Why ChatGPT Never Mentions Your School: The 60-Day Fix

UK institutions lead Europe with a 29% ChatGPT mention rate and 38% on Perplexity — yet the European average sits at just 19%, and most private schools score close to zero (Source: Skolbot GEO Monitoring, 500 queries × 6 countries × 3 AI engines, Feb 2026). That gap is not driven by brand recognition. It is driven by five structural problems that this 60-day plan resolves, one phase at a time.

For the full GEO framework for higher education, see our complete GEO guide for schools.

The 5 Reasons ChatGPT Never Mentions Your School

Large language models do not search for the best schools. They search for the best-documented entities in their training corpus and real-time sources. Here is why your institution is not yet among them.

Reason 1 — GPTBot and AI crawlers are blocked

If your robots.txt contains a blanket Disallow: / or explicitly blocks OAI-SearchBot, ChatGPT cannot index your content in real time. This is the most common cause of zero citations and the fastest to fix. The GPTBot documentation specifies exactly which user agents are involved. Many UK institutions inadvertently block these bots via legacy firewall rules or CDN configurations.

Reason 2 — No Schema.org EducationalOrganization markup

Without Schema.org EducationalOrganization structured data, your institution is an anonymous block of text to AI engines. They cannot identify it as a distinct verifiable entity, link it to its TEF rating, QAA accreditation or UCAS course listings, or compare it meaningfully against competitors. Structured markup is the entry ticket to AI-generated answers.

Reason 3 — No citable content

A "Our Programmes" page listing course names without verifiable data provides AI engines with nothing to extract. ChatGPT cites facts: duration, annual tuition, graduate employment rate, UCAS tariff points, TEF Gold or Silver rating. Without those figures — sourced, dated and visible in HTML — your page is functionally invisible to generative models.

Reason 4 — Absent from reference sources

AI engines cross-reference information across multiple sources before generating a citation. If your institution does not appear on UCAS, the QAA register, the OfS register, the Guardian University Guide or any QS or THE ranking, the AI has no external corroborating signal to support a mention.

Reason 5 — No structured social proof

Verified reviews, Alumni outcome data, TEF rating with a direct link to the OfS register entry, National Student Survey (NSS) scores — these constitute the structured social proof that LLMs use to legitimise a recommendation. Degree-awarding bodies that lead in GEO visibility consistently surface these signals prominently in their structured data. Institutions that omit them are leaving a material citation signal unused.

Summary of the 5 reasons

Days 1–20 — Technical Foundations

This phase removes the blockers that prevent AI engines from recognising your institution at all. Without these foundations in place, the content improvements in the next phase will have limited effect.

Open access to AI crawlers in robots.txt

Open yourinstitution.ac.uk/robots.txt and confirm that OAI-SearchBot (OpenAI/ChatGPT) and PerplexityBot are not disallowed. If a global Disallow: / directive is active, you are excluding yourself from all real-time AI corpora. Add the following explicitly:

User-agent: OAI-SearchBot
Allow: /

User-agent: PerplexityBot
Allow: /

This change takes under an hour to implement. If you use a WAF or CDN, verify that rate-limiting rules are not silently blocking these user agents at the network level. A blocked crawler means zero citations, regardless of content quality.

Implement Schema.org EducationalOrganization

Deploy JSON-LD Schema.org EducationalOrganization markup on your homepage and About page. The priority fields for LLMs are: name, url, logo, address, telephone, foundingDate, numberOfStudents, accreditation (TEF rating, QAA status, AACSB, AMBA, EQUIS where applicable), and sameAs (links to your Wikidata entry, LinkedIn page and UCAS profile). Google Search Central provides the canonical implementation reference and a validation tool.

Schools with structured Schema.org data gain an average of +12 points in AI visibility compared to those without (Source: Skolbot GEO Monitoring, 500 queries × 6 countries × 3 AI engines, Feb 2026). This is the highest-ROI GEO lever: a single technical implementation with a lasting compounding effect.

