Recently, the Aishu 2021 Smart Conference was successfully held in Shanghai, at which Jin Jianhua, founder & ceo of Aishu Analytics, shared the theme of "Enterprise Digital Transformation and Middle Office Strategy", and jointly released the "Unstructured Data Middle Office White Paper v2" with Aishu.
Jin Jianhua believes that the ultimate purpose of digital transformation is to create customer value, and digital technology is a means in the transformation process, and this means requires us to change and reshape from the strategy, from the whole business process, the whole value chain and the organizational level.
The following is the full text of the speech:
Jin Jianhua (founder and CEO of Love Analysis): Good morning, just now Mr. He mentioned Plato's analysis of knowledge, this quote involves the cognition of knowledge by organizations such as Love Analysis, I will briefly introduce the situation of Love Analysis.
Love Analytics is a research consulting firm serving the digital market. In the digital market, the role of Love Analytics is to turn the best practices and knowledge system formed in the process of digital transformation of enterprises into methodologies and tools to help everyone make better decisions, and ultimately to help Party A users solve specific problems in the process of digital transformation.
Love analysis in 2015 was established so far, served more than 2,000 customers, at the same time we also served many Party A, in the process of serving Party A to do digital transformation, service vendors, we have accumulated a lot of knowledge and best practices, I hope that today I share the content can better help you establish a understanding of digital transformation, as well as our theme today - the cognition of unstructured data middle office.
First of all, after the impact of the new crown epidemic in 2020, digital transformation has become a "hot word", and the concept has gained good market education and popularization. However, there are still most of the questions that have not been solved, what is digital transformation? What exactly should digital transformation do? What is its value? This is the question that all Party A enterprises, especially traditional enterprises, face digital transformation.
From the definition of digital transformation, first of all, the ultimate purpose of digital transformation is to create customer value, digital technology is a means in the transformation process, and this means requires us to change and reshape from the strategy, from the whole business process, the whole value chain and the organizational level.
In the process of digital transformation, in fact, it is also shaping a data-driven organization, in the whole transformation process, data has played a vital core role, then, what is the value of data, and in what way can the data value be well presented, this is the first topic we talk about today; the second topic is how to build the middle office; the third is in the process of building the data middle office, what is the unstructured data innovation, what is the value to everyone?
Why data is the foundation of the digital transformation process, in 2020, from the national macro level to make a very important and far-sighted definition, the definition of data as a new type of production factors after the four basic elements of land, labor, capital and technology, basically determines whether it is the global market or the Chinese market, which is the development momentum of the next few decades, is a very clear definition from the policy level.
Data is not only a new factor of production, but also China, as the world's second largest economy, has developed rapidly in the digital economy, and data is becoming the most important engine of China's economic development.
In 2020, the digital economy accounts for 39% of the total GDP, while the GDP growth rate in 2020 is 2.3%, but the growth rate of the data economy reaches 9.7%, and the gap of more than 3 times can be seen that the digital economy has played a very important role in our entire economy.
The value of data as a basic new production factor, not only from the collection of data to the commercialization of early data, data must be managed through data to make it truly valuable, how to manage, we generally divide into three steps:
The first step is data integration; the second step is data assetization; the third step is data value generation. How can data be truly valued in business scenarios? We can roughly divide the value generated by data into three categories, the first is to generate value in the process of operating companies and business operations; the second is to generate value in business innovation; and the third is to better help us make management and business decisions through data. Only in this way can data be truly valued in economies as a new factor of production.
The concept of digitalization has been mentioned for many years, whether it is data governance or data assetization, some domestic enterprises have done it earlier, but there are not many who can really manage and use data well. Forrester surveyed a lot of business leaders, about 71% are doing data management work, but only 8% of them are really effective, so most companies are not getting the results they deserve in the process of using data, managing data, and making data valuable.
The most core reason for this result is, first of all, in the process of data integration and assetization, the original chimney of traditional IT, the system is a separate system, and there is no way to integrate these separate systems well, so that the data becomes a complete data asset; in addition, in the past data application development process, it is basically repeated construction, resulting in data reuse and standardization is difficult to do; finally, the data is not based on the specific business scenario requirements for assetization, As a result, enterprises invest a lot of IT budgets every year, but they can't see roi.
How data can generate value must have an idea - the concept of data management and data to create value. Although the data center platform was proposed as early as 2015, its concept has been evolving continuously, and the data center platform is completely a Chinese localization concept, and it is very suitable for the concept of digital transformation of Chinese enterprises.
Our cognition of data middle office, or our definition of the concept of data middle office, is not actually a technology or a product, if the data middle office is regarded as a product and a technology, I think this recognition is a big misunderstanding. The data center itself is actually a set of methodologies, a set of methodologies for data management and data value creation, with a strong business-driven concept behind it, and of course, it is also necessary to apply many technical platforms or tools to make data generate value. The core point is that the data middle office must take the application scenario as the starting point, and through the governance of data, let the data assetization and data business enable business scenarios, and create better value for customers.
To briefly tell a case, the securities industry is a relatively good industry for informatization, the business development of the brokerage company was relatively simple ten years ago, the business is relatively formed, but in the past decade, especially after the development of the mobile Internet, the business of the brokerage has undergone great changes, and the update and iteration speed of the business has become very fast. The original IT system does not actually have good scalability and secondary development, and when it comes to new business, it is necessary to continuously build new data applications.
In the process of new data application construction, it is a single point application, and the original data capability cannot be reused. Then it proposes to do the data middle platform, the core is to unify and govern the original data on the basis of the big data platform, establish a unified data development specification, form data assets, and ultimately empower the business, there are many specific business scenarios here, such as intelligent investment, intelligent marketing, to help the sales team better develop new customers.
