vendors do not specialize in one activity, but rather provide a broader range of services, sometimes even as network operators for customer networks. Similarly, an activity can also be performed jointly by multiple different entities (eg, gatekeeper network management and management).

Given that the industry is still at a relatively early stage, the boundaries between different roles and activities are not always clearly separated, and in some cases significant overlaps can be observed. For example, today, the different roles of network operators are performed by the same entity in most cases, and in the future, one can imagine the potential separation of these roles among multiple entities, which will lead to an increase in the dispersion of network governance.

2, 3 industry development

Entry point

The enterprise blockchain industry consists of hundreds of entities that provide products and services, from small start-ups to large established companies. However, not all companies enter the industry in the same way: this study distinguishes between native blockchain entities and non-native entities.

  1. Native blockchain entities : These entities, at least initially, specifically serve the enterprise blockchain market.
  2. Non-native entities : Existing entities have expanded their business offerings to include blockchain related products and services.

With a few exceptions, native blockchain entities are often relatively small start-ups, not native entities, often from larger, existing firms in other industries.

Figure 7: Although the supplier is primarily a native blockchain entity, operators are often from existing ones.

P7

The difference between the types of participants, we can also observe that although nearly two-thirds of the surveyed suppliers are considered as native blockchain suppliers, there is a similar proportion before entering the enterprise blockchain ecosystem. Network operators are also active in other industries (Figure 7). This is not surprising given that most blockchain networks serve specific use cases or industries that have existed before. However, more than one-third of network operators have specifically set up a new entity to realize their value proposition.

Industry growth

Although the growth of the enterprise blockchain industry is difficult to quantify due to the lack of reliable data sources, we can use agents to provide approximate estimates. One of the agent data is the number of full-time employees employed by the blockchain company.

Figure 8 shows that since 2016, the total number of full-time employees in the enterprise blockchain industry has grown significantly, with suppliers having the highest growth rate among all types of participants. However, investigators also warned that the market's demand for skilled blockchain talents is growing rapidly, but supply growth is insufficient, leading to a growing shortage of skilled labor.

Figure 8: The total number of full-time employees active in the blockchain industry has grown significantly, with suppliers having the highest growth rate among all participant types.

P8

Note: Based on CCAF survey data, full-time employees who are not fully focused on blockchain activities have been removed from the calculations.

Surprisingly, in general, the growth rate of suppliers in the first two quarters of 2018 is almost the same as the growth rate for the whole year of 2017. During the coverage period, the total number of full-time employees focused on the blockchain employed by network operators has also increased significantly, indicating that the industry is increasingly adopting actual business networks deployed in production.

At the company level, we can also observe a similar pattern: since 2016, the average size of blockchain companies has grown significantly, with entities as suppliers hiring more people (median) and the highest Growth rate (Figure 9).

Figure 9: Vendor and network operator companies are growing significantly

P9

Note: Based on CCAF survey data, only companies that have been operating throughout the coverage period are considered.

However, there are also striking differences between company sizes in the same category: although some suppliers have fewer than 5 blockchain employees, the largest organization has more than 2,000 employees focused on the blockchain, higher than 300 in 2016 and 800 in 2017. The size of network operators is often much smaller: the largest entity employed about 70 people at the end of the second quarter of 2018, up from 35 in 2016 and 50 in 2017.

Alliance and ecological construction

Blockchain networks require cooperation between different parties to maximize their potential. Therefore, we can see that a series of new initiatives have emerged in recent years, trying to focus attention among the parties concerned and work together to use this technology to establish a network effect.

These initiatives can range from simple partnerships between companies and projects to strategic investments and large technology and industry alliances (Figure 10). Survey data shows that the relationship between suppliers and projects is the main type of cooperative activity, although more than 40% of suppliers participate in one or more alliances.

Figure 10: Direct partnership is the preferred mode of cooperation for suppliers

P10

Note: Based on CCAF survey data

Another way to look at the enterprise blockchain ecosystem is to think of it as a collection of smaller (sub) ecosystems that are built around specific platforms, industries, geographies, and use cases. Most of these initiatives are divided into the following three categories:

  1. Technology-centric : The true ecosystem is built around a (mainly open source) core protocol framework and related technology platforms, including a vibrant developer community and hundreds of active contributors. At the time of this writing, the Linux-led Hyperledger ecosystem and R3's Corda platform seem to be the most popular for members and active contributors. Standards bodies such as the Ethereum Enterprise Alliance (EEA) may also fall into this category, although they do not provide specific authorization for technical solutions.
  2. Industry-centric : There have been dozens or even hundreds of industry initiatives that provide relevant industry participants with a way to pool resources to explore blockchain-based use cases and establish common data standards and governance structures. An example of this initiative is the insurance industry's B3i alliance;
  3. Geographically centric : Some initiatives also focus on specific geographic regions or countries to facilitate local development and implementation of blockchain networks. These initiatives can be applied at different levels: for example, regional, national, state, and municipal. For example, the Alastria Network in Spain, the Argentine Federal Blockchain Network, the Australian Department of Industry, Science and Technology, and the Block Link Map of the Ministry of Trade, Tourism and Investment, and the Dubai Blockchain Strategy, and other government-funded blockchain initiatives. .

Section III: Adoption of blockchain

3, 1 transferred to production

Since the release of the first global blockchain benchmark study in September 2017, the corporate blockchain landscape has undergone major changes. Two years ago, although more than half of the surveyed vendors had production-ready technology platforms, only 8% of surveyed operators said they were running active networks: most of them were involved in early testing, proof-of-concept and experimental projects.

The situation in 2019 looks completely different: although it is almost impossible to track all production deployments around the world, we estimate that hundreds of active blockchain networks around the world are being used in production environments across industries and industries. As shown in Figure 11, most of the covered blockchain networks were put into production in 2018 and early 2019.

