Development team at their workstations in a brick loft office, one engineer pointing at a monitor beside a sticky-note board

Workshop Guide

DevOps and CI/CD Workshop

Map how a change really travels from idea to production, measure where it waits, and agree on the automation and practice improvements that will speed delivery without adding risk.

Steps
Map→
Measure→
Assess→
Prioritize→
Automate

01 / Summary

What Is the DevOps and CI/CD Workshop?

The DevOps and CI/CD Workshop is a structured session in which development, test, operations, security, and business stakeholders map their delivery pipeline end to end, measure how long work actually takes at each step, and agree on the improvements that will move changes to production faster and more safely. CI/CD stands for continuous integration and continuous delivery or deployment: the practices and automation that let teams integrate, test, and release small changes frequently.

The workshop uses the Scaled Agile Framework’s Continuous Delivery Pipeline as its map: Continuous Exploration, Continuous Integration, Continuous Deployment, and Release on Demand. Teams record active time, wait time, and the percentage of work that arrives complete and accurate at each step, baseline the DORA software delivery metrics, assess their DevOps practices using SAFe’s CALMR approach (Culture, Automation, Lean flow, Measurement, and Recovery), and leave with a prioritized, owned set of enabler backlog items.

1
Makes the Whole Pipeline Visible
Development, test, operations, and security see the same end-to-end map for the first time, including the handoffs no single team owns.
2
Finds Where Work Really Waits
Measuring wait time next to active time shows that most delay usually sits between steps, not inside them, so the group targets the real bottlenecks.
3
Turns DevOps Into a Plan
Improvement ideas become prioritized enabler backlog items with owners and target metrics, not a wish list of tools.

02 / Application

When to Use the DevOps and CI/CD Workshop.

Use this workshop any time a team or Agile Release Train wants to release more often, more reliably, or with less manual effort, and needs a shared, evidence-based view of what is slowing its pipeline down.

Slow or Infrequent Releases

Features are finished in the iteration but take weeks to reach production, and nobody can say exactly where the time goes.

Painful, Risky Deployments

Releases depend on long manual checklists, after-hours deployments, or a few key people, and failed deployments are hard to recover from.

Launching or Maturing an ART

A new or growing Agile Release Train needs a shared Continuous Delivery Pipeline and a baseline before it scales the number of teams contributing to it.

Planning DevOps Investment

Leaders are deciding where to invest in automation, tooling, or platform work and need data on which improvements will pay back fastest.

Prerequisites

The Right People Across the Pipeline

Someone who knows each step is in the room: product, development, test, operations, security, and release management. A pipeline mapped by one function is only one function’s view.

Real Delivery Data

Recent deployment history, lead time samples, and incident or rollback records are gathered in advance, so step times and DORA metrics are measured rather than guessed.

Roles and Responsibilities

Facilitator
Usually the Release Train Engineer, Scrum Master, or an agile coach; frames the scope, guides mapping and measurement, keeps the conversation blame-free, and keeps the session to time.
Development and Test
Describe how code is built, integrated, and tested today, including manual steps, environment waits, and the causes of rework.
Operations, Security, and Release Management
Describe deployment, approval, monitoring, and recovery steps, and the controls that any improvement must still satisfy.
Product Management or Business Owner
Represents how ideas enter the pipeline and when releases matter to customers, and helps weigh improvements against feature work.
System Team or Platform Lead
Owns shared build, test, and deployment infrastructure, and is usually the owner of many of the resulting enabler backlog items.

Keep the focus on the system, not on people. Long wait times are almost always caused by policies, handoffs, and missing automation, not by anyone working slowly.

Scheduling and Timing

  • Plan for about three hours, including a short break; very large or complex pipelines may need a second session to finish measurement.
  • Run it before PI Planning so the resulting enablers can be planned into the next Program Increment, or at the start of a DevOps or platform initiative.
  • Gather delivery data at least a week ahead: deployment dates, a sample of changes with commit and deploy timestamps, and recent failed deployments.
  • Repeat the workshop every two to three Program Increments to measure progress against the baseline and refresh the improvement backlog.

03 / Preparation

Preparing for the Workshop

The quality of this workshop depends on the room and the data. Without people from every part of the pipeline, the map has gaps; without real timestamps and deployment history, the measurements become opinions and the group optimizes the wrong step.

  • Agree on the scope with the sponsor: which product, system, or ART pipeline is in focus, and where the map starts and ends.
  • Invite people who know each step of the pipeline: product, development, test, operations, security, release management, and the System Team or platform lead.
  • Gather delivery data: deployment dates for the last few months, a sample of recent changes with commit and production timestamps, and records of failed deployments and recovery times.
  • Prepare the space: a long wall or virtual whiteboard with four swim lanes labeled Continuous Exploration, Continuous Integration, Continuous Deployment, and Release on Demand, plus sticky notes in two colors for steps and waits.
  • Download and prepare the Facilitator Presentation and the DevOps and CI/CD Excel Template.
  • Track every workshop activity, from preparation through follow-up, using the DevOps and CI/CD Workshop Checklist. Teams using Jira can import the DevOps and CI/CD Jira Accelerator.
  • Send scheduling communication with agenda and pre-read materials

    Email Template

    Subject: [Team/ART Name] DevOps and CI/CD Workshop – Agenda and Pre-Read

    Hi team,

    We’re holding a DevOps and CI/CD Workshop on [date] to map how our changes travel from idea to production, measure where they wait, and agree on the improvements that will help us release faster and more safely.

    Before the session, please review the attached delivery data and think about the steps you own in our pipeline: what you do, how long it usually takes, what you wait on, and how often work arrives needing rework. Bring rough numbers if you have them.

    This is a blame-free session. We are looking at the system, not at individuals.

