How to read an MBA ranking

The table is the least important page

A ranking is an argument wearing the costume of a measurement. Some are careful arguments; many are not; and the table itself will never tell you which kind you are holding. The methodology page will. So the one genuinely wrong way to consume a ranking is the universal way: read the table first and the method never.

It helps to remember what a ranking is commercially. It is content that institutions cannot ignore, published by organisations that need attention, participation or both. That does not make every list corrupt — it makes every list a product, and products are designed around their buyer. The seven questions below establish who the buyer really is. Sometimes it is you. Frequently it is the institutions being ranked.

Seven questions to put to any ranking

1. Who supplied the data? Figures reported by the ranked institutions themselves, unaudited, reward the boldest form-filler rather than the best school. Look for the word “verified” and for who did the verifying.
2. How many answered? A salary average built from the minority of graduates who chose to reply is not an average; it is a self-portrait by the satisfied. No response rate, no credibility.
3. Are inputs dressed as outcomes? Faculty counts, library size and campus acreage measure spending. They may correlate with your outcome; they are not your outcome.
4. Does the publisher sell anything to the ranked? Badge licences, advertising, “profile enhancement”. Money flowing from the measured to the measurer deserves a permanently raised eyebrow.
5. Is the category coherent? A table mixing two-year residential programmes with online degrees is ordering things that do not compete for the same student or the same recruiter.
6. How much does it swing? Institutions change slowly. A school moving fifteen places in a year is evidence about the method’s noise, not about the school.
7. Is it ranking your question? A research-weighted list answers an academic’s question. If yours is teaching quality or hiring access, a rigorous answer to the wrong question misleads as surely as a sloppy one.

Fast tells of a weak one

• No methodology page at all — the disqualifier that ends the conversation.
• A single composite “score” whose components and weights are nowhere stated.
• Every participant somehow a winner of a sub-category of its own.
• Rank badges available for licensing on commercial terms alongside the list itself.

Using even a flawed ranking well

Demote it from verdict to discovery device. A ranking is a decent source of names you had not considered; it is a poor authority on their order. Cross-check any name across three years and at least two unrelated publishers, and treat agreement between independent methods as weak signal — a single table as none.

And remember what no ranking can carry: your constraints. Fee ceiling, city, mode, the clause in a promotion policy — the variables that actually decide your shortlist appear in no publisher’s weighting. A list can only rank the average applicant, and you are not the average applicant.

Our own instrument sidesteps the contest: we grade programme types against use-cases rather than putting named institutions in a queue — the method is public at how we grade, the results at the grades, and the companion at mba.mbaindia.org explains how the grades map onto a shortlist. Whether any of it should move you to enrol is a prior question — answered here.