Note

Access to this page requires authorization. You can try signing in or .

Access to this page requires authorization. You can try .

Best practices for Agile product management

Azure DevOps Services | Azure DevOps Server | Azure DevOps Server 2022

This guide helps product managers get started with Azure Boards. It summarizes practical recommendations for configuring teams, planning work, and using boards, backlogs, iterations, and Delivery Plans to deliver value predictably.

Note

If your team follows Kanban or Scrum specifically, see About Boards and Kanban or the Scrum tutorials.

Most recommendations apply to both Azure DevOps Services (cloud) and Azure DevOps Server (on-premises). Some newer features, such as the latest Analytics views and some portfolio planning previews, appear in Azure DevOps Services first and might not yet be available on Azure DevOps Server.

Tip

You can use AI to help with Azure DevOps tasks. See Enable AI assistance with Azure DevOps MCP Server to get started.

Configure teams

Define a team for each delivery group that works autonomously. Configure teams along value streams so each team can plan, track, and deliver independently while still rolling up to product-level roadmaps.

Recommendations:

  • Create a team for each feature or delivery group. Keep teams small enough to stay nimble (typically about 5–9 members) and large enough to deliver meaningful value.
  • Give each team its own area path and iteration cadence.
  • Use team settings to assign default area and iteration paths so work items the team adds inherit the correct context.

More information:

Configure iterations

Define iteration paths (iterations) at the product level, and then assign teams to the appropriate iterations. Keep a consistent iteration cadence across related teams when it helps coordination.

Recommendations:

  • Choose a common cadence for teams that deliver together (typically one to four weeks).
  • Create at least six iterations to support planning for the next three to six months.
  • Use iterations consistently for forecasting and iteration planning.
  • Consider continuous flow approaches for teams that can deliver incrementally without fixed time boundaries.
  • For flow-based teams, focus on work in progress (WIP) limits rather than iteration capacity.

More information:

Choose work item types

Pick the work item types that match how your teams plan and deliver work. Map product-level work (epics and features) to team-level work (requirements) and optionally let teams break work into tasks.

The team-level requirement type depends on your project's process:

Process Portfolio types Team-level requirement Estimation field
Agile Epic, Feature User Story Story Points
Scrum Epic, Feature Product Backlog Item Effort
CMMI Epic, Feature Requirement Size
Basic Epic Issue (none by default)

Recommendations:

  • Use Epic to represent large initiatives that span multiple features or releases.
  • Use Feature to represent customer-facing value that ships as a coherent capability.
  • Use the team-level requirement type for work scoped to a single iteration.
  • Use Task for developer work that fits within an iteration.
  • Decide how teams handle Bugs—as backlog items alongside requirements, as tasks under requirements, or tracked separately. For the project-level setting, see Show bugs on backlogs and boards.

More information:

Create and maintain your product roadmap

Use the Features backlog as your product roadmap. Have product managers order and refine features. Let teams decompose features into backlog items and optionally tasks.

Recommendations:

  • Keep the Features backlog ordered.
  • Break features into sized requirements that teams can complete within iterations.
  • Review and refine backlogs regularly through backlog refinement sessions.

Features backlog

Product managers create and order features in the Features backlog. Each feature should represent a shippable capability.

👁 Screenshot showing a features backlog.

Team backlog

Teams add requirements—called Stories, Backlog items, Issues, or Requirements depending on your process—to the team backlog, size them for an iteration, and map them to parent features.

👁 Screenshot showing a product backlog with user stories.

Recommendations:

  • Size requirements so a team can finish them in a single iteration.
  • Keep acceptance criteria and the definition of done clear.
  • Map unparented work to the appropriate Feature.

More information:

Forecasting and roadmapping

Use the Forecast tool and team throughput to forecast when features might ship. Forecast requires estimates (Story Points, Effort, or Size) on requirements. If you prefer simple forecasting by count, assign estimate = 1 for each requirement.

Recommendations:

  • Establish a consistent estimation approach across teams that feed a common product roadmap.
  • Use Forecast to model several iterations ahead and validate assumptions.

👁 Screenshot showing a forecast of a product backlog with velocity settings.

Manage dependencies

Track cross-team dependencies by using Predecessor/Successor links and by surfacing dependencies in Delivery Plans. A Delivery Plan is an interactive multiteam calendar view that shows scheduled work across teams and projects.

Recommendations:

  • Tag dependent work with a consistent tag (for example, dependency) for quick queries.
  • Use Predecessor/Successor link types to capture formal dependencies.
  • Visualize dependencies in Delivery Plans or use query-based reports to triage blocking items.

👁 Screenshot showing dependency lines between linked work items.

More information:

Note

Marketplace extensions, such as Work Item Visualization, can help visualize relationships. However, they're community-supported and aren't maintained by the Azure Boards team.

Work in iterations

Use the sprint backlog and taskboard to plan and deliver iteration work. Update statuses daily so progress charts remain accurate.

Recommendations:

  • Plan each iteration with the team and define a goal.
  • Ensure work items assigned to the iteration have clear value proposition and acceptance criteria.
  • Update remaining work and status throughout the iteration.
  • Monitor dashboards and charts to track throughput or blockers.

👁 Screenshot showing an Analytics Sprint burndown chart.

More information:

Review progress and delivery

Use the Features board, rollup columns on the Features backlog, and Delivery Plans to review progress across teams.

Recommendations:

  • Add rollup progress or totals to the Features backlog to monitor completion at a glance.
  • Customize Features board columns to match your delivery lifecycle (for example: Research, On Deck, In Progress, Customer Rollout).
  • Use Delivery Plans to coordinate cross-team dates and dependencies.

👁 Screenshot showing a customized Features board with multiple columns.

More information:

Process improvement

Make continuous improvement part of your cadence. Use retrospectives, velocity charts, and dashboards to identify improvements and track progress.

Recommendations:

  • Hold regular retrospectives and capture improvement actions.
  • Use throughput and cycle time to understand and improve your flow of work.
  • Track improvement work on a dedicated board or backlog.

👁 Screenshot showing an example team Velocity chart.

More information:

Optimize work flow

Control WIP to improve delivery predictability and reduce cycle time. Whether your teams use iterations or continuous flow, limiting WIP helps teams focus and deliver value faster.

Recommendations:

  • Set WIP limits on board columns to prevent overcommitment.
  • Monitor cycle time from start to delivery for each work item type.
  • Use cumulative flow diagrams to visualize bottlenecks.
  • Focus on completing work before starting new work.

More information:

Next step

Related content


Feedback

Was this page helpful?

Additional resources