Customers

Guangdong Local Tax Bureau

Guangdong logo

"In an era without BI, the taxation department had to make judgments by gut feel and experiences, being unable to make informed decisions. Now with BI, we are able to obtain critical elements of taxation management from reports, data analysis, and data explorations so as to provide the right evidences for the next-step actions or decisions."

Hua Jun
Director Information Center
Guangdong Local Tax Bureau

Challenge

Guangdong Local Tax Bureau (GDLT) serves as the functional department of the Guangdong Provincial government in charge of local taxation work. There are 14 sections within GDLT, in addition to two direct subsidiaries and five non-enterprise institutions that control 21 city-level branches, 127 county-level bureaus, and 1064 taxation offices (sub-branches). In 2006, the cumulative tax income of GDLT reached ¥230.55 billion, hitting a new record high. By the end of 2006, the taxation revenue of GDLT ranked top among its peers for 13 successive years, with total income value accounting for about one-seventh of China's total, and providing a solid financial basis for the economic development and social stability of the Guangdong Province and the nation as a whole.

In 2001, GDLT took the lead in China to initiate the Grand Convergence informatization project. After a 10-month project consultation and assessment, GDLT identified as the project's major targets the following:

Taxation reflects the status of a national economy's operations, and behind taxation data is a nation's or even a whole society's tremendous wealth. It is important for countries to use IT and information management systems to analyze taxation and related data for economic leverage and regulatory purposes, and to support different levels of government in macro decision-making.

In March 2006, GDLT proposed to build a taxation analytical system within the framework of the Grand Convergence informatization scheme. Zhang Jiansheng, senior engineer of the Bureau's information center, recalls that with the gradual improvement of infrastructures and application programs, taxation analytics became a focus in promoting the overall implementation of the project – marking that the informatization development of GDLT had entered a business-driven expressway. He says, "The operation and enhancement of our Grand Convergence production system laid a foundation for business standardization and data convergence in the whole province. It is invaluable to collect and centrally store the huge volumes of levy, management, and inspection data, which are needed to turn information into knowledge." The volume of management and IT data places even higher demand on data analysis and processing capabilities. For GDLT, in the early days of 2006, some business pains were beginning to emerge.

The first issue was the convergence of data as well as vertical and horizontal sharing of data and interoperability. Although Jiansheng's bureau realized the provincial-level convergence of tax levy and management systems, a number of internal sub-systems with data scattered around still require further integration.

Second, within GDLT, the business auditing staff, mid-level managers, and top decision-makers all have urgent needs for query and analysis based on a business intelligence (BI) environment. Business users have to deal with endless tax-related matters required to make judgments and process the details of taxpayers, for which query and simple reporting are absolutely needed. Mid-level management has to face one specific group or type of taxpayers, and get a timely understanding of the relevant data in order to analyze and supervise business development. Top decision-makers not only need objective and timely insight into the facts but also want to grasp various taxation laws and changes under the macroeconomic conditions supported by more complete, comprehensive, and effective analytical methods.

GDLT contends with large volumes of reports and inquiries that have to be done manually. Jiansheng gives an example of the highly complex taxation report – a single report intended to cover all kinds of data, which is summarized horizontally and vertically by different themes. To line up the complicated data in a well-ordered manner, it needs to add multiple layers of themes here and there. Such manually produced reports are inefficient and waste resources. GDLT launched a campaign aiming at turning volume and boring data into social wealth, supported by BI.

Approach

As the global leading supplier of BI, Business Objects, an SAP company, attracted GDLT because of its advanced products and services as well as its expertise in service provision. Yao Xiaojian, CIO of GDLT, says, "The Grand Convergence IT project of GDLT is the first provincial-level project of its kind in the national tax segment. We chose Business Objects because of its advanced experiences in building data warehouses worldwide, its market-proven mature products, and specialized service standard."

When addressing the business issues of GDLT and providing relevant solutions for the first phase of the project, the taxation analytical system of GDLT placed high demand on Business Objects for gathering insight and predictability from its databases. In discussing business requirements, Bill Xiong, Business Objects architect for GDLT, observes that the purpose of the taxation analytical system is to summarize, analyze, drill down, and model the daily increasing tax-payment and declaration data in order to explore supporting information that can be of value to decision-making. The system's main task is to apply the advanced data analysis concept and sophisticated data analytics algorithm in line with the current situation of local taxation systems – ensuring effective analytical processing of all the data and maximizing the value of information contained in the data.

