Paper's quiet problem
When an in-person notarization is challenged years later, the evidence is a stamp on the page and a line in the notary's journal — if the journal can be found. Everything else is memory. That's the gap remote online notarization was designed to close: the statutes that authorize RON also mandate that the session generate and retain evidence.
Understanding that evidence is worth ten minutes, because it changes what you can prove when a signature is questioned — by an heir, a counterparty, a court, or a fraud investigator.
The four pieces a RON session produces
Identity verification results
Before the notary appears: a records-based quiz the signer must pass, and credential analysis that inspects the security features of the government ID. The results — pass/fail, method, timestamps — are preserved with the session record.
The recorded session
The audio-video meeting itself is recorded and retained for a period set by state law, commonly years. The recording shows the signer, the notary, the oath if one was administered, and the act of signing.
The audit trail
Every step — document uploaded, identity verified, session joined, signature applied, seal affixed — logged with timestamps as it happened. On our platform the trail is hash-chained, so the log itself is tamper-evident.
The sealed document
The notary's electronic seal binds a digital certificate to the exact bytes of the finished document. Change anything afterward — a word, a date, a pixel — and verification fails visibly.
What "sealed" means here — precisely
On SignSealShip, "sealed" is a specific claim: the finished document's SHA-256 fingerprint is recorded in our custody system at the moment of sealing, and the artifact can be re-checked against that fingerprint at any time on our verification page. A document is only described as sealed when that verifiable artifact actually exists — it's a claim about cryptographic custody, not a synonym for "done."
The Certificate of Completion that accompanies every order gathers the human-readable version: who signed, how identity was verified, which notary acted under which commission, and when each step happened.
Observed facts vs. inferred ones
Most document-intelligence tools read a finished PDF and infer what probably happened to it — estimates, with confidence scores. A platform that actually ran the signing, the notarization, and the delivery holds something categorically different: first-party records of what did happen, written at the moment it happened, tied to the sealed original by hash.
That's the standard we hold our own records to: a fact is either observed — we performed it and logged it as it ran, no confidence score, because nothing was inferred — or it's clearly labeled as read from a document after the fact. When you hand the package to a skeptical bank, court clerk, or opposing counsel, that distinction is the difference between "here's what our system believes" and "here's what our system did."
What evidence is not
Evidence proves what happened in the session — it doesn't make the underlying document legally sufficient, doesn't guarantee any office will accept it, and doesn't constitute legal advice about what the document means. Recorders, courts, and institutions apply their own acceptance rules, and communications with our platform and its notaries aren't privileged the way attorney communications are.
What it does do is make the factual questions cheap to answer: who signed, when, how they were identified, and whether the document changed since. In most disputes, those are exactly the questions that matter first.