A false measure of success 'I'd rather have an ounce of cure over this 200 pounds of prevention' Web development

摘要

Several years ago, I was working at a major US financial institution on their Web development projects. When I started with this company, I was initially brought in as a consultant to perform Web development. Over the course of the next 5 years, I had converted over to a full time employee with the company and served a variety of roles including developer, team lead, architect, and project manager of Web projects. This experience report examines the changes in infrastructure, workflow and processes during my time with this company and the results of these changes. The companies' Web development teams started with a software process that was not perfect, but did satisfy the teams' internal business customer. Customers were pleased with the rapid turn-around time that the technology teams provided. Any issues or defects that occurred were within acceptable tolerances for the customer's business needs. However, the management teams of the technology division mandated implementation of processes to ensure stability, redundancy, and uptime. Employee goals and financial bonuses were updated to measure qualities such as 99.99% uptime, full redundancy and zero defects. But as the development teams came closer to their technology goals, the cost of projects increased and project turnaround time decreased. Defining "What quality level is acceptable" and "What cost is acceptable" was shifted from the Business teams to the technology teams. Customers were no longer pleased with the performance of their technology counterparts, thus creating a false measure of success.