Agents are not the problem. For most universities recruiting internationally, the agent channel is the only part of the funnel that converts reliably in cities where the university has no staff, no brand recall and no counselor relationships. Removing it rarely improves anything, and the institutions that have tried usually discover which parts of the job the commission was quietly covering.
So this is not an argument against agents. It is an argument about incentives, and about the gap between what a commission pays for and what a university still has to own. A per enrollment fee is an efficient way to buy a conversion. It is a poor way to buy a shortlist, and the shortlist is where intakes are won. If you want the operating model rather than the argument, our international marketing and admissions practice sets out how the two layers fit together.
Every rule below is dated and labeled: law, code of practice, or guidance. Where a change is still a draft, it says pending. Where the only available source is sector reporting rather than a government page, it says so.
What an agent commission actually buys
Four things, and they are worth real money.
- Reach into a market you cannot staff. An agent network already sits in tier two and tier three cities, in school corridors, in WhatsApp groups and at local fairs. Replicating that presence means hiring, travel, local entities and years of relationship building, paid in advance with no guarantee of enrollments.
- A counselor relationship the family already trusts. In many markets the decision is made in a room with a counselor the family has known for years, sometimes through an older sibling. That trust was earned by someone else and it is not transferable by advertising.
- Document handling. Transcripts, English scores, funding evidence, gaps in study, sponsor letters. An agent who knows your requirements sends a file that is complete and checkable, which shortens your admissions cycle and lowers your refusal risk. Admissions teams feel this immediately when an agent is replaced by direct applications.
- A conversion you did not originate. The commission is paid on a result, after the result, which is why finance teams like it. There is no media budget to defend and no attribution argument to win.
Say it plainly: a well managed agent relationship is good value for those four things. Most universities that think they have an agent problem actually have a contract problem, a measurement problem, or a missing demand layer. Those are different problems with different fixes.
Where the incentives part company
The structural issue is simple and it has nothing to do with dishonesty. The agent is paid per enrollment, and paid by many universities at once. A counselor sitting with one family is therefore holding a portfolio of seats to fill. The question that portfolio makes natural is which of these can I place this student into, and the question the family needs answered is which of these is right for this student. Most of the time those two questions have the same answer. When they do not, the incentive points one way and nobody is paid to point the other way.
This is not a theory the sector disputes. Its own codes name the failure mode. The UK National Code of Ethical Practice for UK Education Agents, in its July 2025 version, tells agents not to seek or facilitate students moving between institutions to courses at the same or a lower level in order to gain additional commission, and to introduce a range of possible providers aligned to the student's profile. Australia went further and legislated, which we cover below. Both of those exist because the mechanism is real.
What it costs the university, in order of visibility
- Yield. An applicant recruited on the strength of an offer rather than a fit is an applicant who is still shopping. The offer converts to a deposit late, or the deposit does not convert to an enrollment, and the intake number moves in the last four weeks when you can do nothing about it.
- Withdrawal and transfer. A student placed on a course that was available rather than wanted is more likely to leave in the first terms. That cost lands on a different budget line from recruitment, usually a year later, which is exactly why it rarely gets connected back to the channel that produced it.
- Students on the wrong course. The quietest cost and the most damaging. They struggle, they do not become advocates, and they tell a market you spent years building that your university was not what they were told. Reputation in a source market is largely word of mouth from the last three cohorts.
None of this is fixed by finding better agents. It is reduced by changing what the commission rewards, and by making sure the university is not dependent on the agent for the part of the decision that happens before anyone is contacted.
What the commission does not buy
By the time a family walks into a counselor's office, they usually already have three to five names. Those names came from somewhere: a search result, an AI assistant's answer, a Reddit or Quora thread, a YouTube video from a student a year ahead of them, a cousin in Canada, a course page that either answered their question or did not. That is the shortlist layer, and a commission structure cannot fund it, because a commission is only paid when demand already exists.
The layer a commission does not reach
Concretely, the commission does not buy:
- Presence in search results and AI answers. Families ask assistants direct questions now, in their own words, about entry requirements, funding, work rights and whether a course is worth it for their profile. Whether your university appears in that answer is decided by your own pages and by what third parties say about you. Our guide to AEO for higher education covers how those answers get assembled.
- Course pages that answer real questions. Most course pages describe modules. Families want to know who gets in, what it costs in total, what happens after, and what the alternative is. A page that answers those is a page an assistant can quote and a counselor can forward.
- Forums and third party mentions. The threads a family reads are not on your domain, and no commission covers showing up in them accurately.