Figure 11: Active blockchain network deployed in mid-2018

P11

Note: Dataset based on CCAF 67 active enterprise blockchain networks

Therefore, it can be said that the DLT has been separated from the propaganda and experimental stage to a certain extent, and it is slowly entering the production stage. However, we can also observe the huge differences between networks, which will be mentioned in the rest of this section.

3, 2 project life cycle

Blockchain networks and projects typically go through several stages until they can run in a production environment. The typical life cycle of a production network can be divided into four main phases: preliminary exploration, proof of concept (PoC), trial/trial, and production .

An empirical analysis of more than 60 active networks shows that the intermediate time between launching PoC and ultimately deploying the network in production is approximately 25 months, or approximately two years (Figure 12). Among them, the transition from the PoC phase to the advanced test phase accounts for more than two-thirds, while the transition from the experimental phase to the production phase is generally around half a year (median).

Figure 12: The four main stages that an enterprise blockchain project typically goes through

P12

Note: Dataset based on CCAF67 active enterprise blockchain networks

Looking closely at the total scope, we can see the main differences between the projects: although some enterprise blockchain networks have completed each of these phases in just three months, other corporate blockchain networks are in each It took more than two years for each phase. Scale and ambition play an important role in this: small projects with a limited number of participants and relatively simple applications can be launched within six months of launching the PoC, with complex applications involving multiple participants, full of regulation and law Large projects with obstacles take more than four and a half years to be put into use .

Blockchain project is a journey that takes a long time frame

This supports the view that enterprise blockchain projects are journeys involving core market infrastructure transformations, which naturally require a longer time frame and more patience than simple application development. In addition, successful web publishing does not immediately translate into commercial benefits: most active networks have not seen widespread use after deployment to production, which can take years to observe significant impact .

origin

DLT is often considered a “team movement,” which requires multiple collaborative engagements. However, without strong leadership support and major persuasion efforts, it is difficult to initiate an initiative that creates sufficient momentum among the participants, namely effective coordination. Our empirical analysis identifies three main innovation models: founder-led, alliance-led, and government-led . Given the above challenges, it is not surprising to see that more than 70% of the blockchain network was initiated by an entity (founder-led) that effectively assumed the leadership responsibility of the project (Figure 13).

Figure 13: The overwhelming majority of active networks originate from a single entity initiative

P13

Note: Dataset based on CCAF 67 active enterprise blockchain networks

However, the boundaries between the initiating models may be somewhat inconsistent: as these projects go through multiple phases, different models may be applied at different stages, and new participants are added to them. For example, some projects were originally led by founders, but as competitors join the program, the new structure may evolve into a broader alliance and may evolve into a fully equity-based joint venture. Similarly, a government agency can start a project by bringing together multiple national entities that ultimately take over the project and turning the project into a business-managed enterprise. Section 4 discusses the advantages and limitations of each of the origin models in more detail.

3, 3 project scope and industry application

In recent years, companies and institutions in almost all sectors and industries around the world are exploring the potential of blockchain technology. However, no other industry is close to the financial and insurance industries in the deployment of enterprise blockchain networks: almost half of the coverage networks are initiated by financial institutions (Figure 14), accommodation and food services, healthcare and social assistance. Then ranked in the second echelon, each accounting for 6% of the share.

Figure 14: Nearly half of the active blockchain network is initiated by financial and insurance institutions.

P14

Note: Based on the CCAF's 67 active blockchain network datasets, the domain has been classified according to the North American Industry Classification System (NAICS).

These findings are largely consistent with the conclusions of our 2017 study, which shows that the banking, financial, and insurance industries are related to 42% of all proven use cases. An interesting difference has emerged in the public sector: although 13% of the use cases explored in 2017 were attributed to the public sector, only 3% of active blockchain networks were related to the government in 2019. This suggests that some projects may have stopped or are still under development.

The use cases for identifying a given blockchain network are not always simple

The blockchain network facilitates the secure exchange of information between a group of participants and enables data validation and integrity checks through consistent audit trails. Thus, a particular use case for a blockchain network corresponds to the business domain in which shared data is generated. Interestingly, it is difficult to determine a specific use case for an active blockchain network because its mission statement is still relatively vague and extensive. In addition, the blockchain network can act as a shared data platform that supports multiple use cases.

The data suggests that most active networks are initiated with a specific use case, although it is often emphasized in the official roadmap to gradually expand the scope to include other industry-related usage intentions. Figure 15 shows the use case for overlay production network support: tracking projects in large and complex supply chains (such as food, gemstones, containers) is the most prominent use case for active blockchain networks, followed by market infrastructure for trading tokenized assets.

Figure 15: Supply chain tracking is the most important use case in an active blockchain network

P15

Note: Dataset based on CCAF 67 active enterprise blockchain networks

Key value proposition

Network participants and application providers use enterprise blockchain networks to extract business value for their operations. The blockchain network can provide value to the business in three ways:

(1) Reduce costs by eliminating adjustment steps that can be avoided between company books;

(2) generate revenue by providing new services (by accessing shared data), and

(3) Create a new market model and a type that is only enabled by a shared network that did not exist before.

The data shows that currently more than two-thirds of active blockchain networks are primarily designed to reduce the cost of network participants and end users (Figure 16). Interestingly, the difference between the founder leadership and the alliance leadership model can be observed: 84% of the former is focused on reducing costs, while only 40% of the latter has established a major goal of reducing operating costs, 20% of which is The foundation for a new utility that creates a new market model.

Figure 16: Cost reduction is the primary goal of the current enterprise blockchain network

P16

Note: Dataset based on CCAF 67 active enterprise blockchain networks

This situation is evidence that the enterprise DLT ecosystem is still in the early stages: most production networks are still in the process of eliminating reconciliation efforts to simplify the exchange of information between participants and to optimize existing workflows. As the system matures, the value proposition of these networks may gradually shift from simple cost reduction to incremental revenue generation over time.

