Optimize Agile DevOps
Many software development organizations have adopted both Agile and DevOps. They aim to deliver reliable, high-quality software with a high cadence. However, reality is hard. Service delivery teams are not as productive as wanted, and integration and delivery is by far not continuous. They also fail to deliver the quality that they aim to deliver. And customers complain about the pace of innovation of their software provider.
VDMbee founders have a long track record in developing and delivering world-class software. They also have deep experience in continuous improvement of processes throughout the software development organization. Throughout the years, they advised and helped software development organizations around the world, to significantly increase their performance.
They recently developed tooling to empower their service to continuously optimize Agile DevOps, following a measurement-based and data-driven approach.
The VDMbee Agile DevOps acceleration service follows the journey as depicted in the diagram.
Identify metrics & responsibilities
It is determined which roles are involved in the improvement project. Typically starting at the core: service delivery teams (agile development teams). Other roles are involved, based on priorities of performance gaps found, and dependencies between the roles. Per role, it is determined which performance metrics they are, or should be, accountable for. The scope of the project, in terms of both involved roles and their responsibilities, can be expanded incrementally, during the course of the project.
Per metric in scope, achievable performance targets are determined, over the timeline. These objectives can be finetuned during next iterations.
Enable & automate measurement
In most cases, measurement (data collection, to determine metric values) is performed manually at the start.
Where possible, and as soon as possible, measurement will be automated. Measurement can be automated based on data sources that are typical in the Agile DevOps domain, such as Jira, GitHub, Azure DevOps, Sonarcloud, snyk and Polaris. The level of automation can be extended over time.
Measure & assess performance
Based on data inputs, performance metric values in scope are determined, and performance gaps are highlighted. These gaps may be detected in the area of responsibility of service delivery teams themselves. They may also relate to responsibilities of supplying roles in the organization, such as product owner (PO), project management (PM), or architects. Assessment may be extended by comparison with benchmarks of similar software development organizations.
Explore scenarios for improvement
Different scenarios for improvement are simulated, in order to deepen the shared understanding of what may practically and feasibly be achieved. This is based on a model that fundamentally captures influence relationships between the various factors that influence performance in the Agile DevOps domain. Priorities for improvement are determined, based on resulting insights and based on consensus.
A roadmap for improvement is developed, following the priorities as established above. This includes the specification of which practices to implement in the organization, by whom, and when. This is based on a model that fundamentally captures collaboration relationships and processes, with their factors that influence performance, in the Agile DevOps domain. The model also serves as guided knowledge base for relevant associated practices, at various levels of maturity, in multiple dimensions. Multiple roles in the organization may get involved in the improvement project, based on this.
The improvement roadmap will be implemented step by step. Service delivery teams, and other roles involved, will be mentored to implement new practices. According to the iterative nature of the project, performance is measured in order to evaluate the impact of changes. From here, iteratively, the set of metrics in scope, as well as objectives as set, may be finetuned. Next opportunities for improvement maybe discovered, and/or next roadmap steps may be implemented.
The VDMbee Agile DevOps acceleration service, as outlined above, can be delivered via one or multiple engagements. The service can be delivered recurrently, for as long as needed or wanted. Typically, the first engagement leads to first performance gains. It can be followed by a period of monitoring. On-demand, or at agreed intervals, VDMbee can be involved for follow-on engagements.
In case you are interested in our service, you should submit your request via the form below.