The UK Archivist's
Migration Guide
From CALM, Adlib, and spreadsheets to modern open-source systems.
No catalogue yet? If your archive has never had an online catalogue — still working from Word documents, spreadsheets, or paper lists — see our Get Your Archive Online guide first. This page covers migration from existing systems.
Why Migrate Now?
Many UK archives are still running on CALM (Axiell), Adlib, or bespoke Access/FileMaker databases — systems that are expensive to licence, difficult to integrate, and increasingly out of step with the IIIF and linked-data landscape. The good news: migration to open-source platforms has become significantly more reliable, and the tools and expertise now exist to do it without losing your data.
Step 1 — Audit Your Current Data
Before any migration, understand what you have:
- How many descriptions exist, at what levels (fonds, series, file, item)?
- Are authority records (creators, subjects) maintained separately or embedded?
- Are there associated digital objects — scans, PDFs, born-digital files?
- What fields are actively used vs. legacy populated fields no longer in your workflow?
- Are there any custom fields or locally-defined terms that need mapping?
Export from CALM
CALM can export to XML and CSV formats. Request a full export from Axiell before your licence expires. Ensure you export:
- All catalogue records (all levels)
- All authority files (names, places, subjects)
- All accession records
- Any linked location/location management data
Export from Adlib / Axiell Archive
Adlib uses its own XML structure. Export via the Adlib Designer or request a data extract from Axiell. The XML maps relatively cleanly to ISAD(G) fields, which is the basis for AtoM's data model.
Step 2 — Map Your Fields to ISAD(G)
AtoM is built around the ICA standards — ISAD(G) for archival descriptions, ISAAR(CPF) for authority records. Most CALM fields map directly, but there are nuances:
- CALM "Reference" → AtoM "Reference code" (ISAD 3.1.1)
- CALM "Title" → AtoM "Title" (ISAD 3.1.2)
- CALM "Date" → AtoM "Dates of creation" (ISAD 3.1.3) — watch for non-standard date formats
- CALM "Extent" → AtoM "Extent and medium" (ISAD 3.1.5)
- CALM "Scope & Content" → AtoM "Scope and content" (ISAD 3.3.1)
- CALM "Access Conditions" → AtoM "Conditions governing access" (ISAD 3.4.1)
- CALM "Creator" → AtoM authority record linked via relationship
Step 3 — Clean Your Data
This is often the most time-consuming step, but it's also an opportunity. Common issues in UK archival data include:
- Inconsistent date formats (YYYY, YYYY-MM, circa, undated)
- Authority records with spelling variants or duplicate entries
- Free-text fields containing structured information that should be split
- HTML entities or legacy character encoding from older database exports
- Hierarchy issues — descriptions at wrong levels or missing parent records
We recommend OpenRefine for data cleaning — it's free, powerful, and widely used in the UK archival community for exactly this task.
Step 4 — Prepare Your AtoM CSV
AtoM's import format is well-documented. The CSV import handles all description levels and authority records in separate passes. Key points:
- The
legacyIdandparentIdcolumns establish your hierarchy - Authority records (creators, subjects) are imported in separate CSV files
- Digital object paths can be included to link scans and born-digital files on import
- Run a test import on a staging instance before the full migration
Step 5 — Migrate & Verify
We run all migrations through a staging environment first. You can review the imported data before going live — checking that hierarchies are correct, authority links have resolved, and digital objects are accessible. We then run a record count comparison and spot-check against your source system before the final cutover.
Common Migration Sources We Handle
- CALM (Axiell) — XML/CSV export → AtoM
- Adlib / Axiell Archive — XML export → AtoM
- CONTENTdm — metadata + digital objects → IIIF + AtoM (see below)
- Preservica — preservation packages → Archivematica or AtoM (see below)
- PastPerfect — museum catalogue → CollectiveAccess or AtoM (see below)
- Microsoft Access databases — bespoke mapping
- Microsoft Excel / Google Sheets — CSV mapping
- EAD 2002 / EAD 3 XML — direct import to AtoM or ArchivesSpace
- Older ICA-AtoM (pre-2.x) — database upgrade
- Filemaker Pro — bespoke extract and mapping
- Word / PDF finding aids — structured parsing (see below)
CONTENTdm Migration
CONTENTdm (OCLC) is widely used in UK university libraries and special collections for digital object management. It holds 206 instances in the UK archives research database — the third most common system after CALM and ArchivesSpace.
What the migration involves
CONTENTdm uses Dublin Core metadata, which maps well to AtoM's archival description fields, though some work is required to restructure flat DC records into a hierarchical ISAD(G) model. Digital objects (images, PDFs, audio/video) are extracted via the CONTENTdm API and republished as IIIF manifests, giving you full standards-compliant image delivery without ongoing OCLC fees.
- Dublin Core metadata exported via CONTENTdm API or OAI-PMH harvest
- Digital objects transferred and republished with IIIF Presentation API manifests
- Full-text search preserved (IIIF content search API or AtoM's Elasticsearch index)
- Compound objects (multi-page documents) handled as IIIF ranges
Preservica Migration
Preservica is a commercial digital preservation platform used by some UK archives for long-term storage of born-digital collections. It stores content in OAIS-compliant Archival Information Packages (AIPs) and is well-regarded for preservation workflows, but its annual licence cost is substantial.
What the migration involves
Preservation content in Preservica can be exported as SIPs and re-ingested into Archivematica, the leading open-source digital preservation platform. Archivematica produces OAIS-compliant AIPs stored in Arkivum, DuraCloud, or local storage — with full PREMIS provenance metadata preserved.
- Structured folder export from Preservica with PREMIS metadata
- Re-ingest into Archivematica — full preservation action audit trail preserved
- AtoM integration for public-facing description of preserved born-digital content
- Storage strategy review: local, cloud, or hybrid
We do not suggest Preservica is technically inferior — it is a capable platform. This path is for institutions whose budget priorities have changed or who want to move to open-source infrastructure.
PastPerfect Migration
PastPerfect is a US-developed museum collections management system used by some UK smaller museums, heritage organisations, and society archives. Its web publishing module (PastPerfect Online) is limited and not standards-compliant.
What the migration involves
PastPerfect data is exported in its proprietary format or via MARC/Dublin Core. We map it to CollectiveAccess (the leading open-source museum collections management system) or, for collections better described as archives, to AtoM under ISAD(G).
- PastPerfect export (text export or MARC output)
- Field mapping to CollectiveAccess or AtoM data model
- Media files (images, documents) linked to migrated records
- Public web catalogue — fully standards-compliant, indexed by search engines
Word Documents & PDF Finding Aids
Many UK archives — particularly religious, society, and business archives — have no catalogue system at all. Their finding aids exist as Word documents or PDFs: hierarchically structured text, but not structured data.
What we do
We parse and convert Word and PDF finding aids into ISAD(G)-structured CSV files suitable for import into AtoM. This involves:
- Parsing document structure — headings become hierarchy levels (fonds, series, file, item)
- Extracting dates, extents, and scope notes into the correct ISAD(G) fields
- Identifying creators and building ISAAR(CPF) authority records
- Handling inconsistencies in the original document (which inevitably exist)
- Cleaning and normalising the data before import
This is the most manual process we undertake — it cannot be fully automated — but it is also the most common starting point for UK archives that have never had a catalogue system. The result is a properly structured, searchable public catalogue from what was previously a locked-away document.
If your data problem goes beyond standard migration — bespoke legacy databases, large photographic collections needing batch IIIF processing, or semantic search over transcribed text corpora — see our Advanced Data Services page.