Launch plan accuracy: what to review
| Field | Value |
|---|---|
| Branch | feat/admin-and-merchant-portals, draft PR #976: feat(admin): portal navigation, financials and monitoring, plus the agent-process work behind them, not merged |
| Project | README.md |
| Issue | None. This is repair work on the plan itself, and per project-creation an item gets an issue when the work is actually intended |
| Servers you need | None. There is nothing to click. The review is a diff on two markdown files you already read |
| Local origins | Not applicable |
Where do I look?
The diff on docs/business/launches/ALPHA.md and docs/business/launches/PROTOTYPE.md.
git diff origin/dev..HEAD -- docs/business/launches/- Exactly eight lines differ and no others:
ALPHA.mdlines 38, 62, 64, 73, 75 andPROTOTYPE.mdlines 16, 30, 38. - The changeset was parsed out of
changeset.mdrather than transcribed, so a typo could not enter the edit, and every Current block matched its target file exactly once.
What changed here?
The launch plan is what every ranking reads, and parts of it were not true.
- Rows read In progress while their only issue was closed. Alpha P0 rows linked no issue at all, and one P0 linked two closed issues about a different subject.
- A ranking that reads a stale row ranks a finished thing, or skips a live one.
All eight reconciliation rows landed. None was skipped.
- The
CLAUDE_ALLOW_PM_BUILD=1door was used, scoped to those two files, on your 1120 decision.
Two issues were filed carrying questions rather than answering them.
#964: design(alpha): position "scheduled lanterns" as going somewhere with a buddy, not as an event, so line 100 stops sharing#326: Strategic Considerations: Rollout, Product, and Messaging with an operator-blocked row.#965: business(alpha): decide what redeeming at the counter means, for the no-POS attribution path, which writes out the question the row has been holding and answers none of it.
All 95 rows were audited and exactly one genuine contradiction was found.
- Live state for the 108 issues and PRs both pages cite was fetched in one pass, not read and assumed.
- The contradiction: "Attribution, no-POS path" cited only a closed issue while reading Needs operator input. The sweep goes 1 to 0.
Line 76 gained the two Prototype P0s that gate it.
- It cited only the master tracker, so the actual gating work was invisible from the Alpha page.
What needs YOUR words, and cannot move without them?
Two rows are still waiting on you.
- Attribution, no-POS path: what does redeeming an offer at the counter actually mean for the staff member on the other side of it? Now tracked as
#965: business(alpha): decide what redeeming at the counter means, for the no-POS attribution path. - ToS and Privacy Policy: how far does the legal work go, on counsel review, on publishing, and on whether a dated draft is enough for an invite-only cohort? Still tracked by
#141: Legal: Create Terms of Service (ToS). - Both are written out in full in
changeset.md.
"Every P0 tracked as an issue" appears satisfied, and its status was left alone.
- Verified mechanically: 43 of 43 rows marked P0 carry at least one issue link, 0 empty.
- What would settle it: your word on whether that row closes when the condition is met, or stays open because new P0 rows keep arriving.
What is deliberately NOT done?
docs/business/launches/README.md was not touched, because the door did not cover it.
- Two items sit in
backlog.mdunchanged: the stale "Needs review" section, and three Prototype rows that also appear in the Not-scheduled table.
Six rows read Built while citing an open issue, and all six are correct.
- Each of those issues is an epic with genuine remaining scope. The row is the alpha-scoped slice of the epic, so Built is true of the row and Open is true of the issue.
- No issue was closed. Closing an epic is not an agent's call.
Eleven rows cite no issue at all, and none of them is a P0.
- Filing against a row that may move is how a link ends up pointing at the wrong page.
The Prototype-to-Alpha placement pass is a COMPARISON, not a move.
- Nothing in
prototype-vs-alpha-placement.mdwas applied, and four of its seven findings are marked as your call.
What open findings touch this project?
- Nothing from
../merchant-and-venue-dashboards/branch-review-0828.md. The review covers code; this is a documentation reconciliation. - One thing worth knowing about how a conflict here was resolved:
backlog.mdwas rewritten by a merge on 2026-08-27 and the conflict was resolved as a union, not a pick. Both branches carried a blanket claim that turned out to be half wrong once the second file was checked. A keep-the-destination resolution would have deleted a real finding silently.
What issue does each piece close?
| Piece | Issue | State |
|---|---|---|
| The reconciliation itself | None. Repair work on the plan | Applied on branch |
| The scheduled-lanterns split | #964: design(alpha): position "scheduled lanterns" as going somewhere with a buddy, not as an event | Open, carries the question |
| The no-POS attribution question | #965: business(alpha): decide what redeeming at the counter means, for the no-POS attribution path | Open, needs your words |
| The API test-coverage row | #962: test(api): 109 exported functions across the six API services are imported by no test in their own service | Open, now cited on line 64 |
| The real-device coverage row | #963: test(alpha): install, push and offline behavior verified across real Android and iOS devices | Open, now cited on line 62 |
| The governance sign-off row | #896: business(alpha): entity, Employee Rights Charter, and governance docs signed off | Open, now cited on line 75 |