Before the industry changed, the transformation happened through informatization, but informatization just moved business processes from offline to online, and used computer networks as a carrier to support business activities. The data itself was not analyzed, applied, and played the core. value. With the rapid development of emerging technologies, the transformation of new retail is gradually evolving into digital intelligence, and data, as the fifth factor of production, is playing an increasingly important role.

About the author

Han Bing: OceanBase solution architect, 12 years of experience in the software industry. He has served as application architect and infrastructure architect. He has provided digital and intelligent transformation and consulting services for many well-known large enterprises, and has a national invention patent.

"People, Goods, and Fields" in New Retail

Whether it is retail or new retail, there are three keywords—“people, goods, and market”. In short, we need to sell the right products to the right people at the right time and place. But in the new retail, "people, goods and fields" have changed.

图片

"people"

In the final analysis, retail still revolves around "people", because only when "people" consume can real retail behavior occur.

In traditional retail, we need physical contact with customers to reach them. In new retail, the contact channels are greatly expanded. Most retail activities have almost no physical contact except for logistics delivery. Information flow has become the main means of interacting with customers.

In traditional retail, the information flow basically relies on the market promotion of enterprises to reach customer information in one direction without direct feedback; but in the new retail scenario, the information channels between customers and enterprises, customers and customers, and customers have meshed information. The interaction has formed a unique community after information gathering and identification, and the individualized behavior characteristics of groups including individuals are more significant.

"goods"

With the rapid development of the industry, the retail format has undergone tremendous changes. For example, there are more and more types of goods, not only physical goods, but also many virtual goods. In general, "goods" can be divided into two types, one is the goods that people need to produce, which is to sell known goods to known or unknown people; the other is people who do not know what to produce, but have already There are "fields" with communities and specific domains, and new commodities are produced according to the needs of the communities and "fields". This on-demand production method is actually closer to the essence of new retail.

"field"

In the past, the venues for transactions were relatively fixed, such as e-commerce platforms or brick-and-mortar stores. There are more and more places where transactions are generated, such as hearing an advertisement on a smart speaker and asking it to help place an order. The scenarios of placing orders anytime, anywhere are diversified, and the "field" becomes larger and larger.

But sometimes the "field" becomes very small, because only in a private circle or community can the same "field" be formed to complete the transaction. We gather people with common hobbies and preferences to form a community, and increase the stickiness of the community through socialization . Gather together and interact, and finally form the retail behavior of "goods".

Key business processes of new retail

As shown in the figure below, the key business processes of new retail mainly cover logistics, information flow, and capital flow, that is, from product production, sales, payment, to logistics, and even overseas sales, and finally to users.

图片

In the new retail, the information flow has changed a lot - the original information flow is one-way, the enterprise actively reaches the user, and cannot analyze how the user receives and likes the information flow, and the information flow will affect the user's purchase. What impact does the behavior have; the current new retail is two-way, and companies can continuously iterate their strategies based on user feedback and data-driven.

Digital and intelligent support structure of new retail

At present, many new retail enterprises are upgrading their business models with the help of data, and the support structure of digital intelligence plays an important role in this. Whoever has a higher degree of digital intelligence and who makes better use of data will have better business growth. As shown below:

Let's take a look at "Five Central Taiwan" first. When Alibaba proposed the "big, middle and Taiwan" strategy, many people felt that having "middle and Taiwan" would definitely help business. However, it was later discovered that "Zhongtai" was deified and then corrupted, making it difficult to survive. "Zhongtai" is essentially a tool with resources, just like a car full of gas, you don't know how to drive, where to drive, and what to transport.

Therefore, many applications have grown on the "middle platform", and the upper layer is "N applications" to help the business move forward. Because each user has different needs and diverse scenarios, it is necessary to provide various types of applications to help business processes run efficiently and accurately.

How do these "N apps" really work? This comes to the top "four transformations", which are actually consulting. There are many roads that others have seen and walked through, and many applications need to be consulted before they can be really used and used well.

Therefore, the higher the support structure of new retail digital intelligence, the more intelligent it is. Integrate the wisdom of predecessors, industry experience and business to form its own unique business model. In the era of digital intelligence, the more a company can combine data and intelligence, the more special it will be, the greater the living space, and the greater the possibility of becoming a unicorn.

The new retail enterprises after the digital and intelligent transformation will mainly have the following three characteristics:

One, real-time high frequency. Real-time high-frequency consumer behavior or retail behavior, which means that customers can have this behavior at any time and anywhere;

Second, extensive access. Reach the people we want to reach through a wide range of information, and reach out to thousands of people through user groups, triggering customers' retail behavior;

Third, data-driven. This is the most important point, because most new retail is contactless, and the source of our information is often data. Only through data-driven can we achieve higher-order business growth purposes.

