Flow metrics explained

Velocity tells you how much your team estimated. Flow metrics tell you how your team actually works. Here's what each one means and why it matters.

The big picture

What are flow metrics?

Flow metrics measure how work moves through a system — not how much someone estimated. They come from the Kanban world, but they apply to any team that moves tickets from left to right on a board.

The four core metrics are Cycle Time, Lead Time, Throughput, and WIP. Add Flow Efficiency as a derived metric and you have everything you need to understand how work actually flows through your team.

The metrics

Five numbers that matter

Cycle Time

The time from "work started" to "work done" on a ticket. It measures how fast the team moves once they pick something up — not how long it sat in the backlog. Cycle time is the metric the team itself can most directly influence.

Deep dive: How to measure cycle time in Jira →

Lead Time

The time from "request entered the system" to "work done". It includes backlog wait, prioritisation, everything upstream. Lead time is what stakeholders and customers actually feel — the number they'd quote if you asked them how long something took.

Deep dive: Lead time vs cycle time →

Throughput

The number of tickets finished per unit of time — usually per week. Don't confuse it with velocity, which counts story points (an estimate). Throughput counts completions (a measurement). It's also the raw input for Monte Carlo forecasting.

Deep dive: Monte Carlo forecasting →

WIP (Work in Progress)

The number of tickets the team has started but not finished. Little's Law says Lead Time = WIP / Throughput. Same throughput, more WIP, longer lead time. That's the math behind WIP limits — they're not a productivity hack, they're arithmetic.

Deep dive: The physics behind flow →

Flow Efficiency

The share of cycle time that's active work vs. waiting. Typical software teams land around 15–20% — meaning 80%+ of a ticket's lifetime is queue time. Developers aren't slow; the system has too many queues. Low flow efficiency is a system problem, not a people problem.

Why not velocity?

Velocity measures story points completed per sprint. Story points are estimates, not measurements. Velocity goes up when teams learn to estimate higher, not when they actually deliver faster. That's why "our velocity doubled" often means "we inflated our points" — not "we're twice as fast".

Flow metrics measure what actually happened, not what someone thought beforehand. They're harder to game because they count real completions against real clock time. That's the whole point.

Try it

Cylenivo measures all five flow metrics from your Jira data. Connect once, understand everything.

Fetching latest release…

Keep reading

Further reading

How to measure cycle time in Jira

Jira doesn't show cycle time natively. Learn what it means and how to get real numbers from status transitions.

Read the guide →

The physics behind flow

Little's Law, Kingman's formula, and why WIP limits aren't optional.

Explore Flow Physics →

The Work-Feedback Loop

A framework for turning metrics into actual change, not just reports.

Read about the WFL →