How Ivy League admissions actually works

Almost every complaint about selective admissions comes from assuming it is a ranking exercise. It is not. It is a selection exercise, and those are different problems. A ranking asks who is best. A selection asks which group of roughly two thousand people, taken together, the institution wants next year. Once you see which question is being answered, most of the apparent randomness stops being random.

Holistic review, described plainly

The eight Ivy League schools all practise what they call holistic review. In practice that means an application is read as a whole rather than scored on a formula, and it usually moves through something like this:

1. A regional reader. Applications are grouped by geography, so a file is normally first read by the officer responsible for that region. They know the schools in it: what a demanding course load looks like at that particular high school, which counsellors write carefully, what opportunities exist locally.
2. A second read. Most files get at least one more reader, partly for consistency and partly because the first read is fast.
3. Committee. Contested files go to a group, where the regional officer presents the applicant. This is the part that surprises people: someone has to argue for you, out loud, to colleagues who have their own candidates to argue for.
4. Shaping. Late in the cycle the committee is no longer only comparing individuals. It is assembling a class, and the remaining decisions are made with the shape of that whole class in view.

Each school does this slightly differently and none of them publish the internal mechanics in full. But the broad structure is consistent, and it is described in similar terms by admissions offices themselves and by the National Association for College Admission Counseling.

The context rule

The single most misunderstood part of the process is that applications are read in context. A reader is not asking whether you took the hardest courses that exist. They are asking whether you took the hardest courses available to you, at your school, given your circumstances.

This cuts both ways, and the second way is the one families rarely hear:

• A student from a school offering three AP courses is not penalised for having taken three.
• A student from a school offering twenty-five, whose family has resources, and who took eight, is not being compared to that first student on raw count. They are being asked why not more.

Context is not a discount applied to some applicants. It is the denominator applied to all of them.

What the pieces are actually for

The transcript is the single heaviest element almost everywhere, and it is read for trajectory and rigour, not just for the final number. Four years of demanding coursework with an upward trend reads differently from a flat, unchallenged record at the same average.

Testing varies by school and by year. Policies moved substantially after 2020, several Ivies have since reinstated a requirement, and others have not. This changes often enough that it should never be taken from an article; check the admissions page of the specific school for the specific cycle.

Essays are not a writing competition. They are the only place in the file where the applicant speaks in their own voice, and readers use them to answer a practical question: what would this person be like to have around for four years, at eight in the evening, in a corridor. That is why sincerity outperforms polish so reliably.

Recommendations are read closely, and read comparatively. A teacher who has taught for twenty years is implicitly ranking you against everyone they have taught. The phrase readers are looking for is some version of one of the best I have worked with, and it cannot be manufactured by the applicant.

Activities are assessed on depth, not length of list. Which brings us to the part worth its own section.

Depth beats breadth, and it is not close

The strongest activity records tend to look narrow from outside. One or two areas, pursued for years, arriving somewhere real: a business that took money, research that produced a result, a team that was actually led, a body of work that exists.

The reason is structural rather than aesthetic. A committee member has to be able to summarise you in one sentence to colleagues who have not read your file. A student who is the one who built the thing gets summarised. A student with eleven clubs and no centre of gravity is genuinely harder to advocate for, no matter how capable they are, because there is nothing for the advocate to say.

Institutional priorities are real, and they are not about you

Every year an admissions office arrives with needs that have nothing to do with any individual applicant: an orchestra missing a bassoonist, a department the university is investing in, athletic recruitment, geographic and socioeconomic spread, the children of alumni and donors at institutions where that still counts.

These priorities are why two students with functionally identical files can receive different answers, and why reading a rejection as a verdict on quality is a category error. It was a selection, and part of the selection had already been decided before the file was opened.

What this means in practice

• The process is legible, but it is not a formula, and anyone selling you the formula is selling something else.
• Doing well requires being unambiguously strong on academics and being easy to describe in one sentence.
• Much of the outcome is outside the applicant’s control, which is an argument for a genuinely broad list of schools rather than an argument for despair.

The most useful thing a family can internalise is that the rejection letter is not addressed to the student’s worth. It is the output of a committee assembling a class under constraints it never explains.