In the end, we may see new products, services and business models emerge.

Network usage

Some readers may be surprised that the enterprise blockchain network can be limited to internal use: in fact, 5% of the production network is dedicated to use within the boundaries of a company or group of companies (Figure 17).

Nonetheless, most networks are designed to be shared between different non-affiliated entities in order to cross organizational boundaries.

Figure 17: Most active enterprise blockchain networks are designed for sharing between partners

P17

Note: Dataset based on CCAF 67 active enterprise blockchain networks

A study of the types of shared use network participants reveals interesting insights: in the current form, more than half of the production networks are established between business partners, while only 19% of the production networks are under the participation of direct competitors. Started. And a quarter of the coverage network has a combination of competitors and partners, and it is not easy to classify considering that they occupy different positions in the value chain (downstream and upstream).

That is to say, in the current form, most networks are built between business partners rather than competitors .

A typical example, competing entities work together to develop a shared industry utility, examples of which financial institutions are looking for secure exchange of relevant information and automation of existing workflows.

In contrast, networks that operate between business partners typically exist in the supply chain, which involves existing networks of collaborative manufacturers, suppliers, and distributors.

The types of participants also vary depending on the originating model: in the network dominated by the founders, a leading entity with a dominant market position is primarily a board business partner (65%), compared to a network led by the alliance. Often dealing mainly with competitors (73% are divided into direct competitors and indirect competitors, see "Other").

A common approach to a federated network operated by direct competitors is to establish a joint venture between the participating entities.

3, 4 network design (protocol framework market share)

Currently, companies and organizations can choose a broad core protocol framework as the technical foundation of their network. However, the data shows that there are only four protocol frameworks in the market that dominate: Hyperledger Fabric has the largest share, and nearly half of the active enterprise blockchain networks use this protocol framework, followed by R3's Corda. , and Coin Science's MultiChain (Figure 18).

Figure 18: Hyperledger Fabric is the most used protocol framework in a deployed enterprise blockchain network

P18

Note: Dataset based on CCAF 67 active enterprise blockchain networks

Another survey data confirms this situation: the hyperbook Hyperledger (where Fabric is the flagship protocol) is the most commonly supported protocol framework for integrators and software development platforms, with a ratio of 53%, followed by Corda (35%), MultiChain (32%) and Quorum (26%).

This distribution also shows that many integrators focus on a set of protocol frameworks rather than a single code base.

Network operators and participants typically experiment with multiple protocol frameworks before choosing a particular platform to build their network. In fact, the survey data shows that 26% of operators have switched the protocol framework during the PoC phase during several tests. However, more than half of the surveyed entities also indicated that they will continue to use the framework originally selected.

Figure 19: Supplier maturity is considered the most important protocol framework selection criteria

P19

Note: Based on CCAF survey data

Network operators and participants select the appropriate core protocol framework based on a set of selection criteria. As shown in Figure 19, the maturity and product readiness of enterprise blockchain suppliers are considered to be the most important factors in the decision-making process. The network operators surveyed also listed performance and scalability as key criteria, and while these are often the result of trade-offs, it is difficult to compare directly (see the Calls box). The selection of standard rankings largely reflects the perceptions of ecosystem participants about the remaining key challenges of the wider use of blockchains (see Appendix III).

" The problem of trade-offs

Each network is optimized for the delivery of specific business objectives based on a unique set of architecture and design choices. Each design configuration is a conscious trade-off between various attributes, the most common trade-off being performance improvement at the expense of "decentralization." In a particular environment, certain attributes may be preferable to other attributes, but we must understand that they are always implemented at the expense of another attribute. This makes it difficult to directly compare the performance and scalability of blockchain networks, as they usually choose different compromises to produce different features.

Decentralization and decentralized control

A blockchain is a tool that distributes control of a shared recording system among multiple participants. Therefore, in theory, no entity can exercise full authority over the system unilaterally. However, in practice, the traditional business environment requires a degree of safeguards to ensure that potential problems are quickly resolved without seriously affecting operations. This concept generally assumes that there is an entity or a small number of entities and then gives them some power.

The rest of this section will explore the degree of decentralization in existing enterprise blockchain networks by considering network consensus (automated agreements for trading orders) and social consensus (governance).

a) Network consensus

Two major participants participated in the network consensus:

  1. Auditor : A complete node that independently verifies transactions and system status;
  2. Block producer : An entity that submits a transaction to a global ledger by establishing a fixed order to avoid double spending;

Determining the network resistance of a blockchain system requires evaluating two dimensions: first, which entities are at the operational node (control), and second, where these nodes are hosted (managed). Figure 20 provides a visual representation of this mental model.

Figure 20: There are various options for controlling and hosting full nodes

P20

If a network participant operates a complete node on its own (that is, has full control and access to the node), they can only benefit from being able to independently verify transactions and system status; otherwise, they are completely dependent on third-party service providers. And third-party service providers have been awarded audit functions.

When nodes are hosted locally by network participants or hosted in multiple cloud environments, the network can be more resistant to failures and external attacks, although the various configurations and settings do play an important role because they are in different organizations. There may be a big difference.

It is very difficult to get reliable data about the network composition of an active network.

It is worth noting that with regard to the exact network composition of the active networks covered, the available information is limited: few people disclose the number of complete nodes and block producers and the hosting environment. Moreover, unclear terms often lead to confusion, such as confusion between "users" and "participants" or "nodes." For networks where data is available, the result is that the number of nodes can range from a single node to more than 70, and in certain cases, all nodes have more than 200 members.

In an active corporate network, it is more difficult to obtain reliable data on the number of producers in independent blocks, which indicates that many networks have so far relied on individual block producers. It should be mentioned that this is consistent with the examination of the details of the use cases, which reveal the fact that there is little disagreement in the ordering of transactions, that is to say, these enterprise blockchain networks lack multiple block producers.