The new requirements of digital and intelligent transformation on IT technology

So, what are the technical requirements for new retail enterprises to successfully complete the digital and intelligent transformation? I've summed up the following four points here:

图片

First of all, fast delivery.

New retail needs to form an advantage at a faster speed. Whether it is a new event or a new function, it is hoped that it can be quickly promoted after being verified by the first batch of users. But what should we do technically to deal with fast delivery? Applying a microservices architecture would be the best choice. Microservices disassemble and refine the fusion function, and form new changes through a little change, which can quickly allow users to experience it.

The second point is to perceive quickly.

In the past, it was necessary to know user feedback and activity feedback through feedback from research companies or advertisers. But now, if you send a coupon to a user, whether the user has used or liked the coupon, you can bury the data where the data can be obtained in all business processes, and then quickly analyze the results to guide the next business activities. .

Third, recover quickly.

With the increasing number of high-concurrency scenarios, for example, the server is down due to too many seckill activities, resulting in the failure of users to place orders. For new retail, which has an impulsive consumption scenario, waiting for a few minutes to resume services may completely lose many users. Fast recovery means that when there are massive, sudden, and concurrent business activities, it must be ensured that even if one server has a problem, another server can ensure the normal operation of business activities.

Fourth point, strong flexibility.

Whether the technology is resilient or not involves costs, and when business activities can bring huge benefits, companies are often willing to invest in IT. But if business activities don't bring benefits, or if the peak season brings huge benefits, why pay for a year for the peak season? Elastic scaling can significantly save IT costs for new retail companies.

To achieve fast delivery, fast perception, fast recovery, and strong elasticity, there are high requirements for databases. However, the databases of new retail enterprises are currently facing problems such as low business continuity, weak real-time analysis capabilities, architectural performance bottlenecks, low resource utilization, and unpredictable large-scale promotion capacity.

The fully self-developed native distributed database OceanBase can help new retail companies better solve the above problems.

First of all, the HTAP capability of the integrated architecture uses a set of computing engines to support the mixed load of TP and AP at the same time, so that a piece of data can be used for both transaction processing and data analysis;

Secondly, OceanBase can achieve transparent expansion, automatic load balancing, the scale of a single cluster can reach thousands of nodes, the data volume can reach PB level, and a single table records trillions of rows; Third, OceanBase adopts a single-cluster multi-tenant design, which naturally supports cloud database architecture , to effectively achieve resource isolation;

Finally, OceanBase pioneered the new standard of "three places and five centers" city-level automatic and non-destructive disaster recovery. The system has no single point of failure and guarantees RPO=0, RTO<30s.

New retail distributed upgrade case

A world-renowned coffee retailer

A world-renowned new coffee retailer with thousands of stores in China is also an internationally renowned coffee brand with a long brand history. Most of the company's original systems are based on traditional Oracle/MySQL databases. With the growth of users and businesses, orders and coupons have high requirements for high concurrent database service capabilities. In addition, the original system database deployment architecture is heterogeneous and decentralized, some are deployed in the own computer room, and some are deployed on the cloud, and the data on the cloud and the cloud are separated, which cannot meet the requirements for low systemic risk, management, operation and maintenance required for business transformation. .

Through OceanBase, the company has integrated multiple systems such as orders and memberships. Existing Oracle and MySQL applications can be migrated without loss, database integration has greatly reduced the difficulty of operation and maintenance management, and distributed upgrades have effectively solved the problem of traditional database concurrency scalability. Through OceanBase's multi-cloud deployment capability, the enterprise expands the single-cloud deployment architecture, effectively solving the problem of data sharing across multiple clouds. The failure recovery time is also reduced from hours to minutes, and business continuity reaches 99.99%.

A leading hot pot new retail company in China

A leading hot pot new retailer in China is loved by users for its excellent service and healthy ingredients, and has gradually gone overseas to become a world-renowned hot pot brand. In the process of building a new online intelligent service system, the data blowout put forward higher requirements for database scalability, service continuity, response time, and resource utilization.

Through OceanBase's native distributed features of high expansion, high availability, and high performance, the enterprise's database has acquired the ability to expand almost infinitely, and based on the Paxos protocol, it provides users with 7x24-hour continuous services. With OceanBase's high storage compression ratio, multi-tenant co-location makes full use of system resources and HTAP hybrid load reduces data redundancy, the company's data storage cost has been greatly reduced.

write at the end

That’s it for today’s sharing. Finally, I want to say that the transformation of digital intelligence to new retail is obvious to all, and it is also a major boost to the future development of the industry. Data-intelligence and data-first require a strong data management and operation platform for support. The product capabilities provided by OceanBase will definitely become an important supporting part of this platform.


OceanBase技术站
22 声望122 粉丝

海量记录,笔笔算数