Audit your site's crawlability

Run a crawl of your key pages using Screaming Frog or Sitebulb. Identify pages returning 4xx or 5xx errors, pages blocked by noindex tags, and content hidden inside JavaScript rendering that crawlers cannot access. Programme pages, FAQ pages and data pages must all be accessible in plain HTML without authentication walls. PDFs locked behind lead-capture forms are invisible to AI crawlers and should be replaced with open HTML equivalents — a requirement also aligned with OfS transparency expectations.

Add Course or EducationalOccupationalProgram markup to your three highest-traffic programme pages. Complete the fields educationalCredentialAwarded, provider, tuitionInfo, occupationalCategory and applicationDeadline. These are the fields LLMs query first when generating programme recommendations.

Days 21–40 — Citable Content

Once AI engines can identify your institution, they need content they can extract and cite. This phase reformats existing pages and creates a small number of purpose-built assets.

Build FAQ pages with direct answers

Every programme page needs a FAQ marked up in Schema.org FAQPage JSON-LD. Questions must mirror real prospect queries: "What is the graduate employment rate for the LLB?", "What UCAS points do I need for the MSc Data Science?", "Is a placement year available?" Each answer should be 50–100 words, contain at least one verifiable figure, and use no marketing language.

FAQ markup serves two functions simultaneously: it answers human prospects and delivers an extraction-ready format to LLMs. An unmarked FAQ page is roughly half as effective. Aim for 8–12 questions per programme, written as if answering a UCAS adviser, not a prospectus reader.

Surface your proprietary outcome data

Your graduate outcome statistics are a direct competitive advantage in GEO. Publish them in HTML — not only in a downloadable PDF — with: employment rate at 6 and 12 months, median starting salary, cohort size, HESA data year. Bear in mind that the +12-point visibility gain from Schema.org only materialises when the structured content contains real, citable data (Source: Skolbot GEO Monitoring, Feb 2026). Markup without substance does not produce citations.

A practical before-and-after:

Before: "Our graduates secure roles at leading employers across all sectors."

After: "92% of the 2024 cohort secured graduate-level employment within six months of completing their programme (HESA Graduate Outcomes 2025, 287 respondents). Median salary at 12 months: £33,800."

The second version contains four verifiable data points. AI engines extract facts, not ambitions.

Build structured data tables for every programme page

Tables are the most extractable format for large language models. A structured HTML table with descriptive column headers and numerical data is cited far more frequently than a paragraph containing the same information.

Each programme page should include a table covering:

Reference accreditations by their full name and acronym — each is an entity that LLMs can verify against the awarding body's own website. Linking your TEF rating directly to your OfS register entry adds a trust graph link that AI engines weight as corroborating authority.

Days 41–60 — Off-Site Amplification

The technical foundations are in place, the content is structured. This phase builds the external citation network that AI engines cross-reference to validate your on-site information.

Wikipedia and Wikidata presence

Wikipedia is one of the sources most consistently cited by LLMs as a primary entity reference. If your institution does not have a Wikipedia article, verify eligibility (sufficient published secondary sources). If one exists, audit it: check that the founding year, student numbers, TEF rating and key accreditations are accurate, current and cited with external references. A Wikidata entry linked via the sameAs field in your Schema.org amplifies the entity recognition signal further.

Complete your LinkedIn Company Page

LinkedIn is indexed by LLMs as a verifiable entity signal. Ensure your page is fully populated: institutional description, specialisms, sector, headcount (student numbers), all campus locations. Post regularly with factual content — graduate outcome announcements, TEF results, league table movements with ranked figures. An incomplete or dormant LinkedIn page reads as a low-authority entity to AI engines. A well-maintained one reinforces the cross-source consistency that LLMs use to decide whether to cite an institution.

Press coverage from Tier 1 sources

A data-rich article in Times Higher Education, The Guardian Higher Education Network, Wonkhe or Tes Higher that names your institution alongside verifiable figures is a citation that both ChatGPT and Perplexity retain. The strongest pitches are built around proprietary outcome data: TEF performance analysis, widening participation statistics, a sector-first initiative with measurable results. A single article citing your institution by name can produce visible improvements in AI citation rates within four to six weeks of publication.