Most active enterprise blockchain networks do not distribute control among participants, but are effectively controlled by a small number of participants.

Our results show that most of the enterprise blockchain networks in production are currently hosted in a single cloud environment, where nodes are managed by a third-party service provider (platform host) that is also responsible for generating blocks. operating. Project members can then connect to the platform host through a trusted API. It was found that in these cases, 69% of the platform hosts are also network technology partners (ie software vendors).

However, there are exceptions: some networks do offer members the option to run their own nodes (local or in the cloud), although it is interesting to note that most participants choose to delegate their audit authority to a third-party service provider. Although this does not seem to have reached the purpose of the blockchain, in the initial stage, there are legitimate business reasons for doing so.

b) Social consensus

Although the diverse network structure and composition can ensure that no one participant can control the transaction sequencing process, we mentioned in the first section that social consensus can always override the network consensus by changing the institutional rules. Therefore, when assessing the degree of decentralization, the governance of the blockchain network must be considered.

Our empirical analysis shows that more than 80% of active enterprise blockchain networks are currently managed by a leading entity (Figure 21). This leadership entity is responsible for coordinating protocol changes, rights management, and joining policies. Governance structures may vary from project to project, while other participants have varying degrees of decision-making power.

Figure 21: Four out of five active networks are managed by leading participating entities

P21

Note: Dataset based on CCAF 67 active enterprise blockchain networks. For more information on the differences between social consensus and network consensus, see Figure 1 on page 14.

Leading entities often take the lead in building networks: in fact, 96% of the founder-led networks are managed by the same entity that initiated the program. In contrast, networks from alliances are often managed more through joint ventures (33%), although one alliance member is still leading the network management of more than half of the coverage networks.

" Case Study: We.Trade

We.Trade is a blockchain network initiated by the International Banking Union to initially provide open account trade finance for SMEs in Europe.

The official We.Trade project was launched on October 17, 2017 and was originally a digital trading chain consisting of seven international banks (Deutsche Bank, KBC, RabBank, Industrial Bank, NATIXIS, UnReCiDIT and HSBC). On July 3, 2018, two other banks (Santander and Nordic Bank) subsequently joined the alliance and joint ventures and announced the first transactions. In October 2018, three banks (UBS, Erste Group and CaixaBank) from the BATAVIA Alliance led by competitor UBS joined the alliance. In addition, European banks allowed We.Trade to expand to Greece, bringing the total number of Union Banks to 13 homes.

Each of the 13 participating banks provides access to SME customers through its commercial banking platform. This network enables SMEs to find trusted trading partners in other countries that have been reviewed and verified by participating banks in the local area. Since all the steps of the transaction are submitted to the blockchain book, the payment notice is automatically carried out through the smart contract, so the SME can be more confident to get paid at the pre-agreed time. We.Trade's motto "More trust, more trade" is the proper embodiment of the network's revenue-generating goals: the more trust between SMEs, the more trade will increase, which will lead to more wealth creation, and In turn, create more income for participating banks.

The We.Trade network is built on the Hyperledger Fabric core protocol framework and is operated by the software and service provider IBM on behalf of the We.Trade joint venture. Initially, banks can choose to run nodes locally or deploy nodes in the cloud. In the early stages of planning, two of the first seven banks plan to run their nodes locally; however, all subsequent banks decided to run their nodes in a cloud environment, with the core infrastructure running on the IBM blockchain platform. In 2016, the original digital trade chain PoC led by KBC was based on Ethereum. Since IBM was chosen as the platform developer, it finally chose Hyperledger Fabric as the base platform.

Social consensus and governance

Because each bank has different requirements in terms of architecture, security, and compliance, the design meeting quickly became very large.

This led to the existence of We.Trade as a joint venture based in Dublin, with 12 of the 13 banks becoming shareholders. Interestingly, Eurobank, which recently joined the network, is a licensee to the We.Trade platform and not a shareholder.

Description of public sector blockchain activities

In recent years, central banks and other public sector institutions (including local municipalities and governments, national ministries and agencies, and international and multilateral organizations) have become increasingly interested in the potential of blockchain technology. However, it seems that the initial enthusiasm has cooled down recently, and people are soberly aware that the actual implementation process takes time and is often full of obstacles and challenges.

Compared with the 2017 survey data, the number of institutions still in the early stages of exploring technology and conducting research and research has been greatly reduced, but there has been no meaningful growth in the conversion to POC or actual trials. This observation also applies to the central bank and OPSI, although the findings confirm the findings of the 2017 study that the central bank is more conservative and less risky than OPSI in experimentation and advanced testing. In the past two years, only a few projects have been actively promoted and made significant progress, indicating that the trial has been basically completed.

Figure 22: Since 2017, there has been a slight change in the focus of the explored use cases

P22

Note: Based on CCAF survey data

A comparison with the 2017 survey data revealed a slight change in the focus on proven use cases (Figure 22). For example, the central bank’s interest in using the blockchain network to issue and manage central bank digital currencies has declined significantly.

This finding suggests that internal research and preliminary experiments suggest that alternative technologies may be more suitable for central bank digital currencies, which confirms the recent announcement by some central banks to abandon the already launched CBDC project . In contrast, blockchain technology remains a popular candidate for improving the resiliency and robustness of critical payment infrastructure such as instant payment settlement systems.

At the same time, OPIS under investigation has shifted the focus of its blockchain activities from ownership and business records management to personal records management (such as birth and death certificates) . In addition, blockchain technology, as a potential tool for developing new payment and value transfer systems, is being explored more and more, which will increase transparency and promote companies and institutions by providing auditable verified records. Regulatory compliance .

Section 4: Blockchain strategy and business model

4, 1 User Motivation: Expected Benefits

Since the beginning of 2015, the ubiquitous blockchain hype has boasted countless benefits that organizations can expect when adopting blockchain technology. But is there a real driving force for corporate and institutional interests in exploring the blockchain as a new tool for its operations? According to survey data, the main motivation is to generate revenue by providing new products and services (Figure 23). In addition, the prospect of improving process efficiency and associated cost savings also encourages organizations to begin exploring blockchains.

