Using Plandek’s end-to-end analytics to improve delivery traceability
Traceability (sometimes spelt ‘traceability’) is an increasingly discussed concept in agile software delivery. The core principles of Agile and DevOps recognise the need for a smooth and regular flow of work through the end-to-end delivery process (or DevOps value stream), resulting in the early and continuous delivery of value to the customer. Hence traceability for teams and functions across the process is a key consideration.
Traceability is often referred to in the context of ‘requirements traceability’ and product validation. This reflects the need to ensure that the product that is delivered to the customer accurately reflects the requirements defined at the outset. Effective requirements traceability ensures that the key functional requirements of the end product can be traced through the design, build, test, and deployment stages, of the live product.
Traceability becomes particularly important in security-dependent and regulated industries where an ‘audit trail’ is a key obligation.
Plandek is a BI and analytics tool that provides end-to-end metrics across the delivery process. Its deep-dive analytics capability is designed to provide improved traceability down to the ticket level – not only for product validation reasons but also to help delivery teams deliver software more effectively.
The end-to-end delivery process is interconnected, hence without traceability, you cannot understand why delivery velocity is slow, team dependability is poor, or defect rates are high – to name but three examples.
Plandek – integrating data from multiple sources for improved traceability
Plandek works by mining data from multiple toolsets used by delivery teams across the delivery process (such as Azure, Jira, Git, CI/CD tools, and Slack), to provide end-to-end delivery metrics/analytics to optimise software delivery dependability, risk management, and process improvement.
Mining data from multiple toolsets creates a unique perspective, enabling Plandek to identify bottlenecks and opportunities for improvement throughout the design, development, integration, test, and deployment processes.
This perspective also enables traceability down to the ticket level across the process.
Some key metrics to improve traceability, thereby improving delivery effectiveness
Plandek can surface a myriad of metrics – many of which directly or indirectly provide improved traceability. The metrics are grouped into six areas:
- Value delivery metrics
- Delivery efficiency metrics
- Delivery dependability metrics
- Backlog health metrics
- Delivery and engineering quality metrics
- DevOps metrics
At the highest level, Lead Time (also known as Time to Value) is an example of a value delivery metric that tracks a story or epic through the end-to-end delivery process, to show the time taken in each of the stages involved and hence the overall time to deliver (from design to live).
As such, this metric is a very good example of traceability at work. It requires data pulled from workflow management tools, code repos, and CI/CD tools to trace the flow of tickets through all the stages of the end-to-end delivery process.
This improved traceability enables teams to increase delivery velocity as they can see for the first time, the bottlenecks in the process and where there is scope to reduce time to value.
Within delivery and engineering quality metrics, measures such as Commits without a Pull Request and Commits without a Ticket Reference both directly improve process traceability and hence improve quality and infosec.
The former ensures that all code is peer-reviewed before being committed (an important security requirement) – and the latter ensures the clear linkage between committed code and Azure/Jira tickets, for security compliance.
There are additional metrics within the DevOps metrics library that are key to improving traceability and thereby improving delivery effectiveness.
DevOps practitioners can track a range of metrics including Number of Builds, Build Failure Rate, Deployment Cycle Time, and Flakiest Files which identify fragile source code files in your codebase causing failed builds and which can be targeted for refactoring.
As such Flakiest Files is a very good example of effective traceability at work.
As shown below in the example of Build Failure Rate, Plandek’s powerful drill-down capability enables traceability of sources of failure by workflow and any relevant filters.
Typically clients can expect to significantly increase deployments per day (per pipeline) through this improved traceability and hence better understanding of the root cause of Build Failures and Deployment Cycle Time using Plandek.