Career guidance in schools: why it reaches the students who ask

Ask a school why careers guidance is hard and the answer is almost never that nobody cares. It is arithmetic. One counselor, several hundred students, and a decision each of those students makes roughly once, usually at the exact moment they know least about themselves.

So the work becomes triage. The students who ask get seen. The students who do not ask, who are often the ones with the least support at home, quietly do not. And the teachers who sit with those students every week have nowhere to put what they notice.

This is not one school having a bad year. In The State of Global Teenage Career Preparation (OECD, 2025), the students whose career thinking gives the greatest cause for concern are also the ones who take part in guidance activities least often. Guidance reaches whoever knocks on the door.

You can see the cost in what young people expect. Across the OECD, 50% of girls and 44% of boys expect to work in one of just ten jobs, and those expectations have barely moved since the year 2000 while the labour market has moved a great deal.

Europe has noticed. Cedefop published an EU reference framework for lifelong guidance in 2026, eighteen guidelines for national systems, and universal access is one of them. Universal access is a staffing question before it is a policy question, which is exactly where a tool can honestly help.

The useful thing a platform can do here is not to replace the conversation. It is to make sure the right conversation happens with the right student at the right time, and to hand the adult something to open with.

There is more agreement on this than a school would expect, and almost none of it is controversial. The British Gatsby benchmarks set out eight things a school should do, and they have become the reference model well beyond the UK. Cedefop published an EU reference framework for lifelong guidance in 2026 with eighteen guidelines for national systems. The OECD's work on teenage career readiness turns the same ideas into indicators that can be measured against later employment outcomes.

Strip them down and the same items keep appearing: guidance should be woven into ordinary teaching rather than parked in one afternoon a year; every student should get personal guidance, not only those who seek it out; students should meet people who actually do the work; information should be real labour-market information rather than a brochure; and the school should know whether any of it made a difference.

In Spain the model that carries this into centres is Xcelence, built by Fundación Bertelsmann on the Gatsby benchmarks and set out as ten claves, with a free self-assessment tool any centre can use. In Sweden the framing is Skolverket's: guidance is the responsibility of the whole school, not one person in one room.

Read those two sentences next to the staffing numbers above and the gap becomes obvious. Nobody disagrees that every student should get personal guidance. The disagreement, if there is one, is about how a school with one counselor for several hundred students is supposed to deliver it.

Career guidance is national in a way that most education software is not. A Swedish student in year nine is choosing between eighteen gymnasieprogram with meritvärde as the currency. Three years later the question is which higher education to apply for and what the entry threshold has been. A Spanish student at the end of 4º ESO is choosing between Bachillerato and FP, then between five modalidades, then facing ponderaciones and notas de corte that vary by university and by year.

The official data is national too, and most of it is free and good. Sweden publishes occupational outlook, salary by occupation, and how graduates of each degree fared in the labour market about a year after finishing. Spain publishes its education registers and catalogues across grado, máster and FP de grado superior, and European skills-demand data sits underneath both.

This is why translating a guidance product between countries does not work. The vocabulary is the smallest part of it. A system that does not know what a ponderación is cannot answer a Spanish student's real question, and one that does not know meritvärde cannot answer a Swedish one.

It is also why the honest answer matters more than the complete one. Admission figures move every year, and a single wrong nota de corte can send a student down the wrong path. Saying that a figure is not known, and giving the year of the one you do have, is worth more than appearing comprehensive.

Careers software sits on data about minors and their futures, so the questions worth asking are the same whoever the vendor is. Who can see an individual student. What leaves the building. Whether anything automated decides something about a child. Where the data lives. What happens to it commercially.

Our answers, for the record. A student is a named person only to the adults at their school who guide them, and a number to everyone else. Nothing about an individual student ever reaches an employer or a university. School-wide figures omit any group smaller than three, because a small count is a name in disguise. A student's conversations with the AI guide are private from every adult.

Nothing decides anything about a child: no score gates a path, no door closes on a student, and a person makes every judgement. All storage sits with providers inside the EU under GDPR, there are no ad trackers, and student data is never sold. The full list of subprocessors is in the privacy notice.

A school should hold any supplier to answers this specific, including us.