ONIX (ONline Information eXchange)

ONIX (ONline Information eXchange) is a family of XML-based metadata standards maintained internationally by EDItEUR, the international group coordinating standards development for book and serials supply chains, working with national/regional bodies including Book Industry Communication (BIC) in the UK and the Book Industry Study Group (BISG) in the US. ONIX standards give publishers, distributors, retailers, libraries, and metadata aggregators a common, machine-readable format for describing products and licensing terms instead of each trading partner requiring its own bespoke feed.

Scope: three related standards under one name

“ONIX” is not a single schema but a family of related standards sharing common design conventions:

• ONIX for Books — the original and most widely deployed member of the family, used to describe print and digital book products (title, contributors, subject classification, pricing, availability, cover images, and related product relationships) for retail and library supply chains.
• ONIX for Serials — a set of related formats (including ONIX-PL for encoding license terms and ONIX-SOH/ONIX-SPS for serials holdings and subscription information) used primarily for journal and serials metadata exchanged among publishers, subscription agents, and libraries.
• ONIX for Licensing Terms — used to express the rights and license conditions attached to content products in a structured, machine-actionable way.

Within the publishing cluster, “ONIX metadata” most commonly refers to ONIX for Books, and that is the primary sense covered on this page.

Current version

ONIX for Books Release 3.0 is the current major version of the standard, first published in 2009 and not backward-compatible with the earlier ONIX 2.1 message format it replaced. Release 3.1, published in 2023, extends 3.0 with additional functionality while remaining largely (though not fully) compatible with it. EDItEUR continues to maintain and periodically revise the 3.x specification, including its associated code lists, which are updated on a regular release schedule independent of the base message format. See EDItEUR’s own Release 3.0/3.1 downloads page for the current specification and code lists.

What makes a record ONIX-conformant

An ONIX record is not simply “XML that describes a book.” To count as ONIX, a metadata feed must:

1. Validate against one of EDItEUR’s published ONIX schemas (XSD) or DTDs for the relevant release (e.g., ONIX for Books 3.0).
2. Use ONIX’s defined composite structure — nested groupings such as descriptive detail, collateral detail (cover images, reviews, descriptions), publishing detail, and product supply — rather than a flat or ad hoc field list.
3. Populate controlled ONIX code lists (for product form, contributor role, subject scheme, and so on) rather than free-text equivalents, so that downstream systems can process the record automatically without human interpretation.

Worked examples

• A trade publisher sends a nightly ONIX for Books 3.0 feed to a national bibliographic database and to major online retailers, each <Product> record carrying an ISBN, title, contributor block, BISAC/Thema subject codes, price, and publication date — a single feed serving every downstream partner that accepts ONIX.
• A journal publisher and a subscription agent exchange serials holdings and licensing terms using ONIX for Serials/ONIX-PL, so the agent’s system can automatically reflect what a library is entitled to access under its license.

Counter-examples

• A publisher’s internally formatted product spreadsheet or a proprietary CSV export is not a metadata schema-conformant ONIX record, even if it contains the same underlying bibliographic facts — it lacks the standard’s schema conformance and code lists.
• MARC records used for library cataloguing are a related but distinct standard; some systems crosswalk between MARC and ONIX, but a MARC record is not itself an ONIX record.
• A DOI or a Crossref deposit XML/DataCite metadata schema record identifies and registers scholarly content for persistent citation; ONIX instead describes a product for supply-chain and rights purposes and does not itself mint or resolve identifiers, though ONIX records commonly carry ISBNs and, increasingly, DOIs as identifiers within the product metadata.

Why it matters for research administration and publishing metadata

For organizations at the intersection of scholarly publishing and library/institutional systems, ONIX is one of several metadata standards that interoperate rather than compete: ONIX handles product description and supply-chain/rights metadata for books and serials, while standards like DOI registration schemas (Crossref, DataCite) handle persistent identification and citation metadata, and NISO standards such as Access and License Indicators (ALI) handle machine-readable access-status signalling. A single scholarly monograph or serial title can carry all of these simultaneously, each serving a different function in the metadata stack.

See also CASRAI’s guide to types of metadata for how ONIX’s descriptive/administrative product metadata compares to other metadata categories used in research and publishing infrastructure.