“Quality is never an accident. It is always the result of intelligent effort” – John Ruskin
Today, in several organizations, QA environments are managed by a combination of development, QA, project, infrastructure, and other teams. Due to multiple team’s involvement in QA environment management, a lot of the tester’s time and effort gets exhausted in coordination and supporting test environment issues. This is predominantly due to the fact that no team takes up complete ownership of test environment activities.
Some organizations have realized and started setting up dedicated test environment management teams. This test environment management team has to perform activities like incident & problem management, service request fulfillment, environment provisioning, monitoring, coordination and other activities under one single umbrella. As these activities have been historically performed in parts by various other teams, the effort estimation for test environment management becomes quite complex.
The below diagram provides an effort distribution of activities typically performed by the test environment management team.
QA environment effort distribution
* Other factors include any governance controls and management effort for QA environment management service delivery
We can estimate the effort required for providing an incident management service with good accuracy if we know the response, resolution times/ SLAs and the historic ticket volumes. The challenge faced here is mainly due to the lack of availability of historic data.
Service request fulfillment may not be very simple to estimate, because each request might take different times for service. For example, a server restart request might be faster to complete as opposed to a complex application deployment. Moreover, deployment times for each application could also vary.
For environment provisioning, there are multiple parameters to consider ranging from infrastructure, middleware and applications, thereby making it the most complex factor to estimate. Environment provisioning becomes even more complex due to the multi-layered configurations, to be handled by one team.
Environment monitoring efforts are estimated based on the tools used, parameters to be monitored and frequency/ time interval for monitoring. As for reports, we have to consider the tool used, number of reports, frequency and parameters, which are not complex to estimate.
These remain the key parameters governing the effort estimation for test environment management.