Bob Schneider is a Partner in 
Wilson So is the Director of Hitachi Consulting Corporation America’s Favorite Pastime: How You Measure Performance Impacts Results
By Mike Willis and Michael Smith Posted on May 9, 2008
Mike Willis was Founding Chairman of XBRL International and is a partner with PricewaterhouseCoopers. Michael W. Smith is a vice president in Gartner Research. In the first installment of this two-part article, they described how baseball’s front office is using an objective-evidence approach (known as “sabermetrics”) to win more ballgames – a method that business managers would do well to emulate in their own operations. In Part Two below, they look at how Enhanced Business Reporting (EBR) coupled with XBRL-GL can help realize for businesses what sabermetrics has achieved for baseball.
Does a pitcher’s ERA or a player’s batting average actually align with the desired outcome, i.e., winning? Not according to the professional opinion of Bill James, a former night watchman for the Stokely Van Camp Pork and Beans plant in Lawrence, Kansas. In Part I of this two-part series, we discussed how James changed the way baseball managers view critical performance measurements. The point is that “what you measure impacts performance,” so why do many financial executives continue to assemble metrics that don’t actually align with corporate performance?
At the source of existing measurements for both baseball and corporate performance is a common thread: tradition. James has been instrumental in scientifically debunking many of baseball’s traditional measures and establishing those more directly aligned with the desired performance outcomes. But he has an advantage that many financial executives don’t: relatively cost-effective access to the underlying transaction-level data from which he can perform his scientific analysis. Typically, financial executives cannot cost-effectively access transaction-level data from within their own organizations.
The limited access to transaction-level data by financial executives results from another tradition, the double-entry accounting system established in 1494 by a Franciscan monk, Luca Pacioli. The traditional use of double-entry accounting methods within reporting systems, based upon the manual summary methods of the 15th century, are focused on organizing and summarizing accounting information.
This double-entry compliance approach to information is based on information summaries (revenues, costs, assets, liabilities) that typically do not facilitate the aggregation of transaction-based unit data that may be more useful for performance measurement purposes. As many financial executives are painfully aware, this limitation of existing double-entry summary-based accounting systems is at the core of their business intelligence and/or corporate performance transformational project efforts.
Let’s look at a simple example. Traditional internal systems can help corporate management with a wide range of compliance-oriented requirements and related questions such as What were sales last period? and What are our inventory levels?. However, these same systems cannot answer more basic and important performance-related questions, such as: How many ‘widgets’ did we sell?, What was the product mix of our sales or orders for the same period?, and What did we sell to XYZ customer last period? At the core of this information problem is the double-entry accounting orientation wherein transaction level details are discarded as the information is summarized and moved upstream from one software system to another.
In order for corporate managers to make the same type of scientific analysis and leverage the outcomes-based measurement approach that James brought to baseball, transparent cost-effective access to underlying transaction level data is needed. However, data warehouses and reporting tools are commonly optimized for accounting-based compliance-oriented reporting (e.g. financial statements), and changing the proprietary design or architecture of the data warehouse to expose this data is simply not a cost-effective solution.
Standardized transaction ledger level taxonomies can expose this transaction unit level data without changing the existing design or proprietary schema of the underlying databases. How to achieve this in a cost-effective manner is where the XBRL Global Ledger (XBRL-GL) comes in. The XBRL-GL taxonomy is a standardized way to describe any ledger-level concept: general ledgers, payable ledgers, receivable ledgers, inventory ledgers, any type of ledger.
The XBRL Global Ledger is NOT a standard description of a “general ledger” containing the 55 million possible ledger account descriptions; rather, it IS a standardized way to describe ledger level information, the database fields common to business operational and accounting systems. As such, the XBRL Global Ledger can be used to both (1) standardize the relevant transaction level details currently included within relevant ledgers, and (2) expose this transaction level data for analysis and inclusion in performance measurement processes.
