—Proof
PreArrive should earn trust with artifacts you can inspect, metrics that show sample size and date range, and customer stories only when the host has approved them. This page is the publication standard.
Last reviewed 2026-05-22. Minimum public metrics cohort: n=30.
01Inspectable today
These are public surfaces a buyer, partner, adjuster, or clerk can inspect now. They prove product behavior and evidence structure; they do not say anything about a guaranteed dispute outcome.
A clearly marked sample showing the packet text, per-item acknowledgments, drawn signature, audit trail, content hash, and QR verification path.
Generated from the same certificate renderer used by signed reservations.
A no-login verify page recomputes the certificate HMAC and returns a public verdict without exposing guest PII.
Backed by functions/api/verify/[hash].js and the PDF QR token.
The public REST API describes reservation, certificate, and webhook surfaces that partners can integrate against.
docs/api/openapi.yaml mirrored into public/developers/openapi.yaml at build time.
02Metrics
No public completion rate, reminder lift, claim export count, or rule category ranking ships without a cohort size, date range, exclusion notes, and PII stripping. A number without that context is marketing decoration, not proof.
| Metric | Source | Aggregation | Minimum | Status |
|---|---|---|---|---|
| Median time from send to signature | reservation sent_at and signed_at timestamps | median duration, excluding test/demo reservations and rows with missing timestamps | n=30 | Not published - Waiting for a permissioned cohort large enough to publish without implying a guarantee. |
| Signature completion rate | reservation lifecycle events | signed reservations divided by sent reservations for the stated date range | n=30 | Not published - Will publish only with sample size, date range, and exclusion notes. |
| Reminder conversion rate | reminder and signature lifecycle events | signatures after reminder divided by reminder recipients | n=30 | Not published - Reminder cohorts must be separated from manual follow-up before reporting. |
| Packets signed before check-in | reservation signed_at and check_in fields | signed before check-in divided by signed reservations | n=30 | Not published - Will publish only as aggregate operating data, not as a host outcome promise. |
| Most acknowledged rule categories | packet rule labels/categories after PII stripping | top categories by packet count, with free-text rules bucketed manually before publication | n=30 | Not published - Free-text rules need category review so no property-specific detail leaks. |
| Most included fee categories | packet fee labels/categories after amount normalization | top categories by packet count, with exact property wording excluded | n=30 | Not published - Fee categories must be normalized and checked against platform-policy guidance first. |
| Claim Kit exports | claim kit export events | count of exports, deduplicated per reservation and incident | n=30 | Not published - Claim Kit is new; export counts need a meaningful reporting window. |
03Stories
Customer evidence is useful only when it is permissioned and precise. A future case study must separate the rule disclosed, the item acknowledged, the evidence attached, and the outcome. A future quote must be approved by the host and caveated if it mentions money, disputes, or claim results.
04Limit
A signed acknowledgment can make a file clearer: what the guest saw, what they tapped, when they signed, and whether the certificate still verifies. It does not decide whether Airbnb, an insurer, or a court agrees with the host. Public proof should make that distinction louder, not softer.
Open the sample certificate, verify the chain, and start with one property free.