By: David J. Wardell

© 1992 By: David J. Wardell.  Reproduction or redistribution in any form without written permission is strictly prohibited.

The lack of data quality is costly because after-the-fact quality control, correction and rerunning of reports are extremely labor intensive — and often yield dubious results and dissatisfied customers.

To best understand the problem, we should focus on management reporting, which includes both accounting and non-accounting elements.

Historically, travel accounting systems have done a fair job of accounting for "finance-critical" elements such as ticket numbers and amounts.

However, these systems have done a fairly miserable job accounting for non-financial data. This deficiency has numerous causes and is responsible for much of the travel data problem.

There are three components to travel MIS.

	Phase One data are input by the agent into the CRS and include elements necessary to understand the transaction but that are not specifically related to accounting. Almost anything the agent inputs is accepted, and in some cases up to 40% of all transactions contain errors.
	Phase Two data are MIS information as received, organized, edited and processed by travel accounting computers. These systems not only fail to completely screen agent input errors but can introduce their own problems because of inadequate data base design, reporting practices and transmission or operational faults.
	Phase Three data are MIS-based reports presented to the customer. Before the client receives the reports, however, the agency takes a number of steps to improve their quality. Typically this involves running a report, manually identifying and correcting problems, then rerunning the report.

Expending resources to do a thorough job can eliminate errors, but this is usually cost-prohibitive, particularly if the account is very large.  It is not uncommon for clients to receive reports with 20% to 30% of the transactions containing errors.

The push toward quality in travel service delivery rarely includes a commitment to the MIS database.  Cost cutting usually puts pressure on data processing organizations to such an extent that report quality receives little attention.

Solutions are not particularly accessible. Phase One data errors could be eliminated through an effective point-of-sale device, but such a product does not yet exist.  The immediate solution is limited to training and basic CRS scripting, which is marginally effective.

Phase Two data error improvement requires investing in a reengineered accounting platform.  While incremental improvements to traditional travel accounting approaches are under development, no such fundamental change is likely.

The need to correct data and reports will exist for a long time and, absent better technology, will only increase in difficulty.  However, there are some ways to better understand and minimize manual corrections.

1. It is essential that data processing managers understand the problem and are supported by corporate management.
   
2. Agencies should develop an ongoing assessment of the exact character and extent of their database and report problems to focus energy and gauge success.
   
3. Agencies should isolate areas vital to report quality and develop procedures to eliminate or control problems, concentrating on error prevention where practical.
   
4. Money may be better-spent upgrading technology than cleaning up reports.
   
5. Error analysis models and tools can help maximize scarce human and other resources.

When data and report quality are elevated to the level of other industry problems that have similar financial implications, more effective tools and procedures may be forthcoming.

While not as glamorous as other projects, your data processing management time is perhaps better spent here than anywhere else.