Very different things are sold under the label “electronic document archive”: anything from storage space with folders to archiving services governed by a legal regime of their own. Here we are talking about what a company needs day to day: a single place where invoices, contracts and personnel files can be found quickly, by those entitled to see them.
One thing we leave aside on purpose. How long each type of document is kept, and under what conditions the electronic version can take the place of paper, are matters you settle with your accountant and a lawyer. We build the archive so that it can follow the rules they give you.
Server folders, emails and lever arch files: the starting point
Almost every company already has an archive in electronic form; it is just scattered. Contracts sit in the inbox of whoever negotiated them, scanned invoices in a shared folder with a name like “scan final new”, payroll records on a computer in human resources, and the originals in lever arch files.
The trouble starts when a colleague leaves, during an inspection or in a dispute with a business partner, when a particular document and its signed version have to be found quickly. There is a quieter problem too: in the shared folder, anyone can open, move or delete anything, without leaving a trace.
Electronic document archive: what happens to a document from the moment it comes in
Documents come in by several routes: scanned, uploaded by hand, sent to a dedicated email address or picked up automatically from other software. On arrival, each one is given a record with identifying details, known as metadata: type, business partner, number, date, period, department.
For scans we use optical character recognition (OCR), the technology that turns the image of a page into text. The text goes into an index, so you can search for a word inside a contract, not just its title. OCR has its limits: handwriting and faded copies are read poorly, so important data is checked by a person.
- Classification: document types you define, each with its own mandatory fields.
- Search: by content, by any field, or both combined.
- Links: the invoice linked to the contract, the delivery note and the proof of payment.
- Versions: a document is never overwritten; each change keeps the previous version.
- Retention periods: a review date for each type, filled in according to the rules given by the specialists.
Who sees the payroll records and who opened the contract
Rights are granted by role, not person by person: accounting, sales, human resources, management. Each role sees certain types of document and can do certain things with them. Documents containing personal data call for stricter rules, which you set with your data protection officer.
The log records every action: who uploaded, who opened, who downloaded, who changed the rights. Its entries cannot be altered from within the application. On arrival we calculate a checksum for each file, a kind of fingerprint of its contents, which can be used later to check whether the file has changed. This is a technical check; how much weight it carries as evidence is something to discuss with a lawyer.
Where the files live and what happens if the server fails
The files are kept in object storage compatible with the S3 interface (a widespread way of storing large numbers of large files, offered by many hosting providers), and the records and the index in a database. We choose the physical location of the data together: a data centre in Romania, one elsewhere in the European Union, or a server on your premises.
Backups are made automatically, in a different place from the archive itself, and are encrypted. A backup that nobody has ever tried to restore from is only a hope, which is why a test restore is part of the maintenance routine.
The right order: the classification scheme first, then the scanner
The most common mistake is to start by scanning everything. The result is the same chaos, moved to a different disk. We start with the list of document types and the fields for each, then the access rights, and only after that do we upload.
If you already have an office suite with shared storage and your needs go no further than tidy folders and simple permissions, setting it up carefully may be enough. Custom development is worthwhile when the archive has to be connected to the company’s software or when the access rules are finer-grained than a general-purpose product allows.
- The inventory: what documents you have, where they are now, who uses them.
- The classification scheme: types, mandatory fields, naming rules.
- Day-to-day uploads: new documents go into the archive from the first day.
- The backlog: older documents are brought in gradually, starting with those searched for most often.
Frequently asked questions
Can I throw the paper away once I have scanned it?
That is not an answer for a software supplier to give. It depends on the type of document and on the conditions under which the law recognises the electronic version, which are things your accountant and a lawyer confirm.
How long does it take to bring older documents into the archive?
It depends on the volume, the condition of the paper and how much data you want filled in for each document. That is why we recommend bringing the backlog in stages, starting with what is searched for often.
What happens to the documents if I stop using the application?
The files stay in their original format, and the records can be exported in an open format that other software can read easily. We build the export from the start, not on request.
This is a typical project description: it shows how we usually approach this kind of work and does not present a project carried out for a particular client. Every real project starts from your company’s situation, and the stages, timescales and price are agreed after the initial discussion.