- Alumni and outcome content. The single most persuasive asset in international recruitment is a recent graduate from the same city and the same course saying what actually happened. Nobody is paid per enrollment to produce it.
- Your own data. When the agent owns first contact, you learn what converted but not what was considered, which city the interest came from, or which question stopped the family. That is the input you need to plan the next intake, and it is not in the commission.
Put the two together and the picture is clear. The agent channel closes demand. Something has to create it, and that something can only be the university.
The 2026 rulebook, and which parts are actually law
This matters because a lot of agent governance conversation treats a code of practice as if it were legislation, and treats legislation as if it were optional. Verified as of September 27, 2026.
| Market | Instrument | What it is | Status and date | Source |
|---|---|---|---|---|
| United Kingdom | Student sponsor guidance, Document 2: Sponsorship duties | Immigration guidance, enforced through the sponsor licence rather than by statute | Updated version reported April 7, 2026; in force. Sponsors must record agent details on the CAS where an agent was used in recruiting that student, and retain evidence of managing agents in line with the AQF and the national code | GOV.UK, ICEF Monitor, British Council |
| United Kingdom | The National Code of Ethical Practice for UK Education Agents | A sector code of practice, voluntary in itself, now referenced by sponsor duties and embedded in provider contracts | July 2025 version. Five standards, including ethical business practice and objective advice | British Council |
| United Kingdom | Agent Quality Framework | Standards, training and resources. It states that it does not provide accreditation, licensing or ongoing assurance of compliance | Established 2022, led by the British Council with BUILA, UKCISA and Universities UK International | AQF, BUILA, UKCISA |
| UK, Australia, Ireland, New Zealand | London Statement of Principles for the Ethical Recruitment of International Students by Education Agents | An intergovernmental statement of principles, not law and not enforceable against an agent | Agreed in London in March 2012. Named as an origin of the UK national code | British Council |
| Australia | ESOS Act 2000 and the National Code of Practice 2018, Standard 4 | Law. Providers must have a written agreement with each agent they engage, must ensure agents act ethically, honestly and in students' best interests, and must act where an agent does not comply | In force | Department of Education, ASQA |
| Australia | Ban on commission for onshore transfers | Law. Registered providers are banned from paying commission for onshore student transfers. Initial enrollments and progression within a packaged course already on the visa are unaffected | Applies to transfers occurring after March 31, 2026; does not apply to students accepted for enrollment at the new provider on or before that date | Study Australia |
| Australia | Education Legislation Amendment (Integrity and Other Measures) Bill 2025 | Legislation amending the ESOS framework, sharpening the definitions of education agent and of agent commission to cover any benefit, monetary or otherwise | Reported as passed on November 28, 2025. Sector reporting, not a government page we could verify directly | ICEF Monitor |
| United States | Higher Education Act section 487(a)(20), 20 U.S.C. 1094(a)(20) | Law. Bans commission, bonus or other incentive payment based on success in securing enrollments or financial aid, except for the recruitment of foreign students residing in foreign countries who are not eligible to receive federal student assistance | In force. The exception is statutory, not a later interpretation | Cornell LII, Department of Education |
| United States | NACAC Guide to Ethical Practice in College Admission | Association recommendations, not a mandatory code. Institutions may use commissioned agents for students residing outside the US who are ineligible for federal aid, and are asked to disclose that agents are used | December 2023 version. Approved as recommendations by the 2020 Assembly after the Department of Justice identified antitrust concerns in the previous code in 2017 | NACAC, NACAC |
| India | Consumer Protection Act, 2019 and the CCPA coaching advertising guidelines | Subordinate guidelines on misleading advertising. Coaching is defined to exclude counseling, so overseas education consultancies are not squarely covered | Guidelines in effect November 13, 2024 | Press Information Bureau |
| India | Draft Overseas Mobility (Facilitation and Welfare) Bill 2025 | Pending. A draft to replace the Emigration Act 1983 and regulate recruitment intermediaries. The published draft addresses overseas employment and does not regulate education agents placing students | Draft published for comment, deadline November 7, 2025. Not enacted | Ministry of External Affairs |
| India | AAERI | Industry self-regulation, not a regulator. A code of conduct and code of ethical practice, and a cap on what members may charge a student in processing fees | Formed October 1996, registered under the Societies Registration Act | AAERI |
Three things follow from that table. First, there is no single agent regulator anywhere, and in India there is no dedicated national licensing or registration regime for agents placing students at foreign universities that we could source. Second, the binding obligations sit on the university, not on the agent: the UK sponsor duty, the Australian written agreement, the US statute are all things you can be held to. Third, the only market that has legislated against a specific commission incentive is Australia, and it did so by removing the payment rather than by writing a better code.