Figure 23: New revenue-generating potential is a key driver of the enterprise blockchain strategy

P23

Note: Based on CCAF survey data.

This is consistent with the findings of Section 3, which shows that the initial focus of the active enterprise blockchain network is primarily on cost reduction, but will gradually shift to providing new products and services with a view to generating additional revenue. In addition, these data confirm our initial assessment that blockchain technology is seen as a “team movement” that develops its full potential through multi-faceted collaboration beyond organizational boundaries. Interestingly, network operators and participants report that “the general reluctance to change established business processes” remains a key challenge in the blockchain (see Appendix III), which indirectly confirms that “blockchain” is more like A powerful catalyst for organizational change, not a new technology.

Finally, it is worth noting that one in seven respondents said they started the blockchain strategy because they were worried about losing their competitive advantage. However, it is unclear whether this means that people are really worried about missing a potentially innovative pilot train, or whether it can be interpreted as a corporate “innovation theater” for marketing purposes to influence external perception.

Internal decision

Blockchain strategies and projects can come from different levels of the company's hierarchy. Perhaps not surprisingly, most of the organizations surveyed said that the internal innovation department has been playing an important role in developing an organizational blueprint for incorporating blockchain technology into its operations (Figure 24).

Figure 24: The enterprise blockchain project mainly originated from the internal innovation unit of the enterprise.

P24

Note: Based on CCAF survey data

It is worth noting that nearly half of the respondents also mentioned that the CEO has been paving the way for the company to participate in blockchain activities, confirming the findings of another study, 53% of executives from all industry companies Managers attach great importance to blockchain activities. Interestingly, 10% of respondents indicated that internal blockchain activities were initiated by external suppliers.

4, 2 Strategy implementation (market approach)

When organizations identify viable blockchain business cases, they will face two market approaches: one is to take the lead in launching a new network, and the other is to join existing blockchain initiatives and networks.

Each method has different strengths and opportunities, but it also has shortcomings and challenges, as shown in Table 2.

Table 2: Analysis of the advantages and disadvantages of different blockchain market methods for new entrants

P25

Choosing the right approach depends on a variety of factors, including internal factors (such as business case, resource requirements and availability, acceptable time horizons) and external factors (such as industry context, the existence of competitive platforms). In addition, an organization's participation in the project plays an important role: the earlier the project, the new entrants can give more flexibility and decision-making power; however, there may be more costs in network development and governance arrangements.

Incentives: Allocating costs and benefits

As with other major technology and infrastructure upgrades, the start-up and maintenance of an enterprise blockchain network can involve significant costs. What are these costs and who will bear them?

Cost factors can be divided into four categories that are closely related to the project phase: initiative initiation, project development, governance arrangements, and operational activities . Depending on the mode of initiation, the cost level and allocation may vary widely: for example, in a coalition-led project, costs tend to be shared among the members of the alliance, and in the network led by the founder, the initial investment is related to the project. The cost is usually borne by the lead entity (Figure 25).

Figure 25: Cost factors and allocations depend on the network model

P25

A major consideration in relation to cost factors and ultimately commercial interests is the network effect: blockchain projects can only realize their full potential when all parties involved in a transaction are on the same network. Choosing the right incentives based on the network model is critical to encouraging other organizations to join this initiative.

If the leading entity has a dominant market position and can take advantage of its position in the downstream and upstream members of the value chain, then the network dominated by the founder can create network effects relatively easily. Alliance-led networks should adopt appropriate business models to avoid over-biasing founding members and early adopters when encouraging others to join.

" Case Study: TradeLens

Launched in January 2018, the TradeLens project is a blockchain network jointly launched by Danish logistics and transportation giant Maersk and software service provider IBM to transform the ship supply chain. Given that Maersk controls about 25% of the world's container traffic, the TradeLens initiative is an example of a blockchain network led by a founder that leverages its vast market share and influence to achieve alliances without the need to form alliances. A minimally viable ecosystem. Then at the end of 2018, media reports showed that other major marine transportation companies refused to join the network led by their main competitors. In May 2019, MSC and CMA CGM announced their participation in the TradeLens network, which enabled the network to support a container shipment ratio of just under 50%. In July 2019, the world's fifth and sixth largest shipping operators Hapag Lloyd and Ocean Network Express joined the network, further accelerating the adoption of TradeLens.

This case study illustrates the challenges faced by the founder-led business model and the time required to reach a critical point in a multi-party network. It also shows that, in practice, organizations that are aware of the situation tend to join existing initiatives, even if leaders and managers are their main competitors, rather than creating their own projects.

Supplier relationship

From a technical perspective, network operators and participants have multiple options for network development and maintenance. While some prefer to develop internally and reduce reliance on third-party vendors, most of the entities surveyed choose to establish relationships with third-party software or service providers: in fact, about 60% of entities say Their POC was the result of a combination of in-house development and outsourced development. Of course, this was different in 2016, when most organizations were experimenting internally with freely accessible open source protocol frameworks.

In addition, network operators and participants often choose the vendor services they originally signed up for. According to the survey data, 61% of organizations engaged in multiple POCs or other tests and trials chose to continue to work with the same supplier, while only 17% chose to switch completely to another supplier (Figure 26).

These findings underscore the importance of suppliers building and maintaining good relationships with potential customers from the start, as customer retention rates appear to be relatively high.

An explanation can be found in the specific protocol framework of a particular vendor (for example, IBM is primarily doing Hyperledger Fabric, ConsenSys is doing Ethereum variants), and the trust of vendors who have successfully delivered projects and PoCs.

