The E-Business Software Weekly is a series profiling trends and developments in software and applications that support e-business, the Internet, and other electronic communication channels. Look for a new story each week in this space.
Streamlining Business Reporting with XBRL
Imagine visiting a crowded outdoor marketplace in some distant, foreign country. You trudge from booth to booth, anxiously seeking buyers for your wares. That's the least of your difficulties. The real challenge lies in the fact that no one speaks your language--and no one seems to speak the same language. And so, at every booth, you must spend hours trying to interpret would-be buyers' words, and even when you've finished, you're still not sure that you have actually understood one another and that a fair and accurate exchange has been made.
Your days are long, filled with the tedious mechanics of just making verbal connections, and not very productive.
Such is the situation that confronts modern businesses as they attempt to exchange transactional and financial data with one another. The arrival of electronic commerce has opened a world of great possibilities, from anytime/anywhere access to customer self-service to real-time data interchange. But that potential often lies locked behind a veil of confusion. Incompatible data formats and proprietary, customized computer languages make it impossible for different organizations' information systems to communicate with one another on their own. For nearly every interconnection, engineers must spend weeks (or more) writing arcane lines of computer code that reconcile data formats and eliminate the barriers to free and intelligible data flow.
The process is long, tedious, and not terribly productive.
Stanley Zarowin, writing in the Journal of Accountancy, compares this translational duck soup to the linguistic chaos that followed the ancient Hebrews' attempt to construct the Tower of Babel. As Zarowin recites from Genesis 11:1-9, the descendants of Noah had "one language, andä one speech." The Hebrews took advantage of this ease of communication to plan and work together to build a tower at the city of Babel designed to "reach unto heaven." Zarowin notes that this handiwork did not please the Lord, who declared: "Behold, the people is one, and they have all one language; and this they begin to do; and now nothing will be restrained from themä" The Lord's solution: to "confound their language, that they may not understand one another's speech."
Corporate America finds itself in much the same disordered condition, cursed with a confounding of computer languages, unable to understand one another's electronic speech. Unlike the early Hebrews, however, we have no one to blame but ourselves.
The Lingua Franca of Business
Enter XBRL. Zarowin, the Journal of Accountancy senior editor, extols: just as English is the universal language of global business communications, "so XBRL soon will be the lingua franca for all business reporting--from issuing financial statements to banks and shareholders to filing 10-Ks with the SEC or uploading business information onto a Web site. The development," he says, "surely will revolutionize how business data is reported, used, and calculated."
So what in the world is XBRL?
Eric Cohen of PricewaterhouseCoopers LLP, a founding member of XBRL's coordinating body, the XBRL Consortium, describes XBRL as a "child of XML," a "subset or dialect of XML." You may recall from last week's column ("Demystifying XML") that XML (eXtensible Markup Language) is an enhancement of ordinary HTML (the language in which Web sites are typically written) that allows for the "tagging" of elements of the page not just with formatting markers, like "bold" or "italics," but also with content identifiers, like "1999 annual income," "price," or "discount rate." As such, electronic parsers can locate, extract, and manipulate data of a particular type, something that is not possible with standard HTML.
Of course, as noted last week, XML faces challenges of its own, not the least of which is ensuring consistent data definitions and consistent data tags, lest corporations slip back into the same linguistic miasma that XML was supposed to correct. Indeed, XML has proven so compelling that there are already more than 200 flavors of the language--200 "standards"--which suggests that another fall of the Tower of Babel may not be too far behind.
