Learn about DORA Metrics & Oobeya Deployment Analytics.
As software engineering organizations scale, fundamental software development and delivery processes become more and more complex. The number of tools used in the software delivery life cycle (SDLC) increases as processes become more sophisticated. Different teams within the company set up and run various delivery pipelines. During this process, visibility into the software development and delivery cycles is significantly reduced.
DORA (DevOps Research and Assessment) sought to find a solution to this complexity and invisibility problem. According to the experts in this team , the definitions for 4 Metric Keys as in the following :
  1. 1.
    Lead Time For Changes: The amount of time it takes a commit to get into production.
  2. 2.
    Deployment Frequency: How often an organization successfully releases to production. High-performing software teams release often and in small batches.
  3. 3.
    Change Failure Rate: The percentage of deployments causing a failure in production.
  4. 4.
    Time to Restore Service: How long it takes an organization to recover from a failure in production.
Now we are ready to start a Deployment Analyses that shows these DORA metrics as both the average and individual with the company with their breakdowns.

Start a Deployment Analysis

1. Start in Beginning

While creating the "Gitwiser" analysis, users can tick the "Deployment Analytics" box if they want and continue the process by specifying the "CI / CD" tool they use. The list of supported "CI / CD" tools is as in the following.
  • AzureDevops (Releases Definitions)
  • Github (Github Actions)
  • Jenkins
  • Gitlab (Gitlab CI / CD itself ,i.e, .gitlab-ci.yml)
  • TeamCity
  • None - Git Release (if merging a pull request into a specific branch (e.g., master, production) means deploying a new version to production.)
Note : Each "CI / CD" tool has its own fields. For instance, while we need the "Release Definition" and "Environment" for AzureDevops , on the other side , the "Job" name is required for Jenkins.
2. Start After the Gitwiser Analysis is Created
While creating Gitwiser, the option for Deployment Analytics is not a mandatory area. Thus, it is possible that the Deployment Analytics can also be started after the Gitwiser analysis has been started.
  • Open Gitwiser Analysis, click on 'Deployment Analytics' and then click on the 'Start Deployment Analytics' button.
  • Choose the 'CI / CD' tool and fill in the fields according to them.
Note: You can update or delete the Deployment Analytics by clicking on "three dots" in the upper right part of the page.

Results of Deployment Analytics

When the Deployment Analytics analysis is finished the page will look as follows :
Note: The bages indicates the translations the DORA metrics to systems-level calculations. (Elite, High etc.)
Note: Also, for example, if the period you have selected is "Last Month", Oobeya not only calculates for "Last Month", but also calculates "Last Month" before the last month and shows how much has changed.

Change Failure Rate

Change Failure Rate is defined as the percentage of deployments causing a failure in production.
You can set your deployment status as a failure manually now. The Comment field is a required field that specifies the cause of the problem in the deployment.

Time To Restore Service

Time to Restore Service is defined as how long it takes an organization to recover from a failure in production.
The deployment that appears at the top of the widget is the newest deployment.
Any of these deployments can be selected as 'Failure'. If a problem is detected for any reason in the 'Production' environment, 'Fix' versions are created for this and this is a normal situation in software lifecycle process. If these later deployments were created in Oobeya to fix a previous deployment that has been selected as 'Failure', these deployments can be manually selected as 'Fix Deployment'. If no selection is made, Oobeya automatically identifies the first deployment after the deployment selected as 'Failure' as 'Fix Deployment'.
Copy link
On this page
Start a Deployment Analysis
Results of Deployment Analytics
Change Failure Rate
Time To Restore Service