What is a DOI? A guide to DOIs, Crossref and ORCID for academic journals

A DOI (Digital Object Identifier) is a persistent identifier assigned to an article or other scholarly output that always leads to the same content, even when its web address changes. Academic journals use DOIs so that their articles can be cited permanently, linked to one another through reference lists and recognised reliably by indexes. For journals, DOI registration is usually done through Crossref membership, and a DOI is only as valuable as the metadata registered with it.

What is a DOI and how does it work?

The DOI system is defined by the ISO 26324 standard. Every DOI has two parts: a prefix starting with 10. that identifies the registrant, and a suffix after the slash that identifies the content. In a DOI such as 10.xxxx/journal.2026.001, the prefix belongs to the journal or publisher and the suffix to a specific article.

A DOI's strength lies in resolution. Placed after doi.org, the DOI becomes a link that redirects to the current article address registered for it. When a journal moves its website, it only updates the address in the registration; DOI links in reference lists all over the world keep working.

So a DOI is not a web address but a permanent route to one. The article page's address may change over time, but once assigned, a DOI never changes and is never reassigned to other content.

DOIs are not only for articles. A whole journal, individual issues, books, book chapters, conference papers, theses and datasets can all receive DOIs. For journals the most common use is the article-level DOI; some also register at issue level, so that full-issue PDFs and tables of contents can be cited permanently as well.

A DOI is not an address but a permanent route to one; how reliable that route is depends on the quality of the metadata behind it.

Why should academic journals use DOIs?

The most basic reason is citation permanence. When authors and readers cite an article by its DOI, the citation is unaffected by changes to the journal's technical setup. Broken links are one of the most common problems in academic reference lists, and DOIs solve it structurally.

The second reason is visibility. The metadata registered with a DOI, including title, authors, publication date, journal details and references, is read by academic search services and reference managers. That way the article is listed with correct details across platforms and citations are matched to it correctly.

The third reason is credibility. Many indexing bodies and research assessment processes see persistent identifiers on articles as a quality signal. A DOI does not prove a journal's quality on its own, but it marks a regular, professional publishing operation.

A fourth benefit is citation linking. When a journal deposits its articles' reference lists with DOIs, readers can reach each cited work in one click; the same mechanism works in reverse, making it possible to see which other publications cite an article. These links turn each article into part of a connected network within the scholarly literature.

How does Crossref membership work?

Crossref is one of the agencies that register DOIs for scholarly publications and is the most widely used among journals. A journal or publisher joins Crossref, receives its own DOI prefix and registers its articles under that prefix. Membership can be direct or, in some cases, through a sponsoring organisation that represents members.

Membership has an annual fee plus a registration fee for each item deposited. Since the fee structure can vary with the size of the organisation and the type of content, check current figures directly on Crossref's official pages.

Membership also brings obligations, chief among them keeping registered metadata accurate and current, updating the record when an article's address changes, and displaying the DOI on the article page as a full link. Depositing reference lists along with the metadata allows citation links between articles to be built.

Having the journal's basic details ready before applying speeds things up: its ISSN, the publisher's official name and contact details, the permanent web address where articles are published and the person responsible for deposits. Once membership is approved, making the first deposits as a small test batch and checking the results catches schema errors before they spread across the whole archive.

Why does metadata quality matter in DOI registration?

A DOI is registered by submitting an XML file that follows Crossref's schema. The file contains not just the article's address but everything that describes it. Missing or wrong metadata does not stop the DOI from working, but it seriously weakens the article's discoverability and citation matching.

• The article title should match the published version exactly, with consistent capitalisation.
• Each author's given name and surname should be recorded in separate fields, in the order shown on the article and spelled consistently.
• Authors' ORCID iDs should be included in the metadata together with their affiliations.
• Publication date, volume, issue and page details should match what appears on the journal website.
• The abstract and licence information, such as the Creative Commons licence used, should be included.
• The reference list should be deposited in structured form, with the cited works' own DOIs wherever possible.

