Flow AnalyticsMarch 15, 2026· 6 min read

Cumulative Flow Diagram: See How Work Moves Through Your Workflow Over Time

The new Cumulative Flow Diagram replays your work item state history day by day to produce a stacked area chart that reveals WIP accumulation, throughput drops, and queue build-up across six workflow stages.

#cumulative flow diagram#CFD#flow analytics#WIP#workflow#azure devops

A Cumulative Flow Diagram (CFD) is one of the most powerful tools in a flow-based agile practice. It shows how many work items sit in each stage of your process on each day of a date range. When a CFD is healthy it looks like a set of parallel bands moving steadily to the right. When something is wrong — WIP is accumulating, throughput has dropped, or a queue is building — the bands diverge, and the problem becomes immediately visible.

Six Workflow Stages

The CFD in Agile Analytics maps work items across six stages:

  • New — items created but not yet in any active queue
  • Queued — items waiting to be started
  • In Progress — actively being worked
  • Test — in verification or quality checking
  • Review — awaiting approval or sign-off
  • Done — completed and closed

Built from State Event History

The chart is not a snapshot of the current board. It is built by replaying each work item's state transition history day by day across the selected date range. This means the chart shows what actually happened over time — not just where everything sits today.

Workflow Mapping Integration

If your team uses custom state names (for example "On Deck" instead of "Queued", or "Code Review" instead of "Review"), the CFD uses your saved workflow mapping from Configuration → Workflow Mapping to assign each state to the correct stage. Teams that have not yet configured a workflow mapping will see generic stage names until the mapping is set up.

What to Look For

A few patterns to watch on your CFD:

  • Widening "In Progress" band — WIP is accumulating faster than it is being completed
  • Flat "Done" band — throughput has dropped, likely due to a bottleneck upstream
  • Narrowing gap between stages — a queue is clearing, often after a blocker is resolved
  • Sudden step changes in any band — a batch of items moved at once, which may indicate work sitting undocumented in a state for too long

Need more than a blog explanation?

Flow buyers usually want cycle time and WIP monitoring next. Use the use-case page or go directly to plan comparison.

Related reading