Avium SignalsTeam tier and above

Cycle Time — how long active work actually takes

Lead time answers "how long from commit to done." Cycle time answers "how long once we actually start." Avium measures both from your real history — no guessing.

Avium SignalsCycle Time
How long work takes, as a distribution — not just an average. The slow tail is the release risk stakeholders feel.
Measured from your real history — P50 is the typical case; P90 is the promise you can actually keep.

"It's in progress" tells you nothing about how long it'll take

A ticket moves to In Progress and… time passes. Is that normal? Slow? You don't know, because "normal" was never measured. Some tickets fly through in two days; some grind for three weeks, and the only signal is a vague sense that this one's been open a while. By the time it's obviously stuck, the sprint has already absorbed the hit.

Cycle time — the clock from first active work to done — is the cleanest read on how fast work moves once it starts. It's sitting in your transition history, unmeasured.

How teams approximate cycle time today

The common proxies all measure the wrong thing:

  • Days between "created" and "closed" — which includes all the backlog-sitting time and isn't cycle time at all.
  • A control chart per board — fine for one team, one board, and useless the moment you want it across teams or filtered by issue type.
  • Story points as a stand-in for time — but points aren't time; a 3-pointer and a 5-pointer routinely take the same week.
  • No outlier visibility — the average looks fine while a handful of multi-week tickets quietly define your release risk.

How Avium measures cycle time

Avium clocks only the active span, from your real history:

  • The clock starts at first active work — the first transition into an in-progress status, not ticket creation — so backlog-sitting time doesn't pollute the number.
  • The clock stops at done — canonical resolution, with a Done-category fallback — and reopened work is counted honestly.
  • P50 and P90 distribution per team / issue type, so the slow tail is visible, not hidden behind an average.
  • Derived from your real transition history (Jira today, or any work management tool) automatically — recomputes as work moves, no manual tracking.

Who reads this Signal

Engineering managers
Know what "normal" actually is, so "this one's taking a while" becomes a number you can act on, not a hunch.
Scrum masters
Catch the slow tail mid-sprint — the tickets drifting past your typical cycle time are tomorrow's spillover.
Delivery leads & PMOs
Pair cycle time with lead time to see where the delay lives — in the doing, or in the waiting before work starts.

See your real cycle time in Avium

Free tier connects your work management tool (Jira today, more integrations on the way) and computes cycle time from first active work to done. Distribution + per-team / issue-type breakdowns on Team; the AI Briefing's interpretation on Business.

Start free — no credit card →

Related Avium Signals