Figure 26: Blockchain supplier's customer retention rate seems relatively high

P26

Note: Based on CCAF survey data

The relationship between network operators and software vendors is also closely related to the project phase: the more advanced the project, the more likely it is to involve third-party software vendors. The respondents mentioned that less than one-third of the pilots and trials were conducted internally without technical assistance or support from external parties. This is similar to traditional software-based projects where companies sign contracts with external software vendors.

4, 3 Evaluation of blockchain activities within the enterprise

Most blockchain experiments, POCs, and experiments have not entered the next phase. Although the launch of the project is usually announced in a big way, including press releases, blog posts, and news, many projects end up with a quiet waiver and are rarely publicly acknowledged. This further fuels the hype, because people artificially believe that blockchain technology has been widely adopted .

Survey data from network operators and participants clearly highlights the main reason behind this: applying blockchain technology to a specific business case fails to achieve tangible benefits (Figure 27). This is not surprising, because many people choose the wrong application from the beginning. In addition, when explaining the benefits that are not realized, respondents often refer to the obstacles of supervision and the difficulty of collaboratively developing business cases across network participants.

Figure 27: Failure to achieve tangible benefits is widely recognized as the main reason for discontinuing blockchain projects

P27

Note; based on CCAF survey data

It seems that quite a few projects have been shut down before they have the opportunity to fully realize their potential: because many active networks and tangible commercial interests are often observed only after the network has been running for a while. This is the result of business model innovation, depending on many unknown parameters and environments. The uncertainty of return on investment associated with blockchain is largely considered by industry participants to be a key challenge for wider adoption (see Appendix III).

Interestingly, more than a third of respondents said that concerns about confidentiality and privacy led them to end their blockchain project, which coordinates key blockchain attributes with the surveyed entity (eg, it is difficult to tamper with history) ) Consistent with the difficulty of regulatory compliance requirements (such as GDPR). However, market sentiment may have changed slightly in recent months after more and more national and regional legislators have enacted favorable legislation.

Overall satisfaction

The network operators and participants involved in the survey, most of the projects they initiated are still in the early stages, waiting for the release of the project's potential after completing the testing phase and being widely used. Nonetheless, survey data shows that two-thirds of organizations engaged in blockchain activities are satisfied or very satisfied with the overall performance and results of their projects (Figure 28).

Figure 28: Network operators and participants, usually satisfied with the results of their blockchain projects and activities

P28

Note: Based on CCAF survey data

Given the high rate of project failures, this self-assessment may initially seem surprising. However, this may also reflect valuable lessons learned from negative outcomes, and it is increasingly recognized that blockchain technology is not a panacea for solving every business problem (although its role is often hyped). These insights gained through self-experiment and practice can add extra value to the organization's overall strategy and avoid improper investment.

Alternative technology

Figure 29 shows that a quarter of network operators have suspended a blockchain project because the technology ultimately proves to be unsuitable for the target business case. When asked if they knew that alternative technologies such as cloud computing or traditional distributed databases might be better suited to deliver the same business case, only 16% were recognized (Figure 29).

Figure 29: One-third of network operators believe that blockchain technology is the best choice for solving their business case

P29

Note: Based on CCAF survey data

One-third of the respondents replied that they believe that alternative technologies do not bring the same benefits, suggesting that they are very confident in the effective capabilities of blockchain technology. 37% of respondents are still investigating alternatives, or are not sure if there are alternatives that offer a greater advantage.

4, 4 supplier strategy (platform openness)

Vendors providing software services face an important strategic decision on the openness of their platforms to third-party developers: the underlying code base can be open source (that is, external developers can freely use or modify the code), Can be proprietary (ie closed source code, externally inaccessible to them). Interestingly, survey data shows that nearly half of the platforms are open source and half are proprietary (closed source) (Figure 30).

Figure 30: Half of the vendor platforms are open to third-party developers, most of which are licensed under the Apache 2.0 license.

P30

From the perspective of the supplier's business model and monetization strategy, both decisions have their strengths and weaknesses: an open platform facilitates the emergence of a community and ecosystem around software projects that encourages experimentation and promotes traditional systems. Integration, but not directly monetization. In contrast, proprietary products can be monetized directly through licensing fees and protect the company's IP, but it can also block potential users.

There are a variety of licenses that can be used to determine the permissions to use and modify open source libraries, while most open source enterprise blockchain protocol frameworks and software platforms are primarily open under the Apache 2.0 license agreement (up to 69%).

Industry goal

From a technical point of view, the blockchain platform is a shared record keeping system that can be applied to any type of information. Given the broad applicability of data management and sharing tools, it is not surprising that 41% of the surveyed blockchain vendors have adopted a cross-industry perspective and positioned their technology platform as a common tool for all industries (Figure 31).

Figure 31: More than half of the suppliers are using their platforms to target specific industries

P31

Note: Based on CCAF survey data

However, the development of some platforms and core protocol frameworks is driven primarily by industry needs, including the introduction of specialized features and capabilities to meet the needs of specific business cases or industries. For example, more than one-fifth of suppliers say their platforms are tailored to specific industries, while banking, financial markets and insurance are the main target industries.

In addition, one-third of suppliers claim to have developed their own platforms for industry-specific ideas while still maintaining the design flexible enough to accommodate the needs of other industries. R3's Corda framework is a good example of this strategic change, which was originally tailored to meet the needs of the financial and banking industries, but has since been repurposed to accommodate unrelated industries and use cases (eg, supply) Chain tracking).

Customer type

So far, large-scale established companies have not only occupied the headlines of enterprise blockchain planning and deployment, but also occupied the customer base of blockchain suppliers. In fact, the surveyed suppliers reported that, on average, 44% of their total customer base consisted of large companies, while SMEs only had 15% (Figure 32). The market pressure of large-scale incumbents to innovate and find new solutions may be an explanation for this difference, and SMEs usually do not face such pressure.

