Key Takeaways
- A Scrum of Scrums is built to surface cross-team dependencies — the second of the three things that erode a committed release date. It surfaces them as status, not as days.
- “Still blocked” sounds identical in week two and week six. The sentence does not degrade. The date does.
- The sync is weekly; dependency state changes continuously. The interval between meetings is where options quietly close.
- The fix is arithmetic, not insight: price each unresolved dependency in days against the committed date and its 50/85/95% confidence band, so the round-robin ends in a decision instead of a nod.
It is Wednesday, eleven o’clock, and the Scrum of Scrums has started on time for once.
Round the circle it goes. Payments is still waiting on the identity service. Platform is still waiting on the schema migration. Mobile is fine. Someone types “still blocked” beside two lines in a shared doc, the meeting ends four minutes early, and everybody takes that as a good sign.
Next Wednesday, the same two lines. The week after that, one of them clears. Nobody in the room is being careless — every single thing that needed saying got said, out loud, to the right people.
And yet when the release slips in week nine, the retrospective traces the cause straight back to that Wednesday meeting, where it had been named eleven times.
The confidence vote was real engineering judgment
Go back to PI planning. Your ART ran a confidence vote — five fingers, a number, then a decision to commit. That vote is the best-informed artefact in the entire plan. The people who would actually do the work weighed which throughput window was representative, whose capacity was genuinely available, what was really in scope, and where the risk sat. It is expert judgment, and it is usually good.
What it is not is durable. A confidence vote is a snapshot. It describes the world on planning day, and planning day ends.
From there, three forces quietly move a committed date: scope creep, dependencies surfaced too late, and capacity changes. Your Scrum of Scrums exists almost entirely to catch the second one — and it does catch it. It surfaces the blocker. What it cannot do is tell you what that blocker has already cost.
“Still blocked” is a status, not a number
The gap is a units problem. A Scrum of Scrums trades in status: blocked, unblocked, at risk, fine. The committed release date lives in days. Nothing in the standard round-robin converts one into the other.
So “still waiting on identity” sounds the same in week two as it does in week six, even though by week six that identical sentence means something very different to a 7 June commitment. The words hold steady. The date does not.
The other half of the problem is the interval. The sync happens weekly; dependency state changes continuously. A blocking link is added on a Tuesday afternoon, and nobody reads its effect until the following Wednesday — by which point the options that were open on Tuesday have quietly narrowed. This is the same structural reason your org chart drew the dependency map long before anyone wrote a ticket: the boundaries are predictable, but the moment they bite is not.
Give the round-robin a number
This part is arithmetic rather than insight, which is exactly why software should be doing it.
Advanced Release Planning reads blocking issue links on in-scope work directly from Jira, re-runs a Monte Carlo simulation over your team’s real completed-work history, and reports where the committed date now sits against its 50/85/95% confidence band. Crucially, it attributes the movement: this many days from scope, this many from dependencies, this many from capacity.
The team still owns every assumption — which throughput window counts, whose capacity is real, what is in scope. The app does not second-guess the vote. It does the vote’s arithmetic at simulation scale and keeps the answer current between meetings. (See how it works in the docs.)
What changes in the room is small and significant. Instead of “payments, still blocked,” the line becomes: payments dependency unresolved nine days, six days against the committed date, confidence now 68%.
Nobody argues with that sentence. They make a decision about it — escalate, re-sequence, or cut something — in week three, while all three are still genuinely available. On a fixed cadence that matters even more, because the train departs on time whatever is on it; the only live variable is scope, and scope decisions are only cheap early.
Renegotiate before you slip
The goal is not a better-run meeting. It is to stop discovering in week nine what was already visible in week three.
That is the entire difference between a date that slips and a date that gets renegotiated. Renegotiation requires knowing the cost while the options are still open — which is why it helps to understand the three things that erode a release date before your next planning event, not after your next slip.
Your Scrum of Scrums is already doing the hard part: getting the right people in one place to say the uncomfortable thing out loud. It just needs the number that makes the uncomfortable thing actionable.
See what your cross-team dependencies are costing your committed date, in days. Advanced Release Planning for Jira runs on Atlassian Forge — your data never leaves Atlassian — and is free to try for 30 days on the Atlassian Marketplace.
Leave a Reply
Your email is safe with us.