Nikhil Kamath called MBA aspirants at 25 'idiots' and it sparked a nationwide debate. Here's what the actual placement and ROI data says.

In November 2025, Zerodha co-founder Nikhil Kamath said something that every CAT aspirant's group chat is probably still arguing about: that doing an MBA at 25 makes you "an idiot." Months later, he was still being asked about it — including at a Columbia Business School event, where he pointed out that the 500 people in the room had collectively paid $90 million to be there (Inshorts).

In a separate March 2025 discussion, a Reddit post arguing that skill-based hiring is quietly making degrees irrelevant went viral, with the line "even an IIM tag won't save you" — sparking real debate about whether an MBA, even from a top school, still guarantees anything (BusinessToday).

If you're prepping for CAT right now, this isn't just internet noise. It's a fair question. So let's actually answer it with numbers instead of hot takes.

What the ROI data actually shows

The honest answer is: it depends on the programme, your goals and the full cost — and the gap between outcomes is enormous.

At the top end, published outcomes can be strong. IIM Ahmedabad’s audited MBA-PGP 2025 report gives a median of ₹34.59 LPA and a mean of ₹35.50 LPA, measured as Maximum Earning Potential—not take-home pay or conventional CTC. IIM Bangalore’s PGP and PGPBA 2023–25 report lists a median salary of ₹32.61 LPA and a mean of ₹34.88 LPA. These are historical batch figures, not promises to individual students. (IIM Ahmedabad; IIM Bangalore)

Costs differ sharply, too. FMS Delhi’s MBA 2026–28 bulletin lists a semester fee of approximately ₹60,818; four semesters would total about ₹2.43 lakh at that rate, before living costs and other charges. That is not comparable to a ₹20–27 lakh tuition bill. A programme charging ₹15–20 lakh with typical offers of ₹6–8 LPA deserves careful scrutiny, but neither a “tier” label nor absence from NIRF alone proves its value. Check the awarding institution, applicable recognition, programme-specific placement record and the roles graduates actually enter. (FMS information bulletin)

Even premium, non-IIM options are not automatically exempt from this math. Someone already earning well must compare the realistic post-MBA salary increase with fees, loan interest, living costs and earnings forgone while studying. A large reported package does not establish a one- or two-year payback. Network access and a career switch may matter, but those benefits need a personal case—not a blanket ROI promise.

Why “which college” is partly a percentile problem

This is where the debate becomes directly relevant to anyone preparing for CAT. A strong score can improve your options, but admission is not decided by percentile alone. Academic history, category, work experience, interviews and other criteria can affect the outcome. Read each school’s current policy rather than treating one percentile as an admission guarantee. (IIM Bangalore admissions overview)

Different IIMs and B-schools use different shortlisting and selection criteria. The score–percentile relationship changes by year, too. For historical context, Cracku reports roughly 84.8 scaled marks for 99 percentile in CAT 2025, while iQuanta reports 111.4 for 99.90. These are third-party estimates, not official CAT 2026 targets. The difference is about 26.6 marks—not single digits—and combining two sources does not create one authoritative conversion table. (Cracku; iQuanta)

This is also why attempting more questions does not reliably improve your score. There is no universal “36–40 