What Is Digital Archiving? Definition, Methods and How to Set One Up

Digital archiving is the practice of keeping selected digital material findable, readable and accountable for as long as an organization needs it, whether that material was created digitally or digitized from physical originals. If someone has asked where your organization’s digital material is kept, and the honest answer is that it sits across a shared drive, a handful of mailboxes and somebody’s local drive, that gap is what the practice exists to close.

The term causes unusual confusion, because three different audiences use it for three different things: organizing personal photographs, European compliance regimes with their own legal definitions, and software that captures corporate messaging for regulators. All three are legitimate. None answers the question a records manager, archivist, librarian or information manager is asking when they inherit material the organization cannot afford to lose.
This article covers what digital archiving means for an organization, how it differs from a backup, how organizations do it in practice, and the six decisions that come first.

What digital archiving actually means

A folder of files is not an archive, and the difference has nothing to do with size or storage medium. Three properties separate them, and any storage arrangement that has all three is doing archival work. An unmanaged shared drive usually has none.
It is selected: Somebody decided this material is worth keeping, and by implication decided that other material is not. An archive that holds everything is a shared drive with a better name, because without selection there is no basis for prioritizing description, storage or access work.
It is described: Metadata is attached to the material so that a person who had nothing to do with creating it can work out what it is, where it came from and whether it is the thing they need. Without description, retrieval depends on somebody remembering, which is a dependency with a resignation date attached.
It is kept accessible and accountable: The file can still be opened years from now, and there is a documented basis for saying what has happened to it since. Integrity in an archive is a fixity check plus the record of what was done and by whom.
That last point is where digital archiving diverges from ordinary file management, and it is where an archive in the professional sense begins.

What a digital archive holds

Digital archives hold digitized material or items created digitally with no physical origin, such as email, case files, CAD drawings, committee papers and captured web pages.
The archiving questions are the same for both: what is kept, how it is described, how long it lives. The preparation is not, and the digitizing of archival materials carries its own capture standards and costs. Organizations whose holdings are mainly documents often call this digital document archiving.

Backup, archive, record and digitally preserved object: how they differ

The most common mistake here is assuming that because material is backed up, it is archived. Four things get used interchangeably, and each is designed to do something the others are not.

A backup exists so you can return to a recent working state after something breaks. It is built for speed of recovery, it is normally overwritten on a cycle, and it is not designed to be searched or understood by a person years later.
An archive exists so that selected material stays findable and readable, sometimes for a defined period. It is built for retrieval by today’s users and by someone who was not there when the material was created.
A record is information created or received in the course of an organization’s activities and kept as evidence of them. The Society of American Archivists defines a record as information used as an extension of human memory or to support accountability. A schedule, regulation or legal obligation governs how long a record is kept, not whether it is a record. Records management handles appraisal, retention and disposition, and is a related but separate field from archives management.
A digitally preserved item is archived material under active management against format and platform change, with integrity checking and planned migration behind it. Digital preservation is a layer applied to an archive rather than another word for it, and it carries its own standards and costs. Its reference model is the Open Archival Information System, published as ISO 14721:2025, which replaced the 2012 edition in March 2025. Building a digital preservation strategy is a project in its own right.
These four are not mutually exclusive. One committee paper can be a record because of how it was created, sit in an archive because it was selected for long-term keeping, have backup copies, and be under digital preservation management because its format is at risk. The question is rarely which one a file is, but which of the four jobs are being done for it, and by whom.
A useful test: could you produce a specific document from 2017, show what has happened to it since, and say who was authorized to see it? Backup and storage are not designed to answer that; whether your arrangements do depends on what is built on top of them.
The National Archives sets three rules for organizations beginning this work: only ever work on copies, document every process, and verify that records remain unaltered using checksum software. Their guidance on getting started is worth reading in full. They exist because the alternative is discovering a problem at the moment somebody needs the file.

Description and access sit in one layer, long-term preservation in another. Soutron’s digital archive management software handles the first, supporting ISAD(G), ISAAR and Dublin Core, while digital preservation is an optional add-on delivered through MINISIS Trusted Digital Repository.

How organizations actually archive digital material

There is no single method. The useful question is what each approach leaves you responsible for once it is in place.

