Overview
Visual management is a DevOps capability that involves creating visual representations of work processes to provide transparency and insights into progress, performance, and potential issues. Visual management allows teams to see at a glance what work is in progress, what has been completed, and what still needs to be done.
This approach can be used to manage the entire software delivery process, from requirements gathering and development to testing and deployment. Visual management tools can include Kanban boards, value stream mapping, and other visual representations that enable teams to identify bottlenecks, prioritize work, and track progress in real-time.
Using visual management, teams can quickly identify issues and adjust their approach to optimize delivery speed, quality, and customer satisfaction. This approach also helps teams to collaborate and communicate more effectively, breaking down silos and improving alignment across the organization.
Visual management is a critical DevOps capability that enables teams to optimize their work processes and achieve better outcomes by providing real-time visibility into work progress, performance, and potential issues.
Visual management practices
In the context of software delivery, various types of visual displays and dashboards are commonly used to provide teams with information:
1. Card walls, storyboards, or Kanban boards
- Can be either physical or virtual, with index cards representing work items that are in progress.
2. Dashboards or other visual indicators
- Like continuous integration systems, use monitors or traffic lights to show whether the build is passing or failing. Effective visual displays are those that are created, updated, and perhaps discarded by teams in response to issues they are currently interested in addressing.
3. Cumulative flow diagrams
- Can show the cumulative status of all work being done, allowing teams to estimate how long it will take to complete the current backlog.
4. Deployment pipeline monitors
- Show the latest deployable build and whether stages in the pipeline are failing, such as acceptance tests or performance tests.
5. Production Telemetry
- Such as the number of requests received, latency statistics, cumulative 404 and 500 errors, and which pages are most popular.
Overall, visual displays and dashboards should be designed to provide useful and actionable information to teams. They should be updated and adapted over time in response to changes in the team's context and challenges. It's important that visual displays are not just used to manage metrics alone but to drive improvements and resolve underlying issues.
Adoption expectations
| Step | MVP | MVP+ |
|---|---|---|
Create Jira board to represent existing process and working in progress items |
+ |
+ |
Create build process status dashboard |
+ |
+ |
Create dashboard to show delivery flow KPIs |
+ |
+ |
Enable production level telemetry dashboards |
|
+ |
Tools
| Functionality | Tool Name |
|---|---|
Version Control System |
Git |
Version Control Collaboration |
Azure DevOps Repo, Bitbucket |
Build Agents |
Azure DevOps Pipeline, Jenkins |
KPIs and Metrics Visualization |
Delivery Central, JIRA, Power BI
|
Roles
| Name | Responsibilities |
|---|---|
Scrum Master/Team Coach |
Visualize delivery process of specific team |
Release Train Engineer |
Visualize delivery process of ART and streamline delivery process encouraging automation and shortening manual verification steps |
Product Owner |
Define scope and provide acceptance of new features on the earliest stages of the process |
Build Engineer |
Integrate monitoring and observability solutions |
System Engineer |
Setup monitoring and observability solutions for infrastructure elements |
App Admin |
Setup monitoring and observability solutions for applications |
