Tamperlens and Snappt

How to read this page

Side by side

Where the two products meet

Their own training-corpus size, on two of their own pages

Two figures for the same thing are published at once, and a reader comparing products should know which one they are being quoted:

• 16+ million: "trained on 16+ million documents" on their document fraud detection page (read 2026-08-13).
• over 19 million: "trained on over 19 million documents" on their "Why Snappt" page (read 2026-08-13).

We draw no conclusion from the gap: corpora grow, and pages go stale at different speeds. We note it because a corpus size is the kind of figure a comparison gets quoted on, and because we would want the same noted about us. For what it is worth, the size of a training set says how much a model has seen, not how often its answer is right.

Where Snappt is the better answer

• You run a multifamily portfolio. If your applicants already flow through Entrata, RealPage, Inhabit, ResMan or Yardi, Snappt plugs into the workflow you have. We do not.
• You want a human ruling, not a signal report. Their "Fraud Analyst Review" layer puts a person behind the result. Nobody reviews a Tamperlens report but you.
• You need more than the document. Income, identity and rental-history verification are the platform; the document check is one part of it. We only ever look at the file.
• Your procurement asks for SOC 2. They hold a SOC 2 Type II attestation and we hold no certification at all.
• Scale is the reassurance you need. They state "over 2.5 million units" protected (their home page, read 2026-08-13). We are a solo-operated product and say so.

Where Tamperlens is the better answer

• You are not a multifamily operator. Snappt's platform, integrations and logos all point at one vertical. A small landlord, an imobiliária, a lender or a marketplace that wants to check one suspicious PDF today has no published path there, no price, no signup, no checker. That buyer is who this page is for.
• You want a price you can read. Every plan and the per-document arithmetic are on /pricing, in USD and in reais.
• You want one answer now. The free checker takes a file with no signup; an API key comes with 50 free documents a month. No call.
• You need the check inside CI. The engine is deterministic, the same bytes produce the same report, so a result can be regression-tested like any other build artifact.
• You want the error rate before you buy. Our false-positive rate is published at /evidence with the population and date it was measured on.
• The file must not linger. Documents are uploaded directly, parsed in memory and never written to disk.
• Your documents are not applicant documents. Any PDF, image or Office file goes through the same endpoint, including redaction recovery by paint order and document-borne prompt-injection signals.
• Brazil. CPF, CNPJ and boleto identifiers checked by arithmetic on the check digits, with prices quoted in BRL.

What we cannot do

• No applicant workflow. There is no screening dashboard for a leasing team and no concept of an applicant: only documents.
• No property-management integrations. Nothing for Yardi, RealPage, Entrata or any PMS.
• No income, identity or rental-history verification. We never verify the person, only the file.
• No human review service.
• No SOC 2 or ISO 27001, disclosed at /security.
• Recall is unmeasured. We know how often we flag genuine documents; we do not know how often we catch real tampering.
• One region. Hosting is EU-only, and this is a solo-operated product.

The number a buyer actually needs

A headline accuracy with no definition behind it does not tell a leasing team the thing that sets their workload: how often a genuine applicant's documents get flagged. That number is the false-positive rate, and Snappt does not publish one (read 2026-08-13). Neither does almost anyone in this market.

Ours is at /evidence, with the population it was measured on, the date, and the engine version that produced it. The other comparisons are at /alternatives.