The application of the XBRL Global Ledger does NOT require the underlying ledger systems to be XBRL compliant. A company’s XBRL Global Ledger standardized description of a ledger can be mapped to virtually any ledger system — from those that are already “XML compliant,” to those accessible via ODBC or SQL, to those that can export a simple text export file — using a simple XML mapping tool that costs less than $500 (here’s an example of how this works).
The heavy lifting here is neither with the technology nor the XBRL Global Ledger; rather, it is in the design of a company-level standardized description of the relevant ledgers. This is an architectural effort that can start simply, but for the greatest benefit requires a fairly deep understanding of the company information processes and requirements, including those that relate to predictive performance metrics.
The double-entry limitations imposed by existing internal compliance processes adversely impacts the metrics used for performance management. Retail sales per square foot (RS/SF) is an example of this limitation in a commonly used key performance indicator (KPI) for the retail consumer products segment. It provides some insights on performance, but is based upon summary level information rather than the more detailed information available at the transaction level. This commonly used KPI does not help management understand the product mix, product through-put, product margins, or rental costs associated with the profitability of the store, business unit, and/or company.
The RS/SF metric is based upon historical data and does little to assist management with predicting future demand for specific products and therefore future net cash flows. The RS/SF KPI is the corporate equivalent of a baseball’s ERA or batting percentage. It is a traditional measure, but its relationship to profitable performance and future net cash flows may be somewhat dubious at best.
Enabling management to cost-effectively obtain and analyze the unit level transaction data from their existing proprietary data warehouses is critical to making the assessments necessary in creating relevant outcome-oriented performance metrics. Ensuring management has the necessary information to timely and cost effectively monitor and manage the increasing complexity of business operations is critical to long-term sustainable performance.
Assessment and analysis of the underlying data leads to development of more relevant and outcome-oriented performance indicators. The XBRL-GL can expose the underlying ledger level transaction data, and XBRL taxonomies can articulate relevant KPIs. Mapping from the underlying data to the KPI reporting summaries used by management is a native “feature” of XBRL.
A corporate equivalent of James’s scientific approach to predictive baseball metrics is needed. The Enhanced Business Reporting Consortium (EBRC), as Robert Eccles has discussed on this blog, provides a collaborative environment for development of an information framework and specific standardized key performance indicators relevant to specific industry sectors. As such, it enables management to cost effectively address the traditional transaction level information access points above.
EBRC is an open collaboration of market participants working to improve the quality, integrity, and transparency of non-financial information. Members of the EBRC are developing a voluntary, global information framework for nonfinancial components including predictive key performance indicators.
The EBRC structured information framework and standardized KPI concepts, articulated via XBRL taxonomies, provide an explicit machine-readable artifact that can be mapped to the wide range of disparate internal data stores, thereby enabling increased transparency of information relevant to KPIs predictive of future net cash flows. The EBRC framework industry sector KPI information can be obtained from the relevant internal data stores in a cost-effective manner, regardless of how the double-entry accounting information is aggregated.
The XBRL and EBRC standards and market-based development processes are now a reality. Company and investor collaboration in the development of predictive standardized KPI concepts is underway. Join this effort to move your performance metrics out of the 15th and into the 21st Century. To learn more about the EBRC KPI collaboration project efforts, please visit http://www.ebr360.org/. To learn more about the application of XBRL to your internal performance processes visit http://www.xbrl.org/.
July 31st, 2009 at 2:20 am
[...] Mike highlighted to me that the traditional “lagging” indicators prevalent in performance management metrics such as EPS, gross revenue increases, current ratio type metrics need to be supported or superseded by more predictive analytics that can achieve deeper “sum of the parts” and “cause and effect” type insight . To illustrate this, he provided a great analogy in the baseball metrics guru Bill James, who, following a deep analysis of contributing factors that led to baseball games being won, established a new set of predictive metrics, called sabermetrics. The traditional “batting average” metric was to give way to the “on base %”sabremetric. This analogy was originally presented in a blog post he wrote with Michael Smith of Gartner, who is spearheading the EBRC consortium. [...]