Figure 32: The supplier's customers are mainly based on large companies

P32

Note: Based on CCAF survey data

An interesting observation is that there is no statistical difference between small and medium-sized suppliers, which indicates that the size of the supplier has little impact on the type and nature of the customer. This is an encouraging sign, which means that small service providers will not face too many disadvantages.

4, 5 blockchain revenue model

Considering the relatively early stage of the enterprise blockchain industry, innovative new business models using blockchain networks and related services have not yet emerged. Therefore, the current revenue model is mainly considering software development and maintenance fees: Table 3 outlines the enterprise area. The blockchain industry is currently applied to different revenue models for service monetization and is classified according to the blockchain layer in which the service takes place.

Table 3: Overview of revenue models for different enterprise blockchains

P32

For example, vendors offer a complex enterprise open source protocol framework and provide service layer protocols and advanced support packages to customers. A popular pattern is to provide "Block Chain as a Service" (BaaS), which corresponds to a complete stack blockchain service package from project conception to full deployment, including managed cloud environments and active maintenance of the platform. This model enables customers to configure and deploy a complete network in minutes to quickly prototype and test applications in a sandbox environment without investing significant time, talent, and development costs.

Service providers can achieve revenue by adding fees, platform maintenance fees, upgrades, and operational support costs. Network participants can charge external parties for transaction transactions that process transactions and set up commercial APIs to provide network insights and data services, as well as broadcast transactions to the network. Application providers can charge end users who access network features through third-party interfaces.

Section 5: Looking into the future, what will happen to the blockchain industry?

Since the release of the first benchmark study report at the end of 2017, the enterprise blockchain industry has slowly surpassed the hype cycle and entered the production phase, although many of the currently deployed networks are initially conceived as blocks of a fully distributed multi-party consensus system. Chain systems have little in common (see Sections 1 and 3).

Most institutions say they plan to expand investment

However, insufficiency of experimentation and short-lived PoC have given way to more serious plans involving clear implementation roadmaps and benefiting from long-term implementation commitments. This development is reflected in the projected budget of the network operators and participants surveyed: only 4% said they would moderately cut the budget allocated to their blockchain activities, while more than half plan to significantly increase the block in the medium term. The chain budget is to expand operations (Figure 33).

This observation reflects Deloitte's recent survey of corporate executives who found that most organizations are planning to allocate significant amounts of money to their blockchain strategy.

Figure 33: Most network operators and participants plan to increase budgets related to the blockchain

P33

Note: Based on CCAF survey data

Estimated situation

Blockchain suppliers have become more cautious, although potential customers may significantly increase their budgets and planned investments, and have revised their optimistic adoption estimates starting in 2017.

In fact, more than half of the surveyed entities believe that it takes at least three years to adopt a broader corporate blockchain in the private sector, and 5% even mention that it may never become mainstream technology (Figure 34). Perhaps less surprisingly, estimates of the wider blockchain in the public sector are more conservative, probably based on the premise that the private sector usually precedes the public sector in adopting new technologies.

Figure 34: Suppliers expect blockchain to be more widely adopted in the private and public sectors over the next five years

P34

Note: Based on CCAF survey data

Interestingly, the supplier's view is very consistent with estimates by other public sector agencies (OPSI), which appear to be more optimistic about the wider adoption of the corporate blockchain by the public sector (Figure 35). In contrast, central bank adoption of blockchain technology is very cautious: only one-third of central banks surveyed believe that blockchain technology will play a bigger role in the public sector, despite the estimated timetable of agencies. differ greatly.

Figure 35: Public sector estimates for widespread adoption of blockchain are significantly more conservative

P35

Expected development

It is worth repeating that despite the significant progress made in recent years, the enterprise blockchain ecosystem and industry are still in their infancy. Below we list a range of potential developments that may emerge in the medium to long term.

1. Collaboration between competitors will be enhanced

Blockchain technology is a team movement that requires extensive participation from different participants to build a network effect and ultimately unlock its full potential. Currently, it is often limited to entities within the same trust boundary: existing partners, suppliers, and customers.

We expect to see an increase in cooperation across all forms of cross-trust boundaries (ie, between directly competing entities): survey data suggests that partnerships remain the preferred method for the foreseeable future, with more than half of the network operators surveyed and Participants also revealed their intention to join an existing alliance or launch a new program (Figure 36).

Figure 36: Network operators and participants plan to increase cooperation by building more partnerships and joining alliances

P36

Note: Based on CCAF survey data

2, "blockchain model" hype will cool down

As the hype fades and the industry matures, it is likely that the industry will adjust the terminology to reflect the actual situation. This means that a "real" multi-party consensus system will be semantically distinguished from systems that use component technology and blockchain toolkit elements.

This development will bring long-awaited clarity to the discussion and greatly reduce accidental misunderstandings and confusion. This will make people more aware of the limitations of blockchain technology and, to a certain extent, prevent the unintentional investment of resources into expensive projects that are doomed to fail from the start.

3. Active networks will gradually release control

Many active enterprise blockchain networks were initially highly centralized, but they plan to gradually assign control of different network functions to multiple participants over time as a long-term goal. The main reason for this approach is that these networks operate in a regulated enterprise environment: participating entities usually choose conservative practices during the start-up phase, which involves the existence of safeguards, such as central coordinators, which can be unforeseen Intervention in case.

In fact, many operators report that participating entities first need to get used to the new architecture, familiar with abstract concepts (such as key management, self-hosted infrastructure) until they feel that they are gradually assigning control and taking on greater responsibility. Satisfied enough. This approach enables network participants and end users to better understand the capabilities and attributes of the system when operating in a secure environment. This creates valuable experience that can be used to move forward until all responsibilities related to social consensus and cyber consensus are fully distributed among the participants so that no entity can make a unilateral decision.

4. Interworking network will appear

Currently, there are a large number of enterprise blockchain networks around the world, but they are currently independent and cannot access each other, but to provide more possibilities, which requires connections and interactions between these independent networks. This interoperability requires different technologies to promote.