These approaches are not mutually exclusive, and running two or three alongside each other is a reasonable outcome. The failure is never deciding which material belongs where, because the default is that everything stays wherever it first landed.

What a digital archiving system is, and is not

A digital archiving system usually means the third row: a system that catalogs material, controls who can see it and makes it retrievable, which is what an archive management system does. The phrase gets applied loosely to all five approaches, which is a significant part of where software evaluations go wrong: an organization buys storage when what it needed was a catalog. If you are comparing products, digital archiving software sets out the categories, and digital preservation software deals with the preservation question.
Document management is a different category again, and the two get shortlisted against each other. A document management system is built for material in active use, with check-out, version control and collaborative editing. An archive is organized around stable identifiers, description and controlled access, which suits material no longer being revised.
Security is not a sixth method. Secure digital archiving is a property of whichever approach you choose, delivered through access rules, encryption and audit logging, and it is far easier to configure before material is loaded than afterwards.

How to set up a digital archive

Creating a digital archive is a sequence of six decisions, and all of them come before any system does. They are set out in a sensible order rather than a mandatory one: sequencing varies with the collection and the organization, and several get revisited as the work proceeds.

1. Decide what the archive is for: Evidence of decisions, a research collection, public access, a legal obligation, or some combination. The purpose determines what gets selected, how it is described and who is allowed near it. Without a stated purpose there is no defensible basis for saying no to anything, which is how archives end up holding everything or conversely the items never get archives at all.
2. Survey what you already hold: Formats, approximate volumes, where the material sits and who created it. The National Archives puts this first for good reason: you cannot plan storage, staffing or migration for volumes nobody has measured. A survey can also surface material that hasn’t been touched in years and material on removable media, usually the most immediately at risk.
3. Agree retention before you agree storage: Which material is kept permanently, which has a defined period, and which is not archive material at all. This is a governance decision, it needs a named owner with the authority to make it, and it has to be documented.
4. Decide how material will be described: Set the minimum metadata, who creates it and when. A workable minimum for most collections is a unique identifier such as project name, a title, the creator or source, the date, the access condition and the retention decision.
5. Set access rules before anything is loaded: Who can see what, on what basis, and how exceptions are handled. Retrofitting access control onto material already in the system is substantially more work than configuring it at the start, and in collections holding personal or commercially sensitive material it can stall a launch.
6. Build in integrity checking and a migration plan: You can generate checksums at ingest with a digital preservation system and verify them on a schedule, in addition to documenting what should happen when a format or platform stops being supported. Over a long retention period the formats and platforms you start with are unlikely to still be current, and planning for that is cheaper than reacting to it.

One further point decides whether this lasts: somebody has to own the archive after the project that created it has closed. Handover is a point of real exposure: the funded work ends and the project team disperses, and unless the archive appears in somebody’s role description, the review cycles quietly stop.

Putting these decisions into practice

Knowing what digital archiving involves is the groundwork. Choosing a system to run it on raises a different set of questions, about standards support, migration, and what happens to your material in ten years. Our white paper on the challenges of choosing an archive management system sets out the criteria worth testing before you shortlist.

Digital archiving best practices, and the problems they prevent

Five risks come up repeatedly in digital archiving projects, some of them governance and some technical. Each has a practice that addresses it, and the practices are more useful stated alongside the problem they solve.
Nobody owns it, so nothing gets decided: The archive sits between IT, legal and the records or library function, and each assumes another holds it. The practice: name one owner with authority over retention decisions before anything else is agreed, and give IT the infrastructure beneath it.
Retention was never agreed, so nothing can be deleted: Volumes and costs grow without limit and the archive becomes hard to distinguish from the shared drive it replaced. The practice: write the retention rules down and review them on a defined cycle, because regulations change and business purposes lapse.
Description was deferred: Material was loaded with metadata promised later, and describing it retrospectively is slower, because somebody has to reconstruct context that was obvious at the time. The practice: describe at ingest, using a recognized standard. ISAD(G), ISAAR and Dublin Core exist so that description is exchangeable rather than local to one organization.
Integrity was never checked: Files were copied forward across platform moves with nothing verifying they were intact, and corruption surfaces at the moment somebody needs the file. The digital preservation practice: generate checksums at ingest and re-check them on a schedule. After a deliberate change such as format migration, the resulting file is a different file with a different checksum, so a new baseline is established and the action is recorded. Fixity detects change; the documentation explains it.
Formats were chosen by habit, and change was never planned for: The practice: choose formats on evidence. The Library of Congress Recommended Formats Statement, in its 2025 to 2026 edition, sets out format hierarchies by content type. It was written to guide the Library’s own acquisition of published content, so treat it as a reference point rather than a ready-made policy. PDF/A is the usual archival choice for text documents. Then plan for technological change rather than assuming the current platform and formats will still be supported when the retention period ends.
Loss at scale is documented rather than theoretical: the Digital Preservation Coalition’s Global Bit List, 2025 edition, tracks categories of digital content at active risk across cultural heritage, government, legal, research and commercial settings.
One further practice: set a maturity target you can reach. The NDSA Levels of Digital Preservation, at version 2.1 since March 2026, map five functional areas (Storage, Integrity, Control, Metadata and Content) across four levels. The framework states explicitly that Level 4 need not be the goal in every functional area, which matters for teams working within real constraints
If you want a quick read on where your own holdings stand, our digital preservation readiness checklist scores itself as you work through it.

Frequently asked questions

What is the 3-2-1 rule, and is it enough for an archive?

The 3-2-1 rule says keep three copies of your data, on two types of media, with one held off-site. It is sound advice, and it is a backup rule. It protects against losing copies, which is one of several ways digital material becomes unavailable. It does nothing about finding material, knowing what a file is, controlling who sees it, or keeping a file readable once its format stops being supported. An archive needs 3-2-1 and the four things it does not cover.

Can you give me an example of a digital archive?

A corporate archive might hold board papers, product photography and historic correspondence, cataloged under a retention schedule with access restricted by department. A university special collection might hold digitized manuscripts and born-digital research papers, described to archival standards and published openly. Both qualify because the material is selected, described and retrievable. A departmental folder holding the same files, with none of those three, would not.

Is digital archiving the same as cloud storage?

No. Cloud storage is where files sit. Digital archiving is the set of decisions about which files are kept, how they are described, how long they stay and how they are accessed. Cloud storage works well as the storage layer beneath an archive, and many archives use it that way. What it cannot do on its own is tell you what a file is, how it is related to other items in the archive, or whether you are still allowed to hold it.

How long should digital archives be kept?

There is no universal period. Retention is set by the schedule, regulation or legal obligation that applies to the material, and periods range from a few years to permanent. Records carry either: some are disposed of once their period ends, and some are designated for permanent retention and become part of the archive. The question for any series is what its retention period is and who set it. An organization that cannot answer has a governance gap rather than a storage problem.

Who should own digital archiving in an organization?

It needs a single named owner with real authority over retention decisions, usually in the records, library or information function, with IT owning the infrastructure beneath it. Two risks are worth guarding against. Where the archive is treated as purely an IT project, retention and description decisions can go unmade, because nobody with authority to make them is in the room. Where it sits entirely with the records function and IT is not engaged, the decisions may be sound while the infrastructure goes unfunded.

Once material is digitized and archived, can we delete the originals?

It depends on whether the digitized version meets the applicable quality and validation requirements, and whether a retention schedule authorizes disposal of the source. US federal agencies have a documented example: under 36 CFR Part 1236 Subpart E, once a digitization project is validated, the digitized version becomes the permanent record and the source becomes a temporary record, disposable under General Records Schedule 4.5 or an agency schedule. A designated official confirms compliance first. Organizations outside the federal system should check their own retention schedule and legal counsel. Validation and authorization both have to happen before anything is destroyed.

Where this leaves you

The difficult part of digital archiving is not the technology. It is deciding what the archive is for, writing down how long each kind of material stays, and naming the person who owns all of it once the project that built it has closed.
An organization that has settled those three questions can implement an archive in almost any of the approaches above and get a usable result. One that has not will struggle in the best system available, because the system will faithfully enforce decisions that were never made.
If your holdings are growing faster than anyone’s ability to describe them, that is the usual signal that the governance work is overdue rather than the storage.