Converting all of this into XML by hand for each article is both laborious and error-prone. Collecting metadata in structured fields during submission and production, and generating the deposit file automatically from that data, is the most reliable way to keep quality sustainable.

Metadata lives on after publication. When an article is corrected, retracted or its author details are updated, the DOI record must be updated too. Publishing corrections and retractions as separate notices linked to the original article ensures readers always reach the current, accurate version.

How are DOI and ORCID related?

A DOI identifies an output; an ORCID iD identifies a researcher. ORCID iDs are persistent identifiers of sixteen characters in four groups, assigned to researchers. They remove the confusion caused by similar names, surname changes and different spellings, gathering all of a researcher's work under one identity.

Used together, the two identifiers form a strong link: when authors' ORCID iDs are included in an article's DOI record, the relationship between article and author becomes machine-readable. This makes it easier to add the publication to the researcher's ORCID record and to attribute it to the right person in academic assessments.

The practical takeaway for journals is that the ORCID iD should be collected as a separate field for each author at submission, displayed on the article page and included in the DOI record. Having authors connect the iD to their own accounts prevents mistyped identifiers.

ORCID is increasingly used for reviewers and editors as well. Being able to show review contributions on an ORCID record helps this voluntary work gain academic recognition. Keeping ORCID iDs in the journal's reviewer and editor records also prevents people with the same name from being confused.

What are the common mistakes with DOIs?

The benefit of a DOI depends on consistent use. The mistakes below are common in journals, and most disappear once the registration process is set up properly.

• The DOI is shown on the article page but never registered with Crossref, so the link does not work.
• The website moves but the addresses in DOI records are not updated, so DOIs point to the old location.
• The DOI is written only as prefix and suffix rather than as a full, clickable link.
• Title or author corrections made after publication are not reflected in the DOI record.
• Suffixes are assigned inconsistently and the journal cannot track which DOI belongs to which article.

The common fix is to make DOI registration a step in the publishing process. Creating the record when an article is published, tracking pending and failed deposits in a queue and resubmitting the record whenever metadata changes turns DOI registration from a one-off chore into routine.

Planning the suffix scheme from the start also helps. A consistent pattern built from the journal's abbreviation, the year and a sequence number makes it easy to track internally which DOI belongs to which article. Avoid putting information that might change later, such as the article title or page numbers, into the suffix; the DOI is permanent, but information embedded in it can become wrong.

Frequently asked questions

Can DOIs be assigned to articles in back issues?

Yes, previously published articles can be registered too. The metadata of back issues needs to be reviewed for this, with author names and page details made consistent.

Does a DOI guarantee that a journal will be indexed?

No. A DOI provides persistent identification and visibility, but indexing bodies assess many criteria together, including publication ethics, the review process, publishing regularity and content quality. A DOI contributes to only one of them.

Can conference papers receive DOIs?

Yes, Crossref also offers record types for conference papers and proceedings. The papers must be published and available at a permanent address, with metadata prepared as carefully as for journal articles.

Can an article be published if an author has no ORCID iD?

That depends on the journal's policy. Many journals require an ORCID iD for the corresponding author and encourage co-authors to get one. Registering for an ORCID iD is free and quick.

How much does a journal pay for DOI registration?

Crossref membership involves an annual membership fee plus a registration fee for each item deposited. Since amounts can vary with the organisation's size, the content type and the publication year, check the current fee schedule directly on Crossref's official pages.

Making DOIs part of the publishing process

Set up correctly, DOIs stop being something the editor has to think about again with every issue. Collecting structured metadata at submission, obtaining ORCID iDs from authors and generating the deposit file at publication ensures that every article in the journal is permanent and discoverable.

Planning your metadata and your DOI suffix scheme before applying for Crossref membership is far easier than fixing things later. At Bir Web Ajans we let journals run this process inside their publishing flow in Nova JournalPort, with a DOI queue that generates the deposit XML Crossref expects from article data.