{"id":3889,"date":"2026-05-13T09:59:47","date_gmt":"2026-05-13T08:59:47","guid":{"rendered":"https:\/\/sites.dundee.ac.uk\/web\/?p=3889"},"modified":"2026-05-13T12:25:33","modified_gmt":"2026-05-13T11:25:33","slug":"making-your-staff-profile-more-visible-for-ai-search","status":"publish","type":"post","link":"https:\/\/sites.dundee.ac.uk\/web\/making-your-staff-profile-more-visible-for-ai-search\/","title":{"rendered":"Making your staff profile more visible for AI search"},"content":{"rendered":"\n

This blog post explains how we’re improving our People pages, so that AI tools can find and reference them. Short on time? Jump straight to how you can help<\/a><\/p>\n\n\n\n

A changing search landscape<\/h2>\n\n\n\n

The way in which people are finding content online is changing. Increasingly, they are getting answers directly from AI-powered tools – such as Google\u2019s AI overview, or from within an LLM (large language model) such as ChatGPT – rather than visiting a website itself. <\/p>\n\n\n\n

Recent studies suggest that around 60% of Google searches now end without a click to any external site, with figures climbing even higher on mobile. This is because people can often get what they need straight from AI summaries or overviews right on the results page and have no need to actually visit the website the information is from.<\/p>\n\n\n\n

Our own website analytics show a substantial increase in AI sessions (which is when someone has followed a link from within an LLM to our site) during the past year. This is likely to be only the tip of the iceberg – for every one person who visits, many more people will have seen that link and the information about the University which is shared from it. These impressions are crucial as they help shape people\u2019s perceptions about the University from the start.<\/p>\n\n\n\n

This means that the role of our website is shifting. In addition to being a place for people to visit, our content is increasingly acting as a source from which the AI tools can draw when they generate answers.  <\/p>\n\n\n\n

How do the bots use our content?<\/h2>\n\n\n\n

AI systems don\u2019t treat all information equally. They prioritise content that comes from clearly identified, trustworthy sources, and they look for strong signals that link expertise to real people and institutions. The more we can bolster both the authority and the discoverability of our content, the more likely we are to see it being used.  <\/p>\n\n\n\n

It\u2019s also important to understand that our content isn\u2019t being reused as whole pages. The bots will break it down into smaller pieces, work out how those pieces relate to each other, and then build something new from them – it\u2019s a bit like Lego where you can use the same blocks to construct different models.<\/p>\n\n\n\n

\n

However, too often websites follow my daughter\u2019s approach to Lego! She will dump all the pieces in one big tin, mix them up, and then ask me to give her the right piece at the appropriate moment. It is definitely not easy to find them as quickly as she would like and not very efficient. 

This is where labelling is important. In the same way that it\u2019s easier to build a model when each piece is organised according to size, shape, colour, or some other system, it\u2019s easier for an AI to extract information from our website when we\u2019ve labelled it.<\/p>\n<\/div>

\"\"<\/figure><\/div>\n\n\n\n

Changes to the staff profile schema<\/h2>\n\n\n\n

Our current staff pages are readable to humans, but importantly, also have a machine-readable layer behind them. This is known as schema, which is a worldwide standard, It marks up content on the page in a way that a bot can quickly grab and re-use. <\/p>\n\n\n\n

Here\u2019s an example on one of the staff pages. This code sits behind the page and isn’t visible to visitors:<\/p>\n\n\n\n

\"\"<\/a><\/figure>\n\n\n\n

By structuring information in this standardised format, search engines and AI systems can quickly recognise who the person is, where they work, and how they relate to other trusted sources. For example, the line \"@type\": \"Person\"<\/code> tells machines that the page is about an individual, while fields such as \"jobTitle\": \"Lecturer\"<\/code> and \"worksFor\": \"University of Dundee\"<\/code> explicitly link that person to their role and institution.<\/p>\n\n\n\n

One change we\u2019ve made recently is adding more sameAs<\/code> fields to staff profile structured data. You won\u2019t see this on the page itself, as it\u2019s a change that is only visible in this machine-readable layer. It enables us to show clearly that the member of staff described on our website is the same person you\u2019ll find on LinkedIn, the same researcher whose work appears on Google Scholar, and the same individual identified by an ORCID. We\u2019d already linked them to their Discovery profile, as you can see in the screenshot above, but now we\u2019ve brought in other top locations where their work is represented.<\/p>\n\n\n\n

This is because the output of our staff doesn\u2019t sit neatly in just one location on our website, making it harder for the bots to link these pieces together and construct the model. Staff publish across multiple platforms, will usually move institutions over time, join collaborations, and sometimes even share the same name as other academics. Without clear signals, AI systems have to guess which bits of information belong together and this is when things go wrong – I\u2019m sure everyone is familiar with AI\u2019s tendency to hallucinate when it doesn’t have the answer! <\/p>\n\n\n\n

We focused on LinkedIn, Google Scholar and ORCID because they cover different aspects of identity. LinkedIn reinforces professional context and continuity. Google Scholar anchors publication history and citations. ORCID provides a persistent identifier that reduces ambiguity, particularly where names are shared. Together, they give machines a much clearer picture of the person behind the profile.<\/p>\n\n\n\n

With the bots being more confident that they can successfully merge information from different sources, our staff are more likely to be recognised as authoritative voices, research is more likely to be attributed correctly when it is surfaced in AI answers, and our institutional content is more likely to be used when relevant questions are asked within LLMs. <\/p>\n\n\n\n

This in turn has a knock-on effect, since strong, well-linked individual profiles help establish the University as a go-to source of expertise. When AI tools confidently attribute research, teaching, or insights to our academics (and link those back to Dundee), they’re much more inclined to pull in related institutional content, such as our courses or research information, rather than using competitor sources. So better-connected staff profiles don’t just benefit you as a person but actually can improve our overall University presence in those invisible AI conversations that prospective students are having every day.<\/p>\n\n\n\n

Adding these sameAs<\/code> fields is one of the many things we do behind the scenes that largely goes unnoticed. However in a small way it can help to strengthen the trust infrastructure beneath our content.<\/p>\n\n\n\n

How you can help<\/h2>\n\n\n\n

If you already have a LinkedIn profile, a Google Scholar profile, or an ORCID in text form on your profile, we have already taken this information and applied it to your People page, so no action is required.<\/p>\n\n\n\n

If you would like to add<\/strong> one or more of these, please share your links with us at web@dundee.ac.uk<\/a>. We can then make sure your profile is accurately connected and hopefully more visible in future AI searches.<\/p>\n\n\n\n

<\/p>\n","protected":false},"excerpt":{"rendered":"

This blog post explains how we’re improving our People pages, so that AI tools can find and reference them. Short on time? Jump straight to how you can help A changing search landscape The way in which people are finding content online is changing. Increasingly, they are getting answers directly from AI-powered tools – such […]<\/p>\n","protected":false},"author":281,"featured_media":3956,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[78],"tags":[],"class_list":["post-3889","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-content"],"blocksy_meta":[],"_links":{"self":[{"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/posts\/3889","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/users\/281"}],"replies":[{"embeddable":true,"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/comments?post=3889"}],"version-history":[{"count":12,"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/posts\/3889\/revisions"}],"predecessor-version":[{"id":3957,"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/posts\/3889\/revisions\/3957"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/media\/3956"}],"wp:attachment":[{"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/media?parent=3889"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/categories?post=3889"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sites.dundee.ac.uk\/web\/wp-json\/wp\/v2\/tags?post=3889"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}