Last week I discussed the objective “Maintain high project data quality”. I’m going to keep on the quality theme but switch to the objective “Provide high quality deliverables”. Deliverables mean different things depending on what business you are in. For the purposes of this discussion I will assume the deliverable is software.
An entire organization wants high quality software but the decisions that need to be made and the metrics to support those decisions are different depending on someone’s role. For a Business Sponsor the decision could be the balance between getting a product to market quickly with delivering it with some defects. For the Manager of a software development shop continually high defect leaves him with the decision as to whether to keep the QA Manager or not.
The table below shows various roles of people involved with delivering a project, the context within their role and a sample metric with target.
| Role | Context | Sample Metric | Sample Target |
| Portfolio Manager | Understanding the level of quality being delivered on projects aids in decision making about the projects. | Percent of projects in user acceptance testing with more than 5 defects. | 5% |
| Business Sponsor | Since the business sponsor ultimately takes the end result to market, they want to ensure it is high quality. | Number of defects in the product when it is in user acceptance testing. | 0 |
| PMO Director | Should quality be suffering, the Director can investigate the processes that are failing to provide the required quality. | Percent of projects in Production with more than 5 defects. | 5% |
| Project Manager | The project manager is responsible for assessing the project status. | Number of outstanding defects. | 0 |
| Team Member | Team members are the contributors to the deliverables created through the project. | Number of outstanding defects on deliverables assigned to me. | 0 |
By implementing this objective and its supporting metrics across roles the overall product quality will increase. When product quality is suffering, it will be able to be quickly diagnosed and corrected. This results in a happier customer and less support costs once the product is deployed.