XBRL is a version of XML defined to meet the requirements of business and financial information. With XBRL, unique identifying tags are applied to financial data items.
More than simple identifiers, these tags provide a range of information and allow labels in any language, as well as accounting references or other subsidiary information. XBRL can show how items relate to one another, how they are calculated, and whether they fall into particular groups for organizational or presentation purposes. XBRL is extensible, so companies can adapt it to meet special requirements.
XBRL won't change what is reported, only how it is reported. With XML tagging, the information in a business report is computer-readable and can be extracted, searched and analyzed (even if it's from multiple sources and written in different languages).
XBRL (Extensible Business Reporting Language) is an XML-based language for business and financial data. XBRL provides identifying tags for individual items, allowing computer programs to sort data, analyze relationships and generate output. The U.S. Securities and Exchange Commission has mandated that all public companies must report their earnings using XBRL by 2010.
How XBRL Works
In XBRL, individual data items are called elements, which are combined into taxonomies (dictionaries) defined by schemas and relationships called linkbases. An XBRL instance document is a business report in an electronic format created according to the rules of XBRL.
Elements are represented in this format: 1,000. The word liability inside angle brackets is called a tag. Between the opening and closing tags is a value. A computer could understand from this example that the number 1,000 is tagged as a liability, but the computer must be programmed to understand the term liability and what values it could have.
Taxonomies are categorization schemes, or dictionaries, that define the specific tags for individual data items. Because different countries have different accounting regulations, each may have its own taxonomy for financial reporting.
With XBRL, the old adage "Garbage in, garbage out" still applies. An April 2009 study at North Carolina State University evaluated the accuracy of XBRL filings for 22 companies that participated in the SEC's voluntary XBRL filing program in 2006. A comparison of XBRL filings against paper-based reports revealed multiple errors in signs, amounts, labeling and classification. The study called these serious problems, because users will have a hard time spotting such errors among the computer-readable XBRL data.
Organizations can also create specialized taxonomies to cover their own business reporting needs, so "there is no need to force a reporting concept into an ill-fitting box," Big Four accounting firm KPMG LLP explains on its Web site. "XBRL is not a standard chart of accounts. It allows the standardization of common business reporting concepts while providing flexibility to extend the vocabulary to meet individual circumstances."
Schemas contain information on the elements in a taxonomy, their names, IDs and other characteristics, and how a computer should treat them. An XBRL schema is a container with an unstructured list of elements and references to linkbase files.
Linkbases provide information about relationships between elements and link them with specified external resources.
U.S. Securities and Exchange Commission, a big proponent of XBRL, has required companies with market capitalizations of $5 billion or more to use XBRL since June 2008, and it will require all public companies to use XBRL by next year. XBRL's standardized tags should allow investors and regulators to more easily analyze and compare financials from various companies.
Kay is a Computerworld contributing writer in Worcester, Mass. Contact him at firstname.lastname@example.org.
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.