    Location/Link: [link]
    Duration: [about three hours, including a break]

    Calendar Invite Copy

    Subject: DevOps and CI/CD Workshop

    Agenda: Frame scope and baseline metrics, map the pipeline, measure flow, assess DevOps practices, find bottlenecks, prioritize improvements, plan enabler backlog items.

    Room/Virtual: [link]

    Please review the delivery data beforehand and come ready to describe the pipeline steps you own.

04 / Facilitation

Facilitation Agenda

Facilitation keeps the group focused on how work actually flows today rather than how the process document says it should. The most common failure mode is jumping straight to tools, buying or configuring automation before anyone has measured where changes wait. The facilitator’s job is to get the real pipeline on the wall, put numbers on it, and keep the improvement conversation tied to the biggest delays.

Presenting the DevOps and CI/CD Process

This agenda guides facilitators through a structured pipeline mapping session built on the SAFe Continuous Delivery Pipeline. Time estimates reflect a three-hour format with a short break; adjust them for the size and complexity of the pipeline. Keep the session focused on a small number of high-impact improvements with owners, not on redesigning the entire toolchain.

  1. Frame the Scope and Baseline the Metrics (15 min)
    • Confirm where the pipeline starts and ends, set blame-free ground rules, and record the current DORA metrics: deployment frequency, change lead time, change fail rate, and failed deployment recovery time.
  2. Map the Current Pipeline (40 min)
    • Walk a recent change from idea to production, placing each step in the Continuous Exploration, Continuous Integration, Continuous Deployment, or Release on Demand lane, including manual steps, approvals, and handoffs.
  3. Measure Flow at Each Step (25 min)
    • For each step, record active time, wait time before the next step, and the percentage of work that arrives complete and accurate; calculate total lead time and flow efficiency.
  4. Assess DevOps Practices (20 min)
    • Rate the team on each CALMR dimension, Culture, Automation, Lean flow, Measurement, and Recovery, from 1 to 5, and note one example behind each score.
  5. Identify the Bottlenecks (20 min)
    • Highlight the steps with the longest waits, the lowest complete-and-accurate scores, and the most manual effort, then agree on the top three to five bottlenecks.
  6. Prioritize Improvements (20 min)
    • Brainstorm improvements for each bottleneck, such as test automation, trunk-based development, environment automation, or deployment pipelines, and score each on impact and effort to pick the highest-value few.
  7. Plan Enabler Backlog Items (20 min)
    • Write each selected improvement as an enabler with an owner, a target Program Increment, and the metric it should move, and agree how progress will be reviewed.

Common Anti-Patterns to Avoid

  • The group maps the documented process instead of the real one; walk a specific recent change through the pipeline and record what actually happened.
  • The conversation jumps straight to tools before measuring where work waits, so money goes into automating steps that were never the bottleneck.
  • Only development is in the room, so operations, security, and release steps are missing from the map and their waits are invisible.
  • Step times are guessed rather than measured; use deployment history and timestamps wherever possible and flag estimates clearly.
  • Improvements leave the room as ideas rather than enablers with owners, target Program Increments, and metrics, so they lose out to feature work.

05 / Follow-Up

After the Workshop

A pipeline map is only valuable if it changes what gets built next. The enablers that come out of this session need to be planned alongside feature work, and the baseline metrics need to be measured again so the team can see whether delivery is actually getting faster and safer.

Expected Outputs

Current-State Pipeline Map
Flow Metrics by Step
DORA Metrics Baseline
CALMR Assessment
Prioritized Enabler Backlog

Distribute the completed DevOps and CI/CD Excel Template and photos or an export of the pipeline map within 24 hours to all attendees and to the sponsor. Add the enablers to the team or ART backlog so they are visible at PI Planning, and post the map and baseline metrics where the team works so progress can be compared against them.

Follow-Up Communication Template

Subject: [Team/ART Name] DevOps and CI/CD Workshop – Pipeline Map, Baseline, and Enablers

Hi team,

Thank you for mapping our pipeline together on [date]. Attached is the pipeline map, the flow metrics for each step, our DORA baseline, and the CALMR assessment.

Our biggest bottlenecks are [bottleneck 1], [bottleneck 2], and [bottleneck 3]. We agreed to start with these enablers:

1. [Enabler] – Owner: [name] – Target PI: [PI] – Metric: [metric]
2. [Enabler] – Owner: [name] – Target PI: [PI] – Metric: [metric]
3. [Enabler] – Owner: [name] – Target PI: [PI] – Metric: [metric]

These enablers are in [backlog / Jira epic] and will be considered at PI Planning on [date]. We will re-measure the baseline at [date / next Inspect and Adapt]. If anything is missing from the map, let [name] know before [date].

Ongoing Next Steps

  1. Facilitator posts the completed Excel Template and pipeline map to [Confluence / Jira / shared drive] within 24 hours.
  2. Product Management adds each enabler to the ART backlog with its owner and target metric so it can be prioritized at PI Planning.
  3. System Team or platform lead confirms capacity and dependencies for the enablers they own before PI Planning.
  4. Teams track the DORA metrics every iteration and review the trend at the System Demo or ART sync.
  5. Release Train Engineer reviews enabler progress and metric movement at the next Inspect and Adapt event.
  6. Repeat the workshop every two to three Program Increments to refresh the map, compare against the baseline, and set the next improvements.

06 / References and Resources

References and Resources

Complimentary Workshops

Value Stream Mapping Workshop

See where work waits between request and delivery.

Read More

Pain Point Identification Workshop

Surface what slows teams down and fix the root causes.

Read More

Capacity Planning Workshop

Balance team capacity against planned PI work so every sprint commitment is realistic and achievable.

Read More

Ready to facilitate?

Download templates and tools to run your own DevOps and CI/CD session.