| MARY LYONS | Pittsburgh, PA |
| Email Address: lyons88@verizon.net | Phone: 412-421-1924 |
Home Quality Assurance Project Management Data Analysis
The first step in QA is to identify a key process that has problems. For simplicity sake, we will determine that our starting point is a monthly report to a major client. The next step is intuitive, yet many companies concentrate a QA program in the wrong place - at the end of the process, or in this case, analyzing and structuring the entire QA process around the final report result. Analysis of the report is essential to determine where the problem lies, however, the correction should be implemented with the raw data that produces the report, not the final report itself.
Implementation of long-term QA processes takes time and resources, therefore, placing temporary fixes is important. In this case, since the fix is only temporary, it usually is a good idea to place it at the end of the process since this requires the least resources. Often it is already being implemented. For example, in our example of an incorrect report, we hold someone accountable to review the report and make the necessary corrections. At this point, we may also be able to make a 'quick fix' to the report by running some additional computerized processes to make the necessary correction.
Documentation of the above process serves two purposes. First, it provides a procedure to follow so the report is corrected. Second, since it is documented, it is available to multiple people and, therefore, can be performed by more than one person. Too often, the information is automatically performed by a single person but if they are out of the office no one knows the process.
Now, the permanent QA process begins - analyze the entire process, beginning at step one, the data sources. Determine the reliability of the raw data, if the data is loading properly, the criteria placed on the data load, restrictions placed on the data load, calculations, etc. Run tests against the raw data to determine if the data is being properly categorized, if calculations are correct, if data is either be eliminated or duplicated, etc.
Look at the process to see if there are any redundancies - either in the automation process or a manual process (for example, two people are performing the same task or two departments are generated the same data).
Automate as much of the process as possible and make the automation as easy as possible for the individual running the report.
Develop audit trails so incorrect data becomes conspicuous. For example, check record counts, subtotals, grand totals, and new records. Develop audit reports that are checked to assure data integrity.
Document the process and train, not only the person accountable for the report, but at least one backup person.
Determine the results of the implementation of the QA program. For example, productivity increases, reduction of errors, and ability to meet deadlines.
Continue to seek ways to improve the process. Often, as the QA process continues, links can be established with other processes to further reduce redundancy. Continual vigilance, improvement of processes, continual communications, and training are vital for QA to be successful. The end result is increased productivity, reduced costs, less stress, improved performance, and meeting the ultimate goal of any company -- improving customer satisfaction.
