Global competitive pressures and the current economic situation have put executive finance and operations leaders in the spotlight for controlling operating costs. To improve responsiveness, many process standardization and improvement efforts are under way and subject to strict scrutiny while others are on the horizon. Strict and improved processes will help, but good data is necessary to address the situation.

CFOs now care about master data and data quality because all accurate knowledge about their organization's assets and liabilities stem from it. Little cash actually changes hands in most organizations; transactions often are nothing more than master data flying back and forth.

Distinctions between line and staff jobs aside, operations and front-line managers care about data quality for the same reason – it provides what they need to know about what they manage.

It is no secret that master data is valued as a strategic asset. Thought leaders and practitioners find communicating the value  of master data and quality less challenging than it used to be. Rich content providing a bird’s eye view and multiple technologies are available, but many such initiatives start with a bang and fizzle out. The real struggle is properly allocating money and resources to support master data and data quality initiatives - and ensuring that these efforts are sustained. This article is an experience-driven perspective of the master data challenge, considering “billing” both as a specific instance of data and an information chain that most organizations should improve, and as an example to illustrate the more general points for similar data and information chains.

Illustrating Impact

Obviously, most organizations are both creators and recipients of invoices, although different departments are involved. Business units or divisions are “billers” (i.e., they invoice their customers for products and services provided), and accounts payable receives invoices from suppliers and pays them. These departments have different perspectives on billing based on their respective roles.

Essentially, there are only two important billing issues: overbilling and underbilling. Obviously, the overbilling issue receives the most attention from customers who will review the invoices they receive. The biller, in most cases, has a large customer service department to address billing questions and complaints, make adjustments and so forth.

Some organizations keep reasonable track of how much revenue they rebate in this way (about 1 percent of revenue is typical). They may also know the cost of this error detection and correction. In large organizations, accounts payable may also track how much time and expense it incurs to find errors and seek corrections. Both the organization supplying the bill and the organization receiving it bear added costs. Neither can close the books promptly, and contention over billing does nothing to advance their relationship. The quality system is as depicted in Figure 1. The data quality enthusiast is usually put off by inspection and rework that involves the customer.

Few customers report underbilling, and most organizations have no way of knowing the extent of this issue. But, if an organization overbills, it must:

  • Underbill by about the same amount (or more), or
  • Systematically overbill, a potential fraudulent act.

A billing system is unbiased if the total amounts of overbilling and underbilling are approximately equal. A billing system is biased toward overbilling if it systematically overbills by more than it underbills and biased toward underbilling if it systematically underbills. Assume that the billing system was unbiased sometime in the past. Because many overbilling issues have been reported, it is possible that the root causes of some issues have been eliminated. If so, there is currently more underbilling than overbilling (and the billing system is now biased toward underbilling). In other words, any correction puts the system at risk of an overcorrection, thus creating another issue on the opposite end of the billing spectrum.
In terms of the customer-supplier model, while the feedback channel for overbilling is usually effective, it is nonexistent for underbilling.

To illustrate the overall impact, assume the following:

Thus, it is evident that the cost of poor master data and quality is at least 2 percent of revenue for a simple information chain whose value-added costs are only 1 percent of revenue. A similar calculation can be constructed by accounts payable. Billers should apply information chain management, and accounts payable should apply supplier management to recover much of this cost. Experience confirms that roughly 70 percent of the total costs of poor data quality can be recovered.

From the customer's perspective, the quality system shown in Figure 1 is simply untenable. Supplier management can replace it with the system shown in Figure 2.

Handoffs Need More Focus

From the biller's perspective, the billing chain is not as simple as it may appear. It involves numerous handoffs of data, the result of normal business activity, including value-added activities (see Figure 3) such as the following:

  • Marketing, which yields new customers and pricing schemes;
  • Sales, which yield customer purchases;
  • New product developments, which yield new products; and
  • Customer payments.

Billing issues occur not because the billers or those upstream are inept, but rather because so many hand-offs in the end-to-end billing chain are usually unmanaged. Process management is strikingly effective because it addresses these weaknesses. If these processes still receive or create data that is not unified or of required quality, desired or projected benefits are hard to achieve.

A critical solution that is often an afterthought, but is still an effective approach, involves billers treating internal departments that supply master data needed for billing as they would external suppliers. This is done by implementing data controls by picking handoffs with high risk of errors. This involves applying risk-based controls and Six Sigma DMAIC methodology. Data mastering may not be needed in some cases, but a data governance framework with decision rights is needed to determine what data quality controls and measures are required, the service level agreements for data suppliers and for data remediation. Experience shows that even informal data mastering and a quality management approach has helped to identify millions of dollars of incorrect bills and payments. Shouldn’t this be an approach to the business case for data quality as well as an eyeopener for executives whose active participation and support is needed? Those critical of data mastering and quality management often get on board after being adequately told what’s in it for them.

Unified Master Data and Quality

Look for opportunities to improve master data and quality “where the money is.” Billing is a natural choice for companies. Many organizations find that they can reduce underbilling for little added cost. Incentive compensation, customer and channel management are other operations with multiple opportunities for errors resulting in myriad over and underpayments as well as ineffective marketing and salesforce management programs. Payment and margin assurance processes can be a natural point of governance in some, but trusted master data with sufficient quality is needed for sustainable success of such process improvement initiatives. Engaging appropriate internal and external members is a critical success factor to identify and connect the dots for sustainable success.
Most importantly, do not forget internal organizational challenges that are unique and most often require more focused efforts to resolve than anything else. Optimized operating costs for these specific processes, reduced stakeholder dissatisfaction and management worries are benefits of employing unified master data and improved quality. These also happen to be critical measures of its success.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access