XBRL (eXtensible Business Reporting Language) is intended to rectify this problem. As Zachary Coffin of KPMG Consulting notes, "XBRL has no competitors: it is the only XML specification uniquely focused on business reporting." Like XML more generally, XBRL "uses tags to give context and structure to content [and] provides a common way for disparate systems to exchange specific information." But XBRL (originally termed XFRML, or "eXtensible Financial Reporting Markup Language) goes even further. It "establishes uniform categories for financial data, yet remains flexible enough to accommodate any company's internal environments, processes, [and] systemsä"
Adding Meaning to Data
Simple as it is in concept, XBRL has profound consequences. It adds intrinsic meaning to otherwise static and sterile data. "In a spreadsheet,ä data can be inserted into the rows and columns," writes the Institute of Management Accountants' Liv Watson in Strategic Finance magazine. "Just as the headings for the rows and columns give meanings to the individual cells, [XBRL] provides meaning to the data within tags that bracket the data. For example, if the cell of a spreadsheet contained the number $45,000, the user of the spreadsheet might not know if the $45,000 was an asset, liability, net income, or the price of the owner's new car."
XBRL instantly clarifies this uncertainty. If the content of the cell or the bracketed data is appropriately tagged with, say, the description "net income for the first quarter," says Watson, then the user and any associated computer applications will interpret the $45,000 to represent exactly what it is--the net income for the first quarter. Moreover, XBRL formats automatically carry the data definitions with them wherever that data is transferred. As a result, explains Watson, "XBRL provides both content and structure (context) to the financial data."
Contrast this clarity will the complications that arise from the current methods of transferring financial data. Traditionally, one might use an Adobe Acrobat document, a Microsoft Word file, or a plain text file to transfer the information. "As convenient as these file formats are for human readers," writes Watson, "they don't make it easy for a computer to extract the information." In fact, to continue the above illustration, the figure "$45,000" as net income in an Acrobat or Word document is virtually indistinguishable from "$45,000" used in any other fashion, because computers generally cannot interpret context except in cases in which that context is precisely structured and coded.
Another option: export the data in a numerical file format, like a comma-delimited file. While this approach eliminates the contextual problem (the data carry the context along), such numerical formats tend to be highly customized, and so require translators at each point of interchange. Moreover, if one of the corporate parties decides at some later point to alter the file format to accommodate other participants, this creates a cascade of interpretive problems for other parties connected to the system, potentially making their former file translators incompatible with the new file formats.
With the common formats and data definitions that are at the core of XBRL, however, no such translators are necessary. "The XBRL language," says Watson, "is designed to export data that's self-describing." In the same way that team uniforms make otherwise indistinguishable players instantly recognizable to sports spectators, XBRL tags make otherwise indistinguishable data instantly recognizable to computer applications. The result is that data can be seamlessly transported among various XBRL-compliant computer systems, which will be able to import and read the data almost as easily as if they were the same system.
The Benefits of XBRL
The bottom line: the ease with which XBRL makes it possible to transfer financial data offers significant business benefits. Analysts have identified three as being the most significant.
First, XBRL minimizes the need to re-enter data. "In most cases," notes the Journal of Accountancy's Stanley Zarowin, "financial information will need to be keyed in only once, reducing the risk of data-entry error. Also, because the information is already XBRL-formatted, users won't need to reformat it when preparing it for any number of presentations, such as to print a financial statement, create an HTML document for a company's Web site, ready an EDGAR document for filing with the SEC, or [prepare] other specialized reporting formats such as credit reports and loan documents." Such an ability not only reduces preparation costs and reduces re-keying errors," he explains, "but it also improves investor or analyst access to information."
These capabilities are particularly critical in view of the increasing use of the Internet as the transfer and display vehicle for financial information. The XBRL Consortium cites surveys showing that approximately 80% of major U.S. companies make some form of financial disclosure on the Web, and predicts that, "soon, every public company will have a Web site where they will make their financial information available to interested parties." The ability to post this information to the Web without re-keying or reformatting--a truly laborious and error-prone process--will ensure the accuracy and timeliness of data that investors and others will come to rely upon for making multibillion-dollar decisions.
Similarly, notes KPMG's Zachary Coffin, cross-national companies will eventually be able to use XBRL-formatted data to quickly analyze and publish the financial statements of its subsidiaries and affiliates in different countries, with different languages, in different currencies, or using different accounting standards and practices, thereby further minimizing the commercial barriers of time and distance that the Internet itself has started to erase.