According to Xiong, the difficulty of such system development is in integrating GDLT's databases, comprising the core database, individual taxation database, inquiry database, and the distributed multi-channel database GDLT requires to ensure data authenticity, integrity, and consistency – quite demanding for extraction, transformation, and load (ETL) tools and also on system performance. Over the past three years, GDLT accumulated 1 terabyte of source data volume, and projects that the data volume will increase even more rapidly. Additionally, 15,000 terminals and many concurrent users pose great pressure at the system’s peak operation period. Therefore, scalability and data warehouse platform performance are stringent requirements. A first phase requirement is to establish a solid foundation for the long-term development of the taxation analytical system, with an eye toward sustainability. All the above difficulties require a perfect integration between the data layer, application layer, and presentation layer of the whole analytical system.

After a careful study of the structure and business needs of GDLT’s application system, Business Objects proposed a solution that integrates GDLT’s data and application layers, delivering information through a portal on a consistent platform. BusinessObjects™ Enterprise XI Release 2 gives GDLT a complete and reliable BI platform for reporting, query and analysis, multidimensional analytics, and exploration of the province's tax levy and management data, which is centrally released, presented, analyzed, and explored over this platform. BusinessObjects Web Intelligence® provides query and analysis capabilities, enabling users to gain greater insight with self-service information access and interaction. End users can easily create data inquiry and calculation without having knowledge of data sources and data structure. A multidimensional database server with dynamic and manageable analytical functions organizes database warehouse data according to predefined themes and analysis demands, making it easy for analysts to conduct comprehensive analysis by combining quantitative and dimensional data. Users are able to partition, separate, and drill down on OLAP data from SAP, Microsoft, Hyperion, and IBM. Crystal Reports® makes it possible to create complex reports, which not only support summarization of multi-content statistics, but also support multiple data sources – for example, one report can support data from asynchronous data sources. In addition, Crystal Reports supports graphical charts, word processing software, databases, and spreadsheets that export reports in PDF, HTML, and Excel formats.

Results

The taxation analytical system of GDLT includes three major functions of system query, microanalysis, and macro-themed analysis. The infrastructure platform is meant to take advantage of advanced computer technology to establish a stable, secure, open, and scalable operation platform for the existing data within the Grand Convergence system, as well as some external data through well-designed integrations. Thus, it ensures the standardization and consistency of the province's basic taxation data, helping to realize the sharing and centralized processing of taxation information in the province and to enhance collection, processing, and decision-making capabilities.

In the first phase, the GDLT analytical system is deployed to several hundred users in the province. Apart from some internal staff, all other employees of GDLT are potential users of the system, including different levels of leaders and management personnel.

The first phase of the project ensures the authenticity, completeness, and centralized control of taxation data. The data warehouse platform provides a solid and trusted foundation for data integration and database analysis. With a consistent BI platform, GDLT has truly realized horizontal and vertical data sharing and interoperability. Thus, it can carry out centralized control over the province’s taxation data. As well, the data authenticity, consistency, and integrity provides an essential guarantee for the establishment of taxation analysis and decision-making support systems, making it possible that the taxation work will always function as the mirror of economic development.

The director of GDLT information center Hua Jun says, "In an era without BI, the taxation department had to make judgments by gut feel and experiences, being unable to make informed decisions. Now with BI, we are able to obtain critical elements of taxation management from reports, data analysis, and data explorations so as to provide the right evidences for the next-step actions or decisions."

Using the taxpaying evaluation as an example, the Foshan Local Tax Bureau carried out a project aiming to make assessment on the tax payment of business tax in the real estate, architectural installation, and hotel and restaurants sectors. The Bureau selected 10 metrics for the evaluation work based on the declared data of taxpayers, financial data, and the annual declaration of company income tax. Among the 10 metrics, three are general items, six are business operational tax-related, and one is concerned with individual income tax. The Bureau conducted analysis on the tax payments from 133 taxpayers under suspicion, meeting with them for clarification. According to the calculation results from the evaluation system, they finally narrowed down to 20 persons for further auditing. Through the auditing and validation on those 20 taxpayers, the evaluation system achieved the expected target in dutifully reflecting the existing issues of taxpayers.

Since its establishment in 1994, GDLT has created many records in the national taxation industry, ranking first in local tax income for 13 years consecutively, with its total income value accounting for over one-seventh of the country’s total. Now through the cooperation with world’s leading BI vendors like Business Objects, the GDLT has become a forerunner in the industry in terms of applying data warehouse solutions. As commented by industry analysts, it is perhaps just a start – and the mindset of developing taxation work with technological means will bring sustained advantages to GDLT, which is going to be a more splendid venture.

Smaller textLarger textPrint this page, properly formatted for paper

Quick Facts