ecoDMS vs PaperOffice

ecoDMS

Intake is genuinely documented: scanners via TWAIN/WIA/SANE and scan-to-folder batch import, separator sheets that auto-split stacks, a smartphone ScanApp, and email among the input types, with automatic OCR full-text indexing and template-based classification that reads invoice data from layout or keyword. But we found no public information on a validation step for the automatic extraction, on XRechnung or ZUGFeRD being read as structured data, on duplicate detection, or on where the ScanApp's AI cropping actually runs.

PaperOffice

Classification determines document type, sender and direction and proposes retention rules, extraction carries confidence values with uncertain fields routed to a review list and documents above a self-set level going to quarantine, and duplicate detection is documented — this is AI capture with a validation step, which is what I look for. E-invoice tools create and validate XRechnung and ZUGFeRD against the EN 16931 profile, and intake runs via scanner SFTP, watched folders, IMAP mailbox and REST API with stated limits (5–15 minute intervals, unread only, OAuth for Microsoft 365 not yet supported). The model provider behind the AI is not named and trainability per document type extends only to a field-suggesting coach and custom terms, which keeps it below the top level.

ecoDMS

The pages promise a traceable change history, versioned and immutable PDF/A storage, clearly controllable access and team access rights — but only at claim level, repeated across pages without mechanics. We found no public information on restoring earlier versions, check-in and check-out, role or directory-based rights, single sign-on, or a tamper-evident, exportable audit trail covering reads and changes.

PaperOffice

Every revision is kept as a version with its own SHA-256 checksum, users and groups hold read, write or admin rights per workspace, and the audit log records accesses, changes and level changes with timestamp, person and old and new field values, with no function in app or API to alter log entries. We found no public information on check-in and check-out, single sign-on or directory sync (Entra ID, LDAP), on whether exports are logged, on field-level rights, or on the log retention period and whether the trail itself can be exported.

ecoDMS

"Revisionssicher nach GoBD" with scanned documents stored immutable, versioned and in PDF/A is stated, and fulfilling retention obligations GoBD- and GDPR-conformingly is claimed. But we found no public information on how retention periods are configured, on a sample Verfahrensdokumentation, on deletion after expiry or legal hold, or on any auditor's report or Testat — the word is on every page, the independent evidence behind it is not shown.

PaperOffice

WORM protection per workspace that holds until the retention term ends, retention periods and legal basis per document type with a user-chosen start date, deletion only by hand with legal hold, retention-end events logged, GDPR access-request handling, and an audit export with a read-only auditor account are all documented. What the captured pages do not show is an auditor's report (IDW PS 880 or comparable) with scope and date, a sample Verfahrensdokumentation, or a Z1–Z3 tax-audit export, and the vendor itself leaves the final GoBD assessment to the tax advisor — a Testat would change this score materially.

ecoDMS

Acknowledgement and approval workflows exist via the optional WORKZ add-on, and a JSON-REST API plus named connections to Lexware, DATEV Unternehmen Online, SelectLine and Sage 100 are published — though the four accounting connections appear in a webinar announcement without any stated scope. We found no public information on graphical workflow modelling, substitutes and escalation, tasks with deadlines, webhooks or e-signature.

PaperOffice

Approval and follow-up steps with owners and deadlines, two-person or SMS approval, e-signature, connectors over SFTP, S3, WebDAV and IMAP with intervals and post-import rules, and a REST API whose endpoints and limits are published are all documented; the captured pages give different figures for webhooks, with one listing them as upcoming while others describe signed events with retries. We found no public information on graphical workflow modelling without code, on connectors to named ERP systems such as SAP, Microsoft Dynamics or DATEV, or on a DATEV handover from an invoice workflow.

ecoDMS

Full-text search is well evidenced: automatic OCR indexing behind every document, "intelligente Volltextsuche", and access via web browser, desktop and the ScanApp's web client search. We found no public information on metadata filters or saved searches, on bulk export of documents with their metadata and versions, or on any documented exit path or import from other systems.

PaperOffice

Full-text, semantic and hybrid search over embeddings at summary, page and paragraph level, filters, saved searches, permission-scoped results and a mobile app are documented, and metadata exports go out as Excel, CSV, JSON or PDF optionally with originals and OCR text, with an import path via SFTP, S3, WebDAV and ZIP named. The captured pages do not show the audit trail included in an export, a documented structure for exporting version chains, or the post-termination retrieval window and deletion terms, so the exit is well begun but not designed in.

ecoDMS

There is real, evidenced European substance: a German contracting entity (ecoDMS GmbH, Aachen, register court Aachen) under exclusively German law, a product installed and run on the customer's own systems on Windows, Linux, macOS or NAS, and an AV-Vertrag offered. But the captured pages say nothing on ownership beyond the holding's name, on where any hosted service or the ScanApp's AI processing runs, or on subprocessors beyond PayPal (Europe) in Luxembourg for payment processing, and we found no public information on certifications such as ISO 27001 or BSI C5.

PaperOffice

The worldwide contracting party is a Spanish entity in Pamplona, processing and storage run in the vendor's own EU data center with own GPU hardware and RAM-disk deletion after completion, index and embeddings stay in that EU data center, ISO 27001 certification is published, and files can remain on your own NAS — though index and metadata then still sit in the data center. We found no public information on ownership, a published DPA, a subprocessor list, or the model provider behind the AI, and a US office in San Jose handles sales and support; I treat an own-data-center claim without a subprocessor list as good but unproven.

ecoDMS

The main products carry public gross prices — ecoDMS "ab 89,00 EUR inkl. 19% MwSt." and ecoMAILZ "69,00 EUR inkl. 19% MwSt." — under a flat, time-unlimited licence with 24 or 48 months of updates included and the scanning features stated as included at no extra cost. The captured facts name no price for the WORKZ add-on, the ecoDMS ONE subscription or additional concurrent connections, and we found no public information on onboarding fees or price-increase terms.

PaperOffice

Three OCR modes are priced to the cent per page, the billing model is stated word for word as no subscription and no base fee with the API billed in credits, and a free trial without credit card is offered. The named plans (Basic, Premium, Ultra, Corporate) carry no published prices on the captured pages, so a buyer cannot compute the cost of users, storage or modules, and we found no public information on storage limits, contract terms or notice periods.