This post is from the Collabnet VersionOne blog and has not been updated since the original publish date.
How Measuring DevOps Performance Increases Enterprise Agility
This is outdated VersionOne V1 post.
Do you know exactly how much value you delivered to your customers last week, month, or quarter? Can you describe, with precision, the cumulative business value contained within each incremental deployment? Here’s how you can start measuring DevOps performance to gain end-to-end visibility from concept to cash and why you can no longer afford not to. Why You Should Measure DevOps Performance Here are 10 reasons you can’t afford not to measure DevOps performance. Measuring the performance of software delivery enables you to: 1- Monitor the flow of business value through delivery 2- Identify and remove bottlenecks and impediments 3- Determine where business value spends time and measure improvement 4- Measure the improvement and reduction of deployment batch size 5- Identify potentially risky packages, releases, and deployments 6- Automate delivery compliance reporting 7- Perform root-cause and post-mortem analysis in real time 8- Quantify delivery performance improvement 9- Prioritize DevOps investments more intelligently 10- Validate the ROI of software initiatives DevOps Performance Measurement (DPM) allows you to track, measure, and analyze the business value flowing through delivery pipelines. Starting with a single commit in a source code repository, DPM methodically traces the granular bits that collectively represent user stories, defects, and features as they progress through delivery streams until finally they can be consumed by your customers. Having insight into how business value is flowing through delivery empowers you to reduce the risk of deployments, automate compliance, optimize bottlenecks, determine the root cause of failures, and make more intelligent decisions about portfolio investments. How Do You Measure DevOps Performance? Simply put, you measure DevOps performance by: 1- Integrating the data generated from software delivery’s manual processes and automated tools 2- Mapping the software delivery data collected across a unified timeline 3- Correlating each bit of the software delivery data back to the originating user stories or defects This includes information that describes source code commits, the continuous integration process, manual and automated testing, artifact deployment, manual approvals, and any other step required to advance value to end users. Unfortunately, gathering this data is not enough. This highly fragmented data must be consolidated into a unified timeline and tightly correlated back to the underlying business value. This final step of associating each bit of data back to a value driver is a critical step — without correlation, you simply cannot measure the end-to-end flow of value. Until now, the DevOps machines we’ve been building have been profoundly “value agnostic". This lack of value awareness has created a short circuit in our ability to effectively measure business value flow. To date, DevOps measurements have mostly focused on operational metrics like build frequency and deployment speed. Not surprisingly, these kind of metrics have little importance to executives and stakeholders beyond those directly responsible for software delivery functions. Tracking the flow of business value has quickly become essential to increasing the strategic importance of DevOps and increasing the overall level of enterprise agility. Increasing Enterprise Agility with End-to-End Visibility With the wide popularity and adoption of agile, enterprises have gotten much better at tracking value through strategic planning and development. Strategic initiatives and themes are prioritized and then decomposed into smaller chunks, epics, features, and finally stories and defects. Teams can track the status of users stories and we now have real insight into how value is progressing through the first half of our value streams. However, once those stories and defects get converted into binary code, the value stream goes dark, creating a black hole into which the organization can’t see. By tracking the user story as it is recomposed into artifacts, packages, releases, and deployments, you extend the understanding of your value streams through delivery, thus providing true end-to-end visibility across value creation and value delivery. End-to-end visibility is the ability to track, measure, and visualize the flow of business value from the initial idea until it is delivered into your customers’ hands. [caption id="attachment_5896" align="alignright" width="600"] View Your Delivery Value Stream with VersionOne Delivery at a Glance[/caption] The progress we’ve made in tracking business value through strategy and development is significant, but it is not enough. Until you fully integrate delivery into your software lifecycle management, you are flying blind. If you are only as strong as your weakest link and you have zero visibility into delivery, then you have no idea how strong you are or how you can improve. Until you fully incorporate delivery into your value stream and measure the flow of value from strategy through development and delivery, you can never be sure where your biggest bottleneck is or the best way to increase enterprise agility. Conclusion I hope this has encouraged you to take a deeper look at how much visibility your organization has into the flow of business value through delivery. To truly increase your enterprise’s agility, you must have full end-to-end visibility and that requires building a strategy around DevOps Performance Measurement (DPM). If you want to learn more, check out the other articles in this series:
- Measuring DevOps Performance Using a Value-Based Approach
- 3 Performance Measures That Can Help Reduce DevOps Wait Time
- 4 Basic Building Blocks Needed to Measure DevOps Flow