Everything about how LaunchClan actually works.

A reciprocity network for founders launching on Product Hunt. You list a launch, back other founders' launches you genuinely support, and mutual backing forms a pact — a real commitment on both sides, verified against Product Hunt rather than taken on trust.

A favour you hope gets remembered has no due date and no record. A pact does: a due date, an automatic verification step, and a trust score that tracks whether the people you pact with actually show up. Matching is on category and launch-date overlap rather than “who's most likely to reciprocate” — a pact with someone whose audience doesn't care about your product wastes the commitment for both of you.

Not to join or list a launch. You'll want one connected before your first obligation comes due, though — without it, verification falls back to a reviewed screenshot instead of the instant automatic check.

Product Hunt prohibits incentivised voting and vote rings — bulk clicking with no real interest behind it, which is exactly what its fraud detection looks for. LaunchClan isn't that: matching is on genuine category and launch-date fit, support is never bought (money changes what you can reach, never whether a vote happens), and every commitment is checked against Product Hunt's own record rather than assumed. See how we compare to vote-for-vote groups for the exact distinction.

With your Product Hunt account connected, we ask Product Hunt directly — using your own access token — whether you voted for a specific post. One request, resolved against your account and no one else's, unforgeable because it isn't self-reported. Full detail on the verification policy page.

You can submit a screenshot instead. It's reviewed by AI first — a confident read resolves your obligation immediately, either way — and only escalates to a human at /admin/review when the AI isn't sure. It's still the fallback rather than the default, because it can't establish whose account cast the vote the way the Product Hunt connection can.

Once their window to back you closes without verification, it's recorded as a no-show and it costs them — trust events are the only thing that moves the score, and a no-show is a real, negative one. You did nothing wrong and owe nothing further.

From your actual record of kept and missed commitments, weighted toward recent behaviour and decayed over time — never set by hand, never adjustable by anyone, including us. One kept commitment scores 81, five in a row score 91, and a single no-show drops a strong record to about 72.

No. If an obligation can't be verified through no fault of yours — your partner never added their Product Hunt link, for instance — it's marked waived, which produces no trust event at all. The score only ever reflects things genuinely within your control.

Score tiers, not extra features. A score only appears after 3 resolved commitments — before that, "New" is more honest than a number with no evidence behind it. 70+ is "Trusted" (shows up reliably), 85+ is "Proven" (a real streak) — thresholds chosen against the actual scoring curve, not round numbers.

Your Product Hunt access token, stored separately from your public profile and never exposed through the app's normal access paths, including to you. A few display fields — username, join date, rough vote count — are mirrored to your profile so matching and the anti-farming check can use them.

There's no self-serve deletion yet — email hello@launchclan.com and we'll handle a deletion or export request directly.

Yes, from your profile settings — pact activity, launch reminders, and digests are separate toggles. Reminders with a real deadline, like an obligation window closing, still send at any hour rather than let it close unanswered; anything without a deadline attached waits for your own morning in your own timezone.