Skip to content

Impulse — Launch Vault

Open this folder (~/Impulse/vault) as an Obsidian vault.

The three things here

  • Timeline — the plan of record, Aug 2026 → Sep 2027. Mirrors vault/docs/marketing/launch-plan.md §6–§7 and vault/docs/marketing/content-push.md. Holds no status; edit it by hand when the plan itself changes.
  • Progress — where you actually are, measured against the Timeline. Rewritten by /parselogs. Don't hand-edit anything above the Manual notes heading; it gets overwritten.
  • Impulse - Daily Logs — one plain-English note per day, YYYY-MM-DD.md. Write normally; the parser does the interpretive work.

The loop

Write logs as you go → run /parselogs in Claude Code from ~/Impulse → read Progress.

Weekly is about right. The launch plan already has a Friday numbers review in the operating rhythm, which is the natural place for it.

/parselogs files beyond this vault. It also routes what the logs contain into the rest of ~/Impulse — a PCB change updates the firmware spec's hardware section, a disproved hypothesis lands in vault/docs/agent-notes.md, a settled decision ticks the open question in the launch plan. So the daily logs are the inbox for the whole project, engineering included, not just the launch.

Everything it changes outside the vault is listed in Progress under Changes filed elsewhere, with the log line that justified it. It won't edit code, and it won't write a claim into a spec that the code contradicts — it reports the conflict instead.

What this is measuring toward

Rescheduled 2026-08-21. The campaign moved from early November 2026 to 1 March 2027 and the ship date from March to September 2027, decided in that day's log because school and the day job took the hours the twelve-week plan assumed. The strategic shape: market and finish the nRF redesign for six months, prove demand with the campaign, and only then start FCC certification. Timeline carries the full reasoning, the old→new milestone ID mapping, and the revised cadence rates.

1 March 2027: Kickstarter. Everything before it exists to clear one gate — 2,500+ emails, 250+ paid reservations, and one creator video that visibly outperformed, checked 2027-02-14, two weeks before launch so that "no" is still a decision rather than a post-mortem. If those aren't there, the plan says don't launch, and the reasoning is that finding out costs only time while a failed public campaign costs the one launch news cycle you get.

A second, earlier gate fires first: 600 emails by 2026-11-30. A single go/no-go at the end of a twenty-seven-week runway tells you in February what you needed to know in November. Missing it is a signal to change the hook, not to post more.

Ship target is September 2027, communicated to backers as October.

There is also an advertising budget now — $2,000–5,000 total, phased across the run-up, with no cold paid social in it at any point. Timeline BUD-ADS has the phasing and the reasoning.

The source documents live in ~/Impulse/vault/docs/marketing/ — launch-plan.md (positioning, unit economics, pricing, FCC, safety) and content-push.md (the channel-by-channel playbook). The Timeline is a mirror, not a replacement: anything about why a target is what it is lives in those two files.