5, the focus will be more and more transferred to the application layer

Since the release of our first research report in 2017, the research focus of the industry has shifted from the protocol layer to the network layer. Once the major large-scale networks are established, more and more business applications will be built on them to take advantage of network data. We can also reasonably assume that applications will become more and more "book unknowable", that is, connected to different enterprise networks according to their needs, rather than being limited to a single network.

6. Further integration between the private chain and the public blockchain

Although this report focuses only on private, licensed “enterprise” blockchains, we should not lose sight of the public, unlicensed blockchain that operates on the basis of complex socio-economic mechanisms. Some overlap between these two systems, we can already observe: For example, some enterprise blockchain vendors have developed a core protocol framework based on the open source public chain protocol (for example, the bifurcation of the Quorum Ethereum code base, The original Multichain was a fork of the Bitcoin code base).

Figure 37: Nearly half of network operators and participants have involved public blockchain activities

37

Note: Based on CCAF survey data

In addition, more and more network operators and participants are experimenting with public, unlicensed blockchain networks to study how enterprise use cases can benefit from integration (Figure 37). A few large companies have begun to use the Ethereum public chain as a new market infrastructure for issuing, recording and trading financial instruments. 34 other companies have been using the Bitcoin blockchain as a notarized scoring device to safely time stamp files to prove their existence and integrity.

Similarly, enterprise blockchains may find it useful to periodically anchor the latest system state to a common blockchain for additional tamper resistance and independent third-party audits. Therefore, for the foreseeable future, the world of enterprise blockchain and crypto assets is likely to merge to the extent that cross-chain communication and asset transfers will become very common.

7. Asset appreciation

From traditional financial instruments to real estate and art, the wave of tokenization of various assets has begun. For some assets, tokenization can separate ownership, which may help expand the investor base and potentially increase market liquidity. For other uses, tokenization can introduce new economic incentives to influence consumer and corporate behavior on digital platforms. As industry insiders point out, tokenization will drive banks, exchanges, asset management companies and custodians to use the enterprise blockchain platform in the future.

Appendix 1: Research Samples

survey data

The Cambridge Alternative Financial Center conducted three online surveys through a secure online questionnaire from July 2018 to November 2018. The survey is written in English and distributed through three main channels: (a) by email directly to potential survey participants, (b) by sharing social survey links via social networks (eg Twitter, LinkedIn), and (c) With the support of the Blockchain Alliance R3 and the Hyperledger project, the relevant companies were investigated.

Data from more than 160 entities in 49 countries

During the survey, the research team communicated directly with the various organizations to explain the research objectives. The research team collected data from corporate blockchain startups, established companies, central banks and other public sector agencies ("OPSI"). The collected data is encrypted and securely stored and only accessible to authors. All the data involved were processed anonymously and aggregated.

  1. Blockchain supplier survey: 60 companies from 25 countries, including start-ups and mature software vendors.
  2. Blockchain operator and user survey: 56 companies from 22 countries;
  3. Central Bank and Public Sector Blockchain Survey: 45 institutions from 33 jurisdictions, 17 of which are central banks and 28 OPIS (including ministries, municipalities, regulatory agencies, multilateral agencies and state-owned enterprises).

Figure 38: Three survey samples are dominated by European entities

P38

Despite the best efforts to ensure the diversity of geographical distribution, the European entity dominated all three survey samples (Figure 38).

Active network data

The research team built an additional data set containing 67 active enterprise blockchain networks that have been deployed in production environments. Data was collected through desktop research from April 2019 to June 2019, including major data sources (official documents and statements on company websites) and secondary data sources (such as news articles, blog posts, podcasts). In addition, the research team conducted telephone interviews with representatives of several projects to cross-check public data sources.

The resulting data set covers the enterprise blockchain network of all five world regions (Figure 39) and a wide range of different industries (see Section 3).

Figure 39: Covering the geographical distribution of active enterprise blockchain networks

39

Appendix 2: Technology Platform

Code base

Figure 40: Most core protocol frameworks are based on the code base of existing protocols, basically based on (or derived from) existing blockchain core protocol frameworks (eg Ethereum)?

P40

Note: Based on CCAF survey data

Figure 41: The platform is increasingly supporting more flexible data broadcasting methods.

How to broadcast data over the network?

P41

Note: Based on CCAF survey data, “global”: each node receives, verifies, and broadcasts all data to each connected peer node; “multi-channel”: transaction data is only shared between parties to the transaction. “Custom”: Network participants can choose from a variety of options, including hybrid scenarios.

Network consensus

Figure 42: Although the current global consensus still dominates, development is increasingly oriented towards local consensus

P42

Consensus algorithm

Figure 43: The core protocol framework supports multiple consensus algorithms

What consensus algorithms are supported by the core protocol framework?

P43

Note: Based on CCAF survey data

Smart contract

Figure 44: Common language (such as Java, Solidity) leading smart contract platform

P44

Note: Based on CCAF survey data, “generic” refers to Turing's complete expression language, while “fixed use” refers to a simple scripting language that enables a limited range of operations.

Privacy and confidentiality

Figure 45: Zero knowledge proof is still experimental

Which of the following privacy enhancement methods are supported by your core protocol framework?

P45

Note: Based on CCAF survey data. "Transaction Visibility" refers to limiting the visibility of counterparties participating in the corresponding transaction.

Appendix 3: Challenge

Operational challenge

Private Sector

Table 4: What are the main operational challenges that hinder the adoption of blockchains in mainstream enterprises?

Figure

Note: Based on CCAF survey data

Public Sector

Table 5: What are the main challenges that hinder the public sector from adopting a broader blockchain?

P46

Note: Based on CCAF survey data

Technical challenge

Table 6: What are the main technical challenges currently facing the enterprise blockchain network?

47

Based on CCAF survey data