The second case is a municipal emergency management bureau, the emergency management bureau is a newly established department a few years ago, the new department is composed of many departments, after the establishment of the department involving the original department of the data scattered everywhere, there is no way to do data integration, from the perspective of emergency management, a new department after the establishment to complete the emergency management of the early warning, including before, during, after the event of all the relationship, which will involve a lot of data. Through the data middle office solution, the Emergency Management Bureau has achieved all-round emergency management, opened up 10 business systems, accessed 160 million emergency data volumes, and realized multi-scenario data fusion, management and value application.
The third case is the construction of an energy Internet by a provincial power company, and the energy Internet is a national strategy in the field of energy. Looking at the energy Internet in a specific scenario, there is a planned power outage in the power system, that is, it needs to be shut down when some equipment is faulted, which will cause regional power outages. Originally, when making a power outage plan, it is necessary for staff to deal with it, such as looking at the ca map, looking at the circuit map, judging which areas and which users will be affected by the problem of power outage through the circuit map, and then sending SMS notifications, this process of manpower consumption is huge, including the manpower investment to deal with the failure, the cycle of the fault and the time when the final failure occurs are inefficient behaviors.
In the energy Internet construction project, the power company introduced a data middle platform, added the application of knowledge graph on top of data governance, basically without too much personnel intervention, you can accurately define which areas need power outages, the second precise definition to which users will blackout, the third based on the judgment of the region and the user automatic text message notification to the user in advance, which is a very specific case.
How to build the middle office? The middle office is a methodology is also a concept, the core of the middle office is that when doing digital transformation, it must start to judge its value from the top-level design, the data middle office is a strategic choice in the process of digital transformation, it not only needs strategic level support, but also needs to support it from the organizational level, from the top design to the support of the middle office technology to the support of organizational operations, and then to the support of data talents, which is a very important starting point for data middle office.
Take the case of a fashion industry company's data center. In 2018, the company formulated a three-year revenue plan to achieve 10 billion, and the final realization of the 10 billion plan is driven by consumer operations, which is a clear business strategic goal. Based on this business strategy, it has developed a strategic plan for consumer-centric digital transformation, in which the data platform base built around consumer operations has played a decisive role. This involves how to dataize the business, operate consumer data assets, and finally turn the data into business empowerment in the application scenario, which can better support member operation activities such as advertising, promotional activities, member pulling, and old customer repurchase through data, and achieve higher business returns.
It can be seen that in specific business scenarios, this company has achieved good business results. For example, the overall repurchase rate of veteran customers in a certain activity increased by 10%, the effect of a double 11 promotion doubled compared with previous years, and a new activity converted 4,000 new members and more than 8 million sales with a budget of 30,000.
Data middle platform to land, in fact, after a large number of practice and methodology summary, love analysis will usually investigate the ability of manufacturers and the needs of Party A, in this process we focus on the study of Party A's landing practice, in the process of research need to turn practical experience into methodology, from the top of the digital blueprint to define what business goals and strategies are what it is, this is the top-level design determines whether the data middle platform can generate value.
It is necessary not only to understand what current data assets are, but also to plan the data middle office based on the needs of enterprise business scenarios and business processes, set the priority of scenario landing, and know which scenarios can generate value in the data.
Of course, the middle is inseparable from the construction of the technology platform, in the process of business value realization, it is impossible for all business processes or the whole value chain to be able to do digitalization immediately, all enterprises, whether it is technical facilities or their own business scenarios, will affect the embodiment of the value of phased data, in this process we generally recommend that in the process of setting the priority of business scenarios, find the scene that can best reflect the business value, give priority to making a closed loop, and see the effect in the short term. It can push organizations to better digitally transform from the top down.
Finally, not only to build but also to operate, the investment of various resources, the expansion of scenarios will help enterprises better realize the data middle office, and ultimately generate value in more business scenarios.
Intelligence is a new stage of enterprise digitalization, there is a very important point in intelligence - cognitive intelligence, cognitive intelligence is essentially how to knowledge the data, and ultimately play a role in business decisions and management decisions. Through AI technology, data can be turned into knowledge, empowering more business scenarios, and better business decisions, which is the next step in the development trend of cognitive intelligence.
The process of data knowledge is actually the process of knowledge management, and knowledge will involve more challenging problems. We generally define this problem into two categories: First, 80% of the data in the enterprise is unstructured data, in the process of knowledge management, how to turn unstructured data into usable knowledge, how to tap the relationship between full data. Second, the real value generated between data and data to help decision-making is to extract its logical relationship and correlation relationship.
The deep relationship between data and data and the global value of data support the final knowledge management at the same time, that is to say, the process of realizing knowledge management needs to consider how to make data or knowledge truly generate value from two levels.
The first is to have a new data architecture or innovation at the data architecture level to deal with the governance of unstructured data; the second is to do the application of AI on the basis of data governance and data assetization, and the knowledge graph will play an important role in the process of turning data into knowledge. In other words, data architecture innovation and AI-enabled knowledge application are very important elements that can help our enterprises do knowledge management.
You can see that this is the data architecture of the unstructured data center, and I think all the platforms are ultimately designed to better empower the organization and the business.
The third part of the unstructured data from the overall methodology to the accumulation of practical experience, to the final case accumulation, we are very detailed in the "unstructured data middle office white paper v2", this white paper is our past few months with the team of Aishu wrote together, we put the unstructured data platform in the field of Aishu's own practice and the practice of Aishu customers summarized into methodologies and cases for you to taste, I hope that this white paper can better help you solve the problem of data application landing in your organization, so that data can produce real value, thank you.