Post-mortems blame the same culprits: scope creep, unclear requirements, technical debt. But ask delivery leads what actually killed the date and a different story emerges — the InfoSec approver was on annual leave, the only person who understood the interface works three days a week, the vendor contact was at their annual conference.
These are capacity collisions: moments when a deadline meets a calendar that was never checked.
Why they keep happening
Every tool in the standard stack is blind to it. The project tracker knows the deadline. The HR system knows the leave. The org chart knows the names. Nobody joins the dots — so the first time anyone notices the collision is when the work stalls and the emails start.
And the failure compounds: the nominated "cover" person is often away too, because nobody checked their calendar either.
What actually prevents them
The fix is structurally simple — it just requires holding three facts in the same place:
- The deadline — on the blocker, or inherited from its project
- The resolvers — the named people who can unblock it, not a queue
- Their real availability — working pattern, booked leave, external commitments, all in their timezone
When those three facts live together, the warning writes itself: "Sarah is unavailable on the deadline. Contact delegate: Marcus." And critically — the check runs on Marcus too, so a delegate who's also away doesn't become the new single point of failure.
Make it a habit
- Name resolvers on every blocker when you log it, not when it bites
- Record delegates for every key person — one named human who covers them
- Review the who's-in view before committing to any date that touches another team
A rose-coloured warning badge in March beats a missed deadline in May — every single time.
That warning is built into OrgBearings — free to start, live in ten minutes.