Standardization of audit data is crucial for achieving efficient and transparent audit processes. Key developments in this field include the Audit Data Standard released by AICPA in 2015 and the accounting data collection standard GB/T 24589 Financial Information Technology ─ Data Interface of Accounting Software by China SAC/TC 341 in 2010. These standards aim to enhance the consistency and quality of audit data.

AICPA’s Audit Data Standard provides standards for the collection and reporting of audit data, supporting data audits. ISO 21378:2019 is greatly influenced by the expertise provided by ANSI in the USA for AICPA’s ADS.

China’s GB/T 24589 enables integrated collection of accounting data at the China National Audit Office and annual accounting detailed data analysis at data centers, contributing to the efficiency of audit processes.

Based on these backgrounds, ISO/TC 295 Audit data services was established in 2015 as a PC (Project Committee) on China SAC’s proposal and formulated ISO 21378:2019 Audit data collection. In 2021, Study Group 1 Semantic model started its investigation under ISO/TC 295. From 2022 to 2023, demonstration projects (POCs) such as the development of the Core Invoice Gateway and a specialized browser for journal entry information were conducted with the cooperation of XBRL Japan. These projects aim to further strengthen the methodology of collecting, converting, and analyzing audit data, aiming for further efficiency in the audit process. UADC was developed against this background of POC. From 2023, based on the proposal of SG1, the formulation of the semantic model standard specification in ISO/AWI 21926footnote:[Semantic data model for audit data services https://www.iso.org/standard/

86895.html] has started.

The role of UADC is particularly important in the auditing of Environmental, Social, and Governance (ESG) reports. Raw data from manufacturing equipment and factory control computers often have proprietary formats and are challenging to output in standard formats. By using UADC, this raw data can be converted into a unified data format, providing consistent information needed for auditing ESG reports. This improves the accuracy and transparency of ESG reporting, providing more reliable information to stakeholders.

Audited companies can quickly respond to accounting audits using detailed data without spending time conforming to specific data formats required by audit authorities or firms. This not only saves time in the audit process but also opens the way to deeper analysis and more accurate audit results.

Using standardized data formats maintains the quality and consistency of data, enhancing the reliability of audit results. The ability to convert data from different systems and formats into a unified format makes it easier to integrate data from various sources.

The conversion and standardization of audit data by UADC provide significant value in accounting audits and ESG report auditing. Audited companies can achieve quicker and more efficient audit processes, further enhancing the quality and transparency of ESG reporting. This improves the quality and speed of auditing and data analysis, providing more valuable insights for companies.

In the Japanese Core Invoice Gateway, we have refined the concept of syntax binding defined in EN16931 by introducing XPath. This allows the mutual conversion of electronic invoices (JP PINT) defined using UBL 2.1 in Open Peppol and those compliant with UN/CEFACT CII for SME common EDI according to binding definitions. In the development process, we developed a new logical model for the Japanese Core Invoice and defined syntax binding with it.

Program page:
https://www.sambuichi.jp/core-japan/

In the Japanese Core Invoice Gateway, we have registered the syntax binding in the XBRL taxonomy and used xBRL-CSV files as intermediary files. The data conversion of JP PINT to standard CSV and standard CSV to common EDI electronic invoices is performed by a generic Java conversion program.

With such a configuration, even if the interface definitions of the source and destination change, it is possible to respond by limiting to changes in the syntax binding definitions, without requiring modifications to the data conversion program. This improves the flexibility and sustainability of the data management process, contributing to the efficiency of audit operations.

Although the input and output of CSV files have been easily programmable and thus not considered much of a problem, there have often been many cases where it was questionable whether the content of these processes was clearly documented as specifications, leading to difficulties in maintenance. Semantic binding significantly reduces the burden of program maintenance by defining the mapping of transformation processes as a correspondence dictionary within the taxonomy definition.
Semantic binding in UADC (Uniform Audit Data Collection) is very important for maintaining semantic consistency of data. In this process, data is distributed according to the definition of semantic hierarchies, such as mapping specific CSV columns to auxiliary account codes like cost department codes or supplier codes.

This approach is flexible and can adapt when the format of the CSV interface changes due to app version updates or when migrating to different software. By changing the semantic binding to the altered interface format and registering it with the XBRL taxonomy, it is possible to respond quickly without waiting for program modifications. This maintains the continuity and consistency of the data management process, improving the efficiency of audit work.

Semantic binding in UADC significantly streamlines the data conversion process for auditing. It facilitates data migration between different software systems, maintaining data integrity and accuracy. Particularly in the fields of auditing and data analysis, this directly translates into saving time and resources, supporting quicker and more accurate decision-making.

According to Wickham’s definition of Tidy Data, data should be organized as follows:
– Each variable forms a column: Each variable (measurement or attribute) in the dataset has its own column.
– Each observation forms a row: Each row represents a single observation.
– Each type of observational unit forms a table: Different observational units are placed in separate tables.

Related Article:
https://www.sambuichi.jp/?cat=68

Wickham’s Paper:
https://www.jstatsoft.org/article/view/v059i10

xBRL-CSV extends the concept of Wickham’s Tidy Data, particularly in handling hierarchical data, with features such as:
– Support for hierarchical data structures: While Tidy Data originally assumes a flat data structure, xBRL-CSV supports hierarchical data structures, allowing more efficient representation of naturally hierarchical data like complex organizational data or accounting data.
– Expression of multi-level variable relationships: xBRL-CSV allows for associating multiple levels of variables within a single record, enabling the representation of accounting information related to a specific transaction and related departmental information in one integrated view.
– Extensibility and flexibility: xBRL-CSV is based on the XBRL taxonomy, making it easily extendable and customizable to existing accounting standards and reporting requirements.

For observational units of journal entries, xBRL-CSV enables hierarchical data representation as follows:
– Information per entry: Each row represents observations like posting date, entry date, and description for each entry.
– Detail line information: Details like amounts for each account and applicable consumption tax are recorded in line items.
– Detailed information per auxiliary account: Detailed information for auxiliary accounts defined in accounts, like bank accounts, suppliers, cost departments, are registered as row observations using the same auxiliary account number and name fields.
– Hierarchical recording: These hierarchies can be recorded in a single file, providing great convenience in data organization and analysis.

Journal Entry Information Dedicated Browser
https://www.sambuichi.jp/core-japan/journal_entry/

In typical relational databases (RDB), JOIN operations are needed to combine related data across multiple tables. However, with the adoption of xBRL-CSV, related data can be hierarchically managed within a single file, eliminating the need for these complex JOIN operations.

Maintaining a relational database involves many tasks, such as ensuring data integrity, security, and backup. However, using xBRL-CSV eliminates the need for these complex database management tasks, significantly reducing the administrative burden.

Using xBRL-CSV simplifies and speeds up data processing. Since the data is contained within a single file, data reading and writing are performed efficiently, improving the overall speed of data processing.

The adoption of xBRL-CSV in UADC greatly contributes to the efficiency of data processing. It allows hierarchical data, like those in XML or JSON, to be represented in a single file, eliminating the need for complex JOIN operations. These extensions maintain the basic principles of Wickham’s Tidy Data while catering to

The liberation from vendor dependence through UADC supports business growth and innovation by enabling companies to manage audit data more effectively and choose appropriate tools and services for various business needs.

UADC’s solution to long-term data storage issues enables the use of past data according to contemporary business needs, significantly contributing to the efficiency of data management and risk reduction for companies.