ORCID and Affiliations: Authorship Basics for Authors

Small identity mistakes create big delays, so lock down your ORCID, your official affiliation wording, and your author list before you submit.

Explainer · 8 min read · Directive Publications Editorial Team

ORCID

• Create or claim your ORCID record
• Use the same ORCID on every submission
• Keep employment and works updated

ORCID helps editors and indexers disambiguate names — especially common surnames.

Claim before you create. Search the ORCID registry for your own name first. Publishers, funders and thesis repositories sometimes generate a record on your behalf, so one may already exist with a few of your papers attached to it. Claiming that record is far less work than finding it two years later, when your output is split across two identifiers and neither list is complete. If you do end up with duplicates, ORCID lets you merge them, and merging is easy while there is almost nothing to reconcile.

Same identifier, always. The point of a persistent identifier is that it persists: new employer, new email address, changed name, change of field, same number. If your name is spelled differently across your existing papers, record the variants in your record so a search under any spelling finds you, and record your name in your own script alongside a Latin transliteration if that applies to you. Where a submission system offers sign-in with ORCID, use it instead of typing the number into a text box. Authenticating links the record properly, and a typed identifier with one transposed digit points at a stranger rather than at nobody.

Keep it worth clicking. An empty record does not help anyone. An editor who follows the link is checking two things: whether the employment section shows the institution printed on your manuscript, and whether the works list looks like the person who wrote this paper. Update employment when you move, with start and end dates, and turn on automatic updates for works where your publisher supports it. That is ten minutes a year, not an ongoing chore.

One practical habit: ask co-authors for their ORCID at the same time you ask for their affiliation, not during submission. Chasing four people for identifiers while a half-finished submission sits open is the slowest possible way to collect them.

Affiliations

• Use the official institutional name the university expects
• Match the affiliation to where the work was done (note moves clearly if needed)
• Keep department naming consistent across co-authors when required

Find the official form and copy it exactly. Most institutions publish a required form of their name, usually through the research office or the communications team, and it is more specific than authors expect: a set order for unit, faculty, institution, city and country, a fixed English rendering for institutions that work in another language, and rules about what may be abbreviated. Use it verbatim, punctuation included. If you cannot find the guidance, look at how papers from your department were published over the past year, match the most common form, and ask the research office to confirm it.

The fussiness has a reason. Affiliation lines are plain text, and the systems that count institutional output match on that text. A variant — "Dept." for "Department", a hospital named without its parent university, a missing city — can quietly fail to group with the rest of your institution's papers. The person who notices is rarely you. It is whoever compiles the annual research report and cannot find your paper in it.

Affiliation follows the work, not your current desk. If you ran the study at one institution and have since moved, list the institution where the work was done and add a present-address footnote for where you are now. Journals expect this and have a convention for it. Dropping the former affiliation because you no longer work there is the common error, and the damage goes past etiquette: the institution that hosted and funded the work cannot count it, and a funder reconciling grant outputs may not find the paper attached to the grant that paid for it. The same applies to a co-author who left mid-project — they keep the affiliation they held while doing the work.

Two real affiliations are fine; list both. An affiliation added because it looks impressive, at an institution where none of this work happened, is not a formatting decision. Institutions and funders read the affiliation line as a claim about where research was carried out.

Agree one string per unit. Three co-authors from the same department will write that department three different ways if left to themselves. Settle on one string in the shared draft, paste it, and use superscript numbering rather than retyping it under each name. Do the same for shared centres, hospitals and grant-funded units, some of which have required wording attached to their funding.

Authorship basics

Authors should meet accepted contribution criteria (commonly ICMJE’s four criteria in medicine). Record roles with CRediT where possible.

What authors most often miss is that the criteria are cumulative rather than a menu. In the ICMJE formulation, an author has made a substantial contribution to the work — its conception or design, or the acquisition, analysis or interpretation of the data — and has drafted or critically revised the manuscript, and has approved the version being submitted, and agrees to be accountable for it. A colleague who generated data but has never read the draft does not yet qualify. The right response is to send them the draft and ask for their revisions, not to quietly drop them.

CRediT gives you standard names for contributions — conceptualization, methodology, formal analysis, writing of the original draft, review and editing, supervision, funding acquisition and others — and drafting the statement is a useful test in itself. Write one line per person before you submit. If you cannot describe someone's contribution in a sentence without stretching, they belong in the acknowledgments.

A decision rule for the awkward cases: funding the study, heading the department, supplying a reagent or an existing dataset, and a single pass of language editing are all contributions worth acknowledging, and none of them is authorship on its own. That cuts both ways. The student who designed and ran the analysis is not a technical assistant because they are junior, and moving them to the acknowledgments is the same failure as adding a name that did nothing.

Have the authorship conversation at the start of the project, repeat it when the scope changes, and write down what you agreed, author order included. The criteria say who is an author; they do not set the order, disciplines differ, and the assumption nobody said out loud is what becomes a dispute in month eleven. Equal-contribution notes are useful when they are true and corrosive when they are handed out to keep the peace.

Read: Who Counts as an Author? · Authorship Disputes

Before you submit

1. All names spelled correctly (including accents)
2. Corresponding author contact is institutional and monitored
3. Equal-contribution notes only if truly accurate
4. Acknowledgments do not hide gift authorship

Check that accents and diacritics survived the move from your manuscript file into the submission form, then check them again on the proofs, which is where they most often break. Enter each person into the given-name and family-name fields the system labels rather than typing a full name as you would write it. For colleagues whose family name comes first, that field mix-up is a persistent metadata error, and it copies itself into every index that later picks up the record.

For the corresponding author, use an institutional address that someone will still be reading in a year. A student's university account often closes soon after they graduate, and proofs and author queries sent to a dead mailbox stall a paper that was otherwise finished. If the journal allows a second corresponding author or an additional contact, use it, and make sure whoever is named actually checks that inbox during the months a submission can sit.

Ask permission from everyone you name in the acknowledgments. Being thanked in print reads as endorsement of the work, and some people will decline. Then read the section in the other direction: a senior name parked there to avoid an awkward authorship conversation is the same problem as gift authorship, moved one line down the page.

If you find a mistake after submitting

Before proofs, most of this is a short message to the editorial office. Say what is wrong and what it should be — a misspelled name, a lost accent, a corrected affiliation, an ORCID that needs adding — and it is usually handled as routine.

Author list changes are treated differently at every stage, and deliberately so. After submission, journals generally ask for a written explanation of why the change is needed, together with confirmation from every author, including the person being added or removed. After acceptance the same request arrives at a worse moment. After publication it becomes a correction notice attached permanently to the paper, where readers, institutions and future employers can see it. None of that is a reason to leave an incorrect list standing. It is a reason to settle the list, in writing, before you press submit.

More resources: For Authors · Directive Academy