One correction worth making, because it circulates: the US allowance for paying international recruiters per enrollment is not a 2013 legal reinterpretation. The exception is written into the statute. What happened in 2013 is that NACAC's Assembly voted to change its own ethical standards to permit members to use commissioned agents outside the United States, which was reported at the time by ICEF Monitor and others. That was an association decision, not a change in law.
What to keep, what to renegotiate, what to build
The useful version of this decision is not agents or no agents. It is layer by layer.
Three buckets
| Layer | Decision | Why | What good looks like |
|---|---|---|---|
| Reach in cities you cannot staff | Keep with agents | Presence and relationships take years and fixed cost to build, and the commission is paid on a result | A named list of agents per city, reviewed each intake on enrolled students rather than applications |
| Counselor relationship and family trust | Keep with agents | Trust is personal and not transferable by media spend | Same day briefings when a rule changes, a link rather than a screenshot, and one dated source of truth |
| Document handling and file completeness | Keep, with an audit trail | It shortens your cycle and lowers refusal risk | A record of who submitted what, so a pattern in one agency is visible before it becomes a compliance event |
| Commission terms | Renegotiate | The fee currently rewards an enrollment, not a student who stays | Payment structured around the enrolled and retained student, not the offer or the deposit |
| Course level and transfer behavior | Renegotiate | The UK national code and Australian law both target moves made for commission | Written terms that pay nothing for a move to the same or a lower level, mirroring the sector codes |
| Attribution and first contact data | Build in house | If the agent owns first contact you never learn what was considered or what stopped a family | Enquiry source, city and course captured on your own properties, joined to deposit and enrollment |
| Search results, AI answers and course pages | Build in house | No commission funds demand creation, because commission is paid on demand that already exists | Course and entry requirement pages written as direct answers, dated, with an owner |
| Market pages and funding clarity | Build in house | One international page cannot answer five markets' questions and five sets of proof | A page per source market with total cost, scholarship criteria and funding evidence stated up front |
| Alumni and outcome content | Build in house | It is your most persuasive asset and nobody is paid per enrollment to make it | Recent graduates by city and course, on video, saying what actually happened |
The sequencing matters more than the list. Renegotiating terms takes one contract cycle. Building the demand layer takes two quarters before it shows up in an intake, which means the work that decides your 2028 numbers is the work you start now. Our higher education practice covers both halves, and the country by country policy picture is in the 2027 intake reset.
What a partner may say, and what only the university may say
Once you accept that a third party speaks to families on your behalf, one line matters more than any other clause in the contract: a partner may repeat what the university has published, and may never create a new fact.
A partner may state published entry requirements, fees and funding as listed, course structure and duration, visa rules with the official link, and graduate outcomes the university itself has published. A partner may never promise admission, promise a visa outcome, invent a scholarship or a discount, guarantee a job or a salary, or paraphrase entry criteria into something softer. The UK national code says the same in its own language, requiring agents not to disseminate false, incomplete or misleading information and not to claim a government endorsement or privileged relationship that does not exist.
Two practical points. First, the university has to publish the facts before anyone can repeat them, which means a dated page with a named owner, not a PDF from the last cycle. Second, transparency about commercial relationships belongs to the family: the UK code expects agents to be open about which services fall under institution pays and which under student pays, while noting that the commission amount itself is treated as commercially sensitive and is not expected to be disclosed. That is worth knowing before you promise a family full transparency you cannot deliver.
There is more to this than fits here, including who signs off on claims and what an escalation path looks like when a partner gets it wrong. We are writing that up separately as a rulebook for anyone speaking on a university's behalf.
Where to start
Pick one intake and one source market. Write down what you paid for, what you received, and what you still had to do yourself. Then check three things: whether the families who enrolled could have found you before they met an agent, whether your published pages are current enough for a partner to repeat, and whether your commission terms reward an enrollment or a graduate.
If that audit is useful, we run it with universities as a working session rather than a pitch. Book a university strategy session and we will look at your agent mix, your shortlist layer and the gap between them, using your own intake data.
Not legal advice
This post summarizes law, codes of practice and guidance as of September 27, 2026, with each source linked and dated. It is not legal advice and it is not immigration advice. Agent rules, sponsor duties and commission restrictions change, and several of the instruments above are reviewed annually. Before you sign a contract or change a policy, read the primary source and take your own legal advice.
