Why document management is suddenly on the agenda
Every few years, an event reminds the market why the ownership model behind a piece of software matters as much as its feature list. This year it is document management. PowerRetrieve, a capable and well-regarded document capture and retrieval product used across investigations, professional services and archiving, has been caught up in the administration of its parent company, 360Globalnet, which entered administration in March 2026. The group's claims-technology business subsequently found a buyer. For PowerRetrieve the public record points the other way: Infocap Technologies Ltd, the company behind the product, applied in July 2026 to be struck off the companies register, and at the time of writing no buyer or continuation of the product has been announced. Its customers, some with decades of scanned records in the system, are now asking how long their service will stay online and where their archives go next.
We know this because they are asking us. Enquiries about Paperless-ngx have climbed steadily since the spring, and the majority trace back to the same story: a proprietary document system, a vendor that no longer exists, and an archive that was always the customer's data but never quite in the customer's hands.
A document archive routinely outlives the company that sold the software it sits in. That is not a criticism of any vendor; it is an argument about structure. When the system holding thirty years of client records is open source, running on infrastructure you control, no administration event, acquisition or product sunset can put the archive at risk.
What Paperless-ngx actually does
Paperless-ngx is the community-maintained successor to the original Paperless project, and it has become one of the most widely adopted document management systems in the world, with over 44,000 stars on GitHub and an active contributor base that ships regular releases. It does the unglamorous work of document management properly.
Documents arrive from wherever they already live: monitored email inboxes, watched network folders, scan-to-folder devices, drag-and-drop upload, or a photograph taken on a phone. Every document is processed with Tesseract OCR, which supports more than 100 languages, and stored alongside a searchable PDF containing the original image plus an invisible text layer. From that point the entire archive is findable by content: any phrase, reference number or name that appears anywhere in any document, surfaced in seconds, with fuzzy matching that copes with the OCR errors that poor-quality scans inevitably produce.
A machine learning classifier watches how you tag and file, then starts doing it for you: correspondents detected from content, document types distinguished automatically, tags applied by rule. Custom fields extend the metadata model to whatever your business tracks, whether that is a matter reference, an invoice value or a year end. Document-level permissions, audit logging and automated retention policies cover the compliance side. And everything is reachable through a full REST API.
The feature list is the entry ticket. What makes Paperless-ngx the right foundation for a professional practice is that all of it, the OCR, the classifier, the API, the storage format, is open, inspectable and yours. The archive is ordinary files plus open metadata. You can leave any provider, including us, and keep everything.
How we apply it: built into the practice, not beside it
A stock document management install solves the storage problem and leaves the workflow problem untouched. Most of our work with Paperless-ngx happens above the platform: we have written wrappers over its API for accountancy and law practices that map the system onto the way each firm actually works, so the archive stops being a destination someone has to remember to visit and becomes part of the working day.
Accountancy practices
Client records map to correspondents automatically, so everything relating to a client, engagement letters, VAT correspondence, HMRC notices, supplier invoices arriving in the practice mailbox, files itself against the right entity without anyone touching it. Working papers are organised by client and year end through custom fields, retention schedules enforce the periods your AML supervisor expects, and when a subject access request lands, full-text search across every document relating to a data subject turns a day of hunting into minutes.
Law firms
Filing is matter-centric: our wrapper keys every document to a matter reference, so case papers, correspondence, exhibits and attendance notes accumulate against the matter rather than in a folder tree that decays over time. Disclosure exercises start from a search query instead of a paralegal's memory, document-level permissions keep ethical walls enforceable, and the audit log answers who accessed which document and when, in the form a regulator or insurer expects to see it.
Beyond the wrappers, we build custom integrations against the same API to connect Paperless-ngx to the systems a firm already runs: practice management, case management, billing and client portals. Keycloak provides single sign-on with your existing credentials and MFA policy, and for firms running Nextcloud the two are complementary, familiar file browsing on one side and content search on the other.
The newest layer is AI. Paperless-ngx now ships optional AI features, metadata suggestions and plain-English questions answered from your own documents, and they work with any OpenAI-compatible endpoint. We connect them to AI models hosted on UK infrastructure rather than a US provider's cloud, which means a fee earner can ask the archive a question without a single client document leaving the country.
The PowerRetrieve migration wave
For PowerRetrieve customers the practical question is not whether to move but how, and how fast, given the vendor's wind-down. The mechanics are straightforward to plan, because the source material is not exotic: documents come out as the PDFs and TIFFs they always were, are bulk-ingested through Paperless-ngx's consume pipeline, re-processed with OCR, and mapped onto correspondents, types, tags and custom fields that mirror the structure the firm had before. The search experience that made PowerRetrieve valuable, internet-style queries over imperfectly scanned paper, is the thing Paperless-ngx does natively, so the capability the archive was bought for survives the move.
Customers on PowerRetrieve's hosted service should treat extraction as urgent rather than eventual: when a wound-down company's hosting stops, it stops without a notice period. Every archive is structured differently, so we start by understanding your use case, your export options, your metadata and your volumes, and quote the migration on that basis; if the first step needs to be a hurried export while the service is still online, we help with that first.
Friendly to the open source world
We run Paperless-ngx as the community builds it. No fork, no proprietary patches that drift from upstream, no repackaging the project under our own name. Our wrappers and integrations sit above the API precisely so that the platform underneath stays stock, upgradeable and honest to the project, and where our work surfaces something worth fixing upstream, it goes upstream. Firms pay us for deployment, operations, integration and accountability, never for the software, because it was never ours to sell. There are no per-user licences, and there is no lock-in we could enforce even if we wanted to: the archive is open files and open metadata on infrastructure you control.
High availability, at any size, fully managed
The same platform serves a three-partner practice and a firm with millions of documents. We deploy Paperless-ngx on our UK Kubernetes estate or into your own cloud, with replicated services and load-balanced ingress where availability targets demand it, encrypted volumes, encrypted daily backups with tested restores, monitoring, and upgrades tracked against upstream releases. Billing is hourly against our rate card with no per-user fees and no commitments, and the deployment scales with document volume rather than headcount.
Document management is a decades-long commitment, and this year has shown what happens when that commitment rests on a single vendor's solvency. Paperless-ngx, run fully managed, integrated into your practice systems, and connected to UK-hosted AI, is the version of that commitment that cannot be taken away from you.
Moving from PowerRetrieve, or just moving on from paper?
Talk to us about your document volumes, your practice systems and what a managed Paperless-ngx deployment would look like for your firm.