AI Visibility & Marketing Agency for Universities

A student in Lagos choosing a master's in public health, a parent in Mumbai comparing engineering degrees, a graduate in Hanoi looking for a funded PhD: more of them now start by asking an AI engine. ChatGPT, Gemini, Perplexity or Google AI Mode replies with a shortlist of five or six universities and a line on each.

That shortlist is the new first round. If your university is not in it, the applicant may never reach your course page. If you are in it but described with last year's fees, the answer works against you. The Rank Master is an AI visibility agency: we track what seven engines say about your institution, prompt by prompt, and work on the facts and sources behind those answers.

The questions students and agents put to AI engines

Applicants rarely ask for the “best university”. They ask with constraints, and the constraints decide who gets named. Typical prompts we map:

• “Best one-year master's in data science in the UK or Ireland under £25,000”
• “English-taught engineering master's in Europe with low tuition fees”
• “Which universities in Canada or Australia have strong nursing degrees for international students?”
• “Universities offering full scholarships to Nigerian students for an MBA”
• “How do employers in India view a one-year master's from abroad?”
• “Top universities for computer science with good graduate jobs, outside the US”
• “Which universities in Dubai offer a British degree in business?”
• “Funded PhD places in climate science starting next autumn”

What global university answers are built from

Engines that search the web before answering typically draw on a familiar mix of sources for university questions:

• Global rankings: the QS World University Rankings, the Times Higher Education World University Rankings, ShanghaiRanking's Academic Ranking of World Universities and U.S. News Best Global Universities.
• Course portals: Studyportals, FindAMasters, FindAPhD and national equivalents, which often hold the cleanest structured record of a programme.
• Reference pages: Wikipedia and Wikidata entries, often used to settle names and basic facts.
• Student discussion: Reddit threads, forums and YouTube reviews, which shape the tone of an answer more than its facts.
• Government pages: official guidance on study visas and post-study work.
• Your own site: course, fees and scholarship pages, when engines can read them.

Which source wins varies by engine and by market, so we test each engine separately.

Where universities lose the answer

The reasons a university is missing or misdescribed tend to repeat:

1. Course pages engines cannot read. Fees, entry requirements and start dates sit in PDFs, JavaScript tabs or course finders behind search forms, so engines quote a portal or an agent's page instead.
2. Stale facts that keep circulating. Last cycle's tuition fee, a closed programme or an old English language requirement can be repeated for months after you changed it.
3. Inconsistent names. A university known by an acronym, a former name or a city variant can be split into two entities, or confused with another institution.
4. Rankings without context. You may sit in a subject table, but if no page explains it, engines rarely connect the ranking to the course being asked about.
5. Thin coverage in the applicant's market. An answer for a student in Vietnam may draw on regional sites where you barely appear.

How we help universities recruit worldwide

What an engagement looks like

We work beside your international office, admissions and marketing teams on one question: what AI engines tell applicants in the markets you recruit from. Every engagement follows our method, adapted to the institution: we map the questions, measure seven engines, fix the facts and sources behind the answers, and report monthly against the universities you lose applicants to. Plans start at $2,000 a month, with custom engagements from $1,500 a month for one campus or one market; see pricing.

What we deliver

• A prompt map by course, market and stage. Each priority programme crossed with your recruitment markets and the applicant's stage, from first shortlist to comparing offers, approved by your teams before tracking starts.
• Market and agent prompt sets. A separate question set for each country you recruit in, phrased the way students there ask about scholarships, visas, recognition at home and post-study work, plus the shortlisting prompts education agents run.
• Course-page fact blocks. Entry requirements by home-country qualification, English language scores, fees, intakes, scholarships, outcomes and the subject rankings that apply, as plain text with structured data. Our guide to AI crawlers and robots.txt covers the access side.
• A source plan for portals, rankings and forums. For each market, the course portals, ranking pages, reference entries and student threads engines cite for your prompts, what each says about you, and the placements we will work to earn on merit.
• A fact-correction log. Every wrong statement, such as last cycle's fee, a closed programme or a confused acronym, with the engine, the source behind it, the fix and the date it stopped appearing.
• Monthly answer-share reporting. Your share of answers by subject, market and engine, against the rival institutions you name.

Your first 90 days

Weeks 1 to 2: map and measure.

• We agree the prompt map: priority programmes, recruitment markets and the rivals you lose applicants to.
• We run the baseline across ChatGPT, Perplexity, Gemini, Claude, Grok, Google AI Overviews and Google AI Mode, asking as an applicant in each priority market would.
• An audit in the first week ranks wrong facts, unreadable course pages and crawler blockers by how many target questions they affect.

Weeks 3 to 6: fix the facts.

• The first course-page fact blocks go live, starting where engines quote a wrong fee or requirement.
• Corrections begin on the portal listings and reference pages behind the worst errors.
• We start on the sources engines cite most in your top two or three markets.

Weeks 7 to 12: build presence and measure movement.

• Fact blocks roll out to the next courses and markets.
• The first earned citations appear on portals, guides and pages the engines read, and we check whether they are quoted.
• At month 3 we report movement on your target questions, market by market. By month 6 you see answer share engine by engine against your named rivals.

What you see each month

A live dashboard shows, for every prompt, what each engine says, which universities it names and which sources it cites. The monthly report adds answer share by subject, market and engine, the facts corrected, the pages and citations delivered, and next month's plan. Where your analytics allow, it shows which pages assisted enquiries and applications.

Illustrative example, not client data:

What we need from you

• Named approvers: a lead in admissions or the international office for facts, a marketing contact for pages, and compliance review for outcome and ranking claims. Nothing goes live without their sign-off.
• Access: CMS publishing rights or a web developer, plus Search Console and analytics.
• Your priorities: the courses and levels that matter most, the markets you recruit from, and the rival institutions you lose applicants to.
• Current facts: this cycle's fees, scholarships, entry requirements by country and visa guidance, with any outcome claim sourced to published data.
• Approvals within a few working days, so pages and corrections go live while each market's application window is open.

International recruitment, agents and markets

International applicants research in their own language, through engines that may favour local sources, and sometimes with an education agent who uses the same tools to draft a shortlist.

The same question asked from India, Nigeria or Vietnam can produce a different answer from one asked in London, because the engine reaches for different sources. That is why each market gets its own prompt set and its own line in the monthly report, rather than one global score.

Visa rules change, and engines can be slow to notice. When your pages state the current rules clearly and with dates, engines have a better source to quote than an old forum thread.

If you recruit mainly from one system, our pages for US universities and colleges and UK universities cover the Common App, UCAS, Clearing and the domestic sources in detail.