Case Study of Company C

The last case study this research conducted is Company C, and the person this research interviewed is the R&D and data science team director. Company C was founded in 1996. At that time, it was the first and only job bank in Taiwan. Company C aimed for the most efficient services for the hiring companies and job applicants to match the job demand and needs. In 2007, Company C has expanded its service coverage to China. According to Company C, they served more than 230000 corporate customers, and acquired more than 5 million resumes on their database. Company C offers professional service for career planning. The Company C Job Bank is not only a website for job listing, but also a professional provider of human resources service.

There are overall three main services of Company C Job Bank other than just job searching. First, Company Hunter provided thousands of businesses to recruit talents. Then, Company C Tutor offers the opportunity of tutor vacancies in various professional fields. Lastly, Company C Temp provided temporary job listings and match jobseekers with user companies.

57

Company C developed the Job Bank website, which is a unique service in Taiwan and China in contrast with rest of the world. Several reasons why this case study is included in this research. Company C is an internet company, which they receive and analyze tremendous data every day. In addition, Company C involved structure or semi-structure data, which is similar to the financial institution. Moreover, they have a data science team, which emphasize in

developing data-driven new products and services. Because of the data similarity and

professional data science team, this research decided to conduct the interview with Company C.

Although Company C is not an actual financial institution, this research believes that Company C can provide a unique insight in Big Data implementation.

4.3.2 Implementation and Process

Before the interview questions, Company C data science team director gave a brief explanation on why they decide to implement Big Data in year of 2011. The director explained with Gartner’s 2012 Hype Cycle for Emerging Technologies as shown in Fig. 14, The 2012 Hype Cycle evaluated the maturity and benefits of emerging technologies as well as their future directions. The director mentioned that Big Data emerged approximately on 2002. According to Gartner’s 2012 Hype Cycle for Emerging Technologies, Big Data will reach its maximum in 2 to 5 years. Moreover, several open-source technologies became mature and sophisticated such as Apache Hadoop. Therefore, the director decided to begin the Big Data project on 2011.

58

Figure 14. Hype Cycle for Emerging Technologies Source: Gartner, 2012

Company C data science director indicated that they only analyzed the data as market research, but rarely on data-driven product development. Therefore, they have definite objectives for the implementation that they want to develop a more competitive new services and products by using data-driven technique.

Company C divided Big Data implementation project into two groups. The first group involved with daily operation analysis such as competitor analysis. Because of insignificant data amount, Company C implemented third party software solution for this group. The second group involved with developing new products and services. This group is the core team in Big Data implementation. This group usually needs to deal with tremendous amount of data. The director

59

indicated that the most current business intelligent technique on the market will not be able to process. Therefore, the director decided that they will process this amount of data by using open-source technique such as Hadoop and Spark.

Lastly, the director emphasized that Company C is an internet company; hence they don’t have strict Standard Operation Protocol (SOP) to implement a new technology. At the beginning, the core team is assembled with around 3-4 employees. Then, in the Prove of Concept stage, they functioned as a project team. They first organized all the benefits and risks in Big Data. Later, they will execute an actual project. If the project success, they will perform the next development.

However, if the project failed, they will learn from the mistakes. The director said the best way is

“Learning by doing.” The director also defined that in the implementation process, this group work as a project team. However, they established a data science department after the Big Data implementation.

4.3.3 Enterprise Requirements

As a R&D director, he supported the implementation at the beginning. The director also indicated that top management directors were all waiting for the maturity of the Big Data technology. Therefore, without a doubt, the top management directors were all supported in this implementation as well.

The main challenge of the whole implementation is human resource for Company C. The director revealed that they have only 5 employees so far after the establishment of the data science department. He indicated that the Big Data ecosystem should not emphasize in the hardware or the software, but the way to process the Big Data. The technique or method to process the Big Data is the key to solve the problem, and invented new data-driven products. In

60

order to process the data, the potential candidates must have basic technical skills, such as python, statistical analysis and mathematics. In addition, they must be able to find the problem from complicated data, and provided an innovative method to solve the problem. He mentioned that it’s difficult to find a candidate with both coding skill and statistical analysis.

During the recent years of implementation, the director responded that they have mostly internal Big Data training, and periodically they will have training outside. They also attend social network meeting to learn the newest technique in Big Data.

4.3.4 Benefits after Implementation

As mentioned above, the maturation of the Big Data ecosystem made data processing easier and more transparent. The Company C R&D director revealed one of the benefits is that the customer satisfaction increased noticeably. In addition, they released two data-driven new services on Company C Job Bank, which are School-Career Development Map (Fig. 15), and Career Wikipedia (Fig. 16). In the School-Career Development Map service, it statistically showed the salary and employment opportunities for all of the college and master school in Taiwan. This service also provided specific salary and job information for all majors in either college or master school. In the Company C Career Wikipedia, it has all the information people need about a specific job. For example, if people like to know information about what

administrative staff is, the Company C Career Wikipedia service will list out the entire key component regarding administrative staff such as the job description, personality, stability, salary range, capability, and job vacancies.

61

Figure 15. Company C School-Career Development Map (104, 2011)

Figure 16. Company C Career Wikipedia (104, 2011)

62

In conclusion, they defined Big Data as an ecosystem that process the tremendous dataset.

They are convinced that what and how they process the Big Data should be focused and

emphasized more instead of compare between the Big Data hardware and software system. Later, they indicate cross-functional and diverse skills are two important components for successful candidate to process Big Data proficiently.

63

Chapter 5. Research Findings and