Avoid opinion pieces without supporting data. AI engines cannot extract a citable claim from an argument. They extract statistics.

Secure and update aggregator profiles

Verify and update your profiles on the platforms that AI engines treat as authoritative corroborating sources for UK higher education:

• UCAS: Complete all course listings with accurate employability data and entry requirement fields
• Guardian University Guide: If you feature, confirm that the figures in the Guide match your published data
• OfS Register: Verify your TEF rating and course-level data are current
• QAA: Ensure your Enhancement-Led Institutional Review outcome is accurately described on your own site with a direct outbound link to the QAA
• QS and THE: Claim your institutional profiles and populate all available data fields

Mismatches between your website and your UCAS listing or OfS entry reduce your AI citation probability — they are trust signal failures in the cross-reference graph LLMs use to validate claims.

Set up your GEO monitoring protocol

Without measurement, you have no way of knowing whether days 1–40 produced any effect. Our article on monitoring GEO visibility for schools covers the full methodology. The core protocol: test 20 strategic queries (10 branded, 10 generic) across ChatGPT, Perplexity and Google AI Overviews monthly and track mention rate, accuracy of cited data and which pages are cited as sources.

For the KPIs to prioritise, see our guide on ChatGPT and Perplexity visibility metrics for schools. For maximising AI citation on programme pages specifically, our programme page BOFU checklist is the companion resource. For a deeper read on every signal LLMs evaluate before recommending an institution, see our analysis of 15 LLM signals for school recommendation.

60-Day Dashboard

At day 60, run the same 20-query test you used to establish your baseline. Perplexity reacts within three to six weeks of content changes. ChatGPT is slower — expect six to ten weeks because its corpus updates less frequently. If your branded query citation rate remains below 50% at day 60, return to phase 1: the entity recognition layer is incomplete. If generic query citation is below 10%, the content in phase 2 needs additional data density.

For an extended 90-day plan covering advanced amplification and measurement, see our 90-day plan to get cited by ChatGPT and Perplexity.

FAQ

Why is ChatGPT citing my competitors but not my institution?

The institutions being cited almost certainly have at least two of the three elements your school lacks: active Schema.org markup, mentions on authoritative external sources (UCAS, OfS, QAA, sector press), and factually dense content with verifiable figures. All three are reproducible advantages. This plan addresses each one in sequence.

Does a Russell Group affiliation automatically improve AI visibility?

It helps, but it does not guarantee citations on specific queries. Russell Group universities benefit from large representation in AI training corpora through research output, press coverage and rankings. Post-92 institutions and specialist providers can close the gap on niche queries — a music conservatoire with a complete TEF Gold Schema.org entry will appear on "best conservatoire in the UK" queries before a research-intensive university whose website has no structured data on its music programmes. Specificity and structuration outweigh brand history on long-tail queries.

How does our TEF rating affect AI citation rates?

The TEF rating is cited by AI engines — particularly Perplexity — as an official UK quality signal drawn from the OfS register. A Gold or Silver rating that is clearly stated in your Schema.org markup, visible on your key pages, and linked directly to your OfS register entry acts as a corroborating authority signal that makes citations materially more likely. Many institutions hold a strong TEF rating but bury it in prose — a missed opportunity that this plan corrects in phase 1.

Do we need a technical agency to run this plan?

Not necessarily. The content and aggregator work in phases 2 and 3 can be managed by an in-house marketing team. The Schema.org implementation in phase 1 requires either a developer or a CMS plugin supporting JSON-LD structured data — most modern university CMS platforms (WordPress, Drupal, Sitecore, Contentful) have this capability built in or available as a plugin. The robots.txt change takes under ten minutes. The primary investment is time and editorial rigour, not budget.

AI engines are already recommending institutions to your prospective students. The question is not whether your school should appear in ChatGPT — it is whether it will appear before your competitors do.