A second key benefit of XBRL is that it enables fast and accurate Internet searches. "Because all the data in an XBRL-formatted file is tagged and related information is linked--say, fixed assets to balance sheet and depreciation," explains Zarowin, more than half the job of conducting a search for specific information is already done. For example, if you search the Internet for information about General Motors' fixed assets, you likely will end up with thousands of sites to explore, and that's even with the best search tools. But with XBRL-tagged data," he goes on, "the search is immediately narrowed to your specific target data--fixed assets as they relate specifically to balance sheets and depreciation."
Ian Wright, U.K. head of corporate reporting at PricewaterhouseCoopers, LLP, provides an even more pointed illustration of this capability. "Say, for example, that you wanted to compile the research and development expenditure for the top 100 UK companies. You could take one of two routes: write to each company secretary for a copy of their annual report; or find each company's Web site, access its online reports and accounts, and trawl through to find the relevant figure. Amassing the figures would take hours, perhaps daysä" Compare that process with one supported by XBRL. "Imagine how much easier it would be," he says, "if you could simply access a specially designed financial reporting browser and tell it to åfind R&D expenditure' for these companies. And then it automatically slots the data into a spreadsheet." The outcome: "Instant, relevant, accurate information at almost a single click."
Benefit three: improved financial analysis. "XBRL helps companies, analysts, investors, and other audiences to make better informed financial and management decisions," Barry Melancon, president and CEO of the American Institute of Certified Public Accountants (AICPA), bluntly states. The reason, says Glen Gray, professor of accounting at California State University, Northridge, and a member of the XBRL Steering Committee, is that "XBRL enables users to access financial reports in a matter of seconds and move the data to the users' analytical software with literally åa click of a mouse."'--significantly reducing the mechanical aspects of financial analysis and providing more time for analysis and review.
Among the consequences of this enhanced data flow, Gray offers, is that it "could decrease credit costs and increase analysts' coverage of smaller public companies because of the lower costs of analysis." Walter Hamscher of PricewaterhouseCoopers, LLP, suggests another possibility. He notes that, "when financial reports are hard to interpret, to analyze, and to compare, investors perceive greater investment risk and become more reliant on secondary (perhaps unreliable) sources." Increasing the comparability and reliability of financial information, as XBRL would do, could therefore improve the efficiency of investment markets, and so better protect stockholders' investments.
Short History, Many Achievements
For all its potential, XBRL actually boasts an incredibly short history. It did not even originate as a concept until April 1998, when Charles Hoffman, a certified public accountant (CPA) with the Tacoma, Wash., accounting firm Knight Vale and Gregory began investigating XML as a possible tool for enhancing the electronic reporting of financial information. After securing the interest of the AICPA, the primary U.S. professional organization for CPAs, Hoffman spent the rest of the year developing an XBRL prototype and, by mid-1999, had won the AICPA's funding and full endorsement.
Since that time, the XBRL Consortium has published the official XBRL Financial Statement and the U.S. GAAP Commercial and Industrial taxonomy, and is actively working on more than a dozen other projects, including XBRL specifications for regulatory filings, assurance schedules, tax filings, and general business reporting. Already, six Global 2000 companies have begun field-testing the XBRL specifications in a variety of financial reporting contexts. Indeed, the XBRL Consortium has garnered the support of more than 65 top accounting, accounting software, management consulting, computing, and technology companies--numbers that are growing monthly. Included in this roster are such industry leaders as IBM, Microsoft, Oracle, PeopleSoft, Arthur Andersen, Deloitte & Touche, Ernst & Young, KPMG, and PricewaterhouseCoopers.
But best news, concludes Stanley Zarowin, "is that XBRL is free"--a characteristic that should greatly accelerate its adoption as a financial and business reporting standard. Because of XBRL's open-source nature, says Zarowin, "companies will not have to pay to include the XBRL code" in their software or financial reporting systems, "and each new editionä and extension developed for specific industries will be available free for downloading off the Internet."
That's a price even the beleaguered builders of the Tower of Babel could have afforded to pay.
ADVERTISING FEATURE
