Title: Determining your Four Key Metrics
Time: 1 hour
Introduction: Improving operability is a worthy aim but where are we improving from and what are we aiming at. The best starting point I have used is the āFour Key Metricsā in the book Accelerate. They are evidence based, data gathered from lots of different organisations as a benchmark. The metrics are:
- Deployment frequency - For the primary application or service you work on, how often does your organization deploy code to production or release it to end users?
- Lead time for changes - what is your lead time for changes (i.e., how long does it take to go from code committed to code successfully running in production)?
- Time to restore service - how long does it generally take to restore service when a service incident or a defect that impacts users occurs (e.g., unplanned outage or service impairment)?
- Change failure rate - what percentage of changes to production or released to users result in degraded service (e.g., lead to service impairment or service outage) and subsequently require remediation.
Purpose: Create a starting point for current operability capability with reliable metrics
Activity:
- Involve your team - its always nice to share your thoughts with your team, especially when looking at metrics. They can be associated with scrutiny, so say you are going to investigate these metrics. Provide the defintions above and the links to the State of DevOps Report and Accelerate. Even better if its a team effort!
- Next step is to look for data. Try the following sources:
- Your deployment pipeline
- Your product backlog
- Your bug tracker and support tools
- Any summary reports prepared for releases
- Analysis of live incidents
- For each of the metrics above, find examples from the last 2 weeks to start with (as many as you can).
- Record your findings in the worksheet that accompanies this lesson:
- Add a summary of what you discovered for each of the key metrics, there are examples within the worksheet.
- If you couldnāt determine through lack of access or the data simply isnāt collected, add that to the barriers section.
- Finally, next to the metric name, try to match your current performance to Elite, High, Medium or Low levels. The details of each level are appended to the worksheet.
- If you donāt believe you have enough to determine a level, just add a question mark. Itās OK not to know. That in itself is useful information!
- Next, either:
- Communicate out to the team, see what conversations get started!
- If you prefer, this is a means for you to check your current operability so use it for your own reference.
- Speak to someone outside the team you check in with for review and advice.
Eventually, this data collection can be automated. With all automation though, its important to understand the subject really well before you dive in.
For more reading, check out the 2019 State of DevOps report:
And Accelerate itself: