Archival Cataloguing Standards: Which One Should You Use?

If you have started researching how to catalogue a collection, you have probably encountered a confusing alphabet soup: ISAD(G), SPECTRUM, EAD, Dublin Core. Each has advocates, each has documentation running to dozens of pages, and none of them comes with a plain-language explanation of when you should actually use it.

This guide cuts through the noise. It explains what each of the main archival cataloguing standards does, who each one is designed for, and how to make a sensible decision — without spending weeks reading standards documentation.

The main UK archival cataloguing standards

There is no single international standard that every archive, museum, or heritage organisation must follow. Different communities — museums, local authority record offices, community archives, academic libraries — have developed their own frameworks, and they overlap in places while diverging sharply in others.

SPECTRUM is the UK’s collections management standard for museums, maintained by Collections Trust. It defines 21 collection management procedures — from object entry and loans through to cataloguing and conservation — and specifies what information should be recorded at each stage. SPECTRUM is not purely a cataloguing standard; it is a broader management framework. But its cataloguing procedure is the benchmark UK museum accreditation bodies assess against, and compliance is expected if you hold Accredited status or are working towards it.

ISAD(G) — the General International Standard for Archival Description — is the international standard for describing archival holdings. Published by the International Council on Archives, it defines seven areas of description and 26 elements. Its core principle is multilevel description: you catalogue the archive collection as a whole first (fonds level), then series, then files, then items. Information is recorded at the level where it is most relevant, rather than repeated at every level. This hierarchical model is the standard framework for professional archive collections held by county record offices, university archives, and similar institutions. For a detailed explanation of how ISAD(G) maps to EAD encoding, see our EAD metadata standard guide.

Dublin Core is a simpler, lighter standard developed for the web. Its 15 metadata elements — title, creator, date, description, subject, and so on — are broad enough to describe almost anything. That flexibility is both its strength and its limitation. Dublin Core is widely used for digital collections and online catalogues because it is easy to implement and works well with search engines. It does not support archival hierarchy or the nuanced provenance information that professional archive description requires.

EAD (Encoded Archival Description) is an XML standard for encoding finding aids in a machine-readable format. It is not a description standard in the same sense as ISAD(G) — it does not tell you what information to record at each level of description, only how to encode it electronically. EAD is used when you need to share finding aids with external systems: national aggregators, funder reporting systems, or other institutions. It is increasingly expected as an output format for digitisation projects funded by bodies such as the National Lottery Heritage Fund.

Which archival cataloguing standard should you use?

Your choice is often determined by external factors before you start: who funds you, what accreditation you hold, and what systems your catalogue must be compatible with.

If you are a museum holding Accreditation — or working towards it — SPECTRUM is not optional. Use it as your procedural backbone, applying Dublin Core or a local schema for your online catalogue where appropriate.

If you are a county record office, university archive, or professional archive service, ISAD(G) is the standard you should work to. Its hierarchical, multilevel description model is designed precisely for the provenance-based archive collections you manage. Most collection management systems used in UK archive services — ArchivesSpace, AtoM, Axiell — are built around ISAD(G) and its successor frameworks. For more on how this applies in practice, see our guide for professional archivists.

If you are a community archive, local history group, or small heritage organisation without a formal accreditation requirement, you have more flexibility — but that can make the choice harder, not easier. Dublin Core is a practical starting point because it is simple to implement and makes catalogue records discoverable online. If your long-term aim is to share records with a county record office or a national aggregator, aligning with ISAD(G) item level descriptions from the start will save significant rework later. Our community archives page covers tools and workflows suited to volunteer-run collections.

If you are producing finding aids for external sharing or funder reporting, you will almost certainly need EAD output at some point, regardless of which description standard you use internally.

Minimum description vs full archival description

Standards documentation presents ideals. Real cataloguing involves backlogs, limited staff, and collections that have sat undescribed for decades. The question of how much description is enough is one every archivist faces.

A minimal record — reference number, title, date, and a brief scope note — is far better than no record. It makes an item findable, captures basic provenance, and gives the next researcher or cataloguer something to work from. Full description to ISAD(G) or SPECTRUM standard adds significant value: related materials, access conditions, custodial history, detailed scope and content notes. That level of description is worth the effort for high-priority collections, items with complex rights situations, or material being prepared for external access to collections.

For bulk backlogs of routine administrative records, a consistent minimum standard applied across the whole collection is usually the right trade-off. Draw up a list of mandatory fields, apply it uniformly, and document your choices. A catalogue where every record meets the same minimum is far more useful than one where description is uneven.

How AI is changing the cataloguing workload

The practical barrier to proper archival cataloguing has always been time. Describing a collection of several thousand items to ISAD(G) standard is a substantial project — trained cataloguers are expensive, and most archive services carry long backlogs because the manual labour involved in creating quality catalogue records outpaces available resource.

AI-assisted tools are beginning to change that calculation. Rather than a cataloguer creating each record from scratch, an AI system can read a document, pre-populate the standard fields — title, date, scope and content, creator, place names — and present a draft for human review. The cataloguer’s role shifts from data entry to verification and editorial judgement.

The Archiver is built on this model. It generates metadata to the standard you choose — ISAD(G)-aligned archival description, Dublin Core, or a custom schema — and exports in EAD3, BagIt, and CSV so your catalogue records are ready for whatever system or funder requires them. Explore the full cataloguing features to see how it fits into your workflow. Request early access to try it on your own collection.

---

For broader context on cataloguing practice, see our complete guide to archival cataloguing. If your collection includes museum objects, our guide to cataloguing a museum collection covers SPECTRUM compliance and object-specific fields. For funded digitisation projects, our guide to NLHF metadata standards including ISAD, EAD3, and Dublin Core explains what funders expect.

Ready to start? Request early access to try The Archiver on your own collection — no technical setup required.