LINKEDIN:55 support 2,700 employees, 200 million users of large data analysis

As the originator of professional social networking sites, LinkedIn is essentially an enterprise that generates business value by running big data. From its inception in December 2002 to the beginning of 2013, LinkedIn registered users to 200 million, an average of a new user per second, 86% of the "Fortune 100 Companies" are using LinkedIn's pay solution, 2.7 million company home page here, Users initiate more than billions of searches a year. It is reported that in the last quarter of LinkedIn's 161 million dollar revenue, the "talent solution" has contributed more than half. Unlike Facebook and Twitter, LinkedIn is a relatively low-key, but profitable situation is very good. A few days ago, csdn cloud computing channel reporters at the 2013 Teradata Big Data Summit, interviewed the LinkedIn business analysis director Zhangxi Dream (Simon Zhang), in-depth understanding of the platform behind LinkedIn and the business value of large data analysis.

Data analysis supports LinkedIn's five departments

With the deepening of enterprise's understanding of large data analysis, data analysis has been mentioned in the agenda of Enterprise information construction. But more companies are simply positioning data analysis as an extension of the business and IT department or project management, rarely as an independent department, not to mention the five core departments of research and development, products, markets, sales and operations. LinkedIn did the opposite.

Zhangxi Dream (Simon Zhang), director of business analysis at LinkedIn Corp.

Q: How does LinkedIn define the Business Analytics department?

Zhangxi Dream: LinkedIn has three large business sections, one is the human business, last quarter earnings showed that accounted for 53% of the revenue, and the second is the marketing of advertising, about 27%; third, advanced subscription services, and LinkedIn members have relations, accounting for 20%. All three businesses need data analysis support. The team I led is now 55 people (the target is 60 people), the focus is on 200 million registered users through the integration of data architecture, BI, data Mining and analysis, direct support for nearly 2700-3000 LinkedIn internal staff (there are currently about 3500-4000 employees in LinkedIn) , which means meeting 70% of employees ' data analysis requirements, which cover five major business functions, such as research and development, product, marketing, sales and operations.

Q: Why do so many internal employees have the need for data analysis?

Zhangxi Dream: LinkedIn is defined as a job search platform, but in fact, according to LinkedIn's data, only 20% of people are more active in finding jobs, and about 80% of LinkedIn users ' main business purpose is to establish contacts/relationships with other companies. They want to connect with all the professionals in the world, making work more efficient, more influential and more successful. These needs are reflected in LinkedIn, where the five professional sectors have different data analysis needs and vary widely.

Q: How are LinkedIn's products designed to be built on a data platform?

Zhangxi Dream: Right. LinkedIn is essentially using data to generate value. All product design principles are derived from the data analysis of the "relationship between people." For example, LinkedIn's membership data is changing, it is clear that there will be more valuable new products, such as the 2008 launch of "People you may know", and we are now developing "you will be interested in the work, you will know people, you may like to see, etc.", these products will promote business growth, Enhance user stickiness. On this basis, the core of the business (accounting for 40%-50%) headhunting business, as a pure data business, the purpose is to help enterprises find the most suitable, the most outstanding talent. In LinkedIn's view, the core is to ensure the success of the company is the talent, from the data analysis of people, to help these talents succeed, is our constant philosophy.

Q: These new products from the principle of data logic, basically the same, what is the difference between each other?

Zhangxi Dream: The basic principles of the data are identical, and the essence is to discover the correlation through the data, and then form the social network (Social receptacle). Prior to LinkedIn's more basic products for members, now that the data platform has been built and profit margins are available, it is time to do more deep digging. Cross-analysis of these products will be new products, such as through the relationship between people to do different colors of the partition, the formation of a new analysis of diversity is more valuable.

(Responsible editor: The good of the Legacy)