We have spent the last couple of months asking what should be reported from testing and to whom it should be reported. However, one of the major issues is that a lot of information and statistics are generated and often ignored until the end of the project. We end up with reams of results, counts, trends (sometimes), comments, analysis and opinions along with change documents and other project documentation with no obvious home and no-one looking at it.
The end result is that decisions are taken without all the required information.
This is obviously not just a QA issue but a project issue, however, our concern is with the QA information that has been generated. Although it may seem obvious, we need to make sure the items that need attention are brought up at the correct time and place for a decision. Many people will colour code their results to make sure that any critical issues are addressed but that does not help if no-one is paying attention. We tend to go back to the Quality Assurance aspect of determining the process that should be followed and of Process Improvement as we tweak that process to work for us. We need to identify who needs the information and from whom decisions must be obtained. Statistics and impact analysis also always help here to identify what could go wrong. A few weeks ago we mentioned trend analysis and thresholds. These are also critical for any of the information generated.
