Before the Next Collaborator Clicks: A 2026 Lab Website Integrity Review

A prospective student, a program officer, a journalist, or a collaborator often meets a lab through its website before meeting its people. That first encounter is not merely promotional. It is part of the lab's public scientific record. It identifies who works there, what the group has published, where data and code can be found, and whether outside readers can use the site at all. When that record is out of date or inaccessible, the cost is not cosmetic. It can misattribute credit, obscure reusable outputs, and signal a level of care that the lab may not intend.

As of November 2026, many lab sites still function as set-and-forget brochures. A principal investigator may update a page after a major grant or paper, then leave it untouched for a year or more. Students graduate, postdocs move on, software repositories change addresses, and images remain online without recorded permissions. The result is a public interface that quietly drifts away from the lab's actual work.

A full website redesign is rarely necessary. What is needed is a short, repeatable integrity review: one hour, a small group, and a shared record of what was checked. The aim is not perfection. The aim is to make the public record accurate enough to be useful and accessible enough to be used.

The Website Is Part of the Research Record

Lab communication is often discussed in terms of papers, talks, and press interviews. The website sits alongside those channels, but it has a different function. It is persistent, searchable, and available outside normal working hours. It is also frequently the only public page that connects a lab's people, publications, and outputs in one place.

That persistence creates an obligation. If a publication list omits a paper, a junior researcher may lose visible credit. If a data link points to a defunct repository, a reader cannot reproduce the work. If a people page lists a former student as current, a reference check or collaboration inquiry may begin with false information. These are small failures individually. Together, they weaken trust in the lab's public communication.

Accessibility belongs in the same review. A site that cannot be navigated by keyboard, read by a screen reader, or understood without downloading a malformed PDF excludes qualified readers. Many US institutions now reference WCAG 2.2 AA in their digital accessibility policies, and campus accessibility offices can usually provide guidance. The lab does not need to become a web development team. It does need to check the basics and fix what it can.

A One-Hour Integrity Review

The review works best with two or three people: one who knows the publication history, one who knows the current personnel, and one who can inspect the site itself. Divide the hour into five short passes. Record findings in a single shared document rather than making changes on the fly.

1. Publication and project inventory, 15 minutes

Open the publication page and compare it with the lab's reference manager, Google Scholar profile, or internal annual report. Check the last two years first, then spot-check older entries. Confirm author order, year, venue, and DOI links where available. Look for papers that have moved from preprint to final publication. Check whether project pages still describe active work or have quietly become historical records. If a project has ended, label it as completed rather than deleting it; the record may still be useful.

2. People and roles, 10 minutes

Review every person listed. Confirm current status, role, and, where appropriate, pronouns and name pronunciation. Remove or archive former members who have asked not to be listed, and update alumni entries with current positions when the individuals have made that information public. Check that undergraduate researchers, technicians, and visiting scholars are not omitted. A people page that recognizes only graduate students and faculty gives an incomplete picture of the lab.

3. Data, code, and materials links, 15 minutes

Click every link to a repository, dataset, protocol, or software package. Note broken links, moved repositories, and items that now require access requests. Check whether a DOI or persistent identifier is present. If a repository has changed, update the link and add a short note explaining the move. Confirm that licenses are visible where they exist. A reader should not have to guess whether code is reusable or data are restricted.

4. Images, permissions, and credit, 10 minutes

Review photographs, figures, and graphics. Confirm that the lab has permission to use each image and that credit lines are accurate. Pay particular attention to images of people: a group photo may include someone who has since left or who did not consent to public use. Check whether figures from papers are reproduced with the publisher's permission or under an appropriate license. If provenance is unclear, remove the image until it can be documented.

5. Baseline accessibility, 10 minutes

Run through the site using only a keyboard. Can you reach the main navigation and every link? Check page headings for a logical order. Confirm that images have alternative text, that link text describes its destination, and that color is not the only way information is conveyed. Open one or two PDFs and check whether they are text-based rather than scanned images. If the site has video, confirm captions. These checks will not resolve every accessibility issue, but they catch common barriers and create a list for follow-up with campus support.

From Audit to Habit

A one-hour review is useful only if it recurs. Put the next review on the lab calendar before the current one ends. A quarterly pass of fifteen minutes for people and links, plus an annual hour for the full review, is usually enough for a modest lab site. Assign one owner, but do not make the owner solely responsible for content they did not create. The principal investigator, lab manager, and a student representative can each claim a section.

Keep a short changelog. Record the date, the pages reviewed, the problems found, and the fixes made. This does not need to be public. It is evidence that the lab treats its website as maintained infrastructure rather than a forgotten poster. If the lab has a communication committee, add the website review to its standing agenda. If it does not, the annual review can be folded into an existing lab meeting.

What Good Enough Looks Like

The goal is not a flawless website. It is a site that a careful reader can trust for the basics: who is in the lab, what the lab has published, where the outputs are, and how to reach the group. Accessibility is part of that trust. So is credit. A lab that documents its public face with the same care it applies to a methods section is not indulging in web design for its own sake. It is extending the norms of reproducible research to the place where many readers first encounter the work.

Before the next collaborator clicks, spend an hour on the record they will see. Then schedule the next hour.