Trust & data handling

Where your documents live — and what happens to them.

Bookkeeping doesn't need marketing promises, it needs verifiable commitments. This page explains where NameQuick Invoices stores your data, how original documents are protected, what gets logged, and which service providers are involved.

Hosting & data location

Your invoice data is stored in the EU. Concretely:

Application: Vercel, EU region

The web application is served through Vercel. Server functions are pinned to the EU region in Paris (cdg1).

Data: Google Cloud europe-west1 (Belgium)

The database and document storage run on Google Cloud in the europe-west1 region (Belgium). AI-assisted field extraction uses Google Gemini (see service providers).

Strict workspace isolation

Originals, derived data, review state, and events stay attached to the workspace that received the invoice.

Immutable original documents

The document you upload or forward by email is the source record — and stays that way:

  • The original is stored unmodified and never overwritten. Corrections are created as derived records with their own event entry, never by editing the original.
  • Structured e-invoice components (such as the XML inside an XRechnung or ZUGFeRD document) are preserved as well.
  • You can download the original directly from the invoice record at any time.

Derived artifacts — OCR text, extracted PDF text, parser output, thumbnails — are stored but treated as regenerable. The canonical records are always the original and your reviewed values.

Audit trail: every step is traceable

Every relevant action on a document is recorded as an event and surfaced in the invoice history:

Intake

Every document is logged on receipt — via upload or email forwarding, including acceptance or rejection.

Processing

The start and outcome of every processing run, including extraction and normalization of invoice fields.

Field corrections

Every manual correction to an extracted field is recorded.

Duplicate decisions

Detected duplicate candidates and the decision you make about them.

Approval

The moment an invoice is marked ready for export.

Export

Every export batch and every invoice exported in it.

Reprocessing

Every requested reprocessing of a document, including support-initiated retries.

Every event carries the workspace, a timestamp, and the actor. For extracted values, the data provenance is recorded as well: structured XML, digital PDF text, OCR, or an AI suggestion.

Retention & GoBD

We never delete your documents or invoice data on our own — there is no silent deletion and no hidden expiry. Only you can delete invoices, and every deletion is logged in the audit trail. Your data is fully removed only when your workspace is deleted.

A GoBD-supporting workflow

NameQuick Invoices provides the technical building blocks that GoBD-compliant processes are built on:

  • Immutable originals: the source document is preserved untouched.
  • A complete event history: intake, edits, approval, and export are traceable.
  • Complete, open exports: CSV and DATEV EXTF, at any time — your data belongs to you.

Honesty is part of trust: NameQuick Invoices is not a certified archiving solution. We do not advertise certified audit-proof (revisionssicher) archiving and we make no tax or legal compliance guarantee. Responsibility for GoBD-compliant processes remains with the taxpayer — we make the technical side as solid as possible.

Service providers (subprocessors)

We rely on a small, deliberately chosen list of service providers. Details on data processing are in the privacy policy (in German).

  • Google CloudDatabase, document storage, and document processing — database and storage in the europe-west1 region (Belgium).
  • VercelHosting and delivery of the web application — server functions in the EU region in Paris (cdg1).
  • Google GeminiAI-assisted field extraction for scans and edge cases — no training on your data in the default configuration.
  • ResendTransactional email and email invoice intake.
  • PolarPayment processing and subscriptions — payment details are handled directly by Polar and never reach our systems.
  • SentryApplication error monitoring (when enabled).
  • OpenPanelProduct and web analytics.

No AI training on your documents

Your documents are not used to train AI models by default. AI is used exclusively to extract invoice fields from your documents.

Structured e-invoices (XRechnung, ZUGFeRD) are read directly from the XML — no AI involved. AI extraction handles scans and ambiguous cases, and it never overrides XML values or fields you have already reviewed.

See for yourself

Process your first invoice free and see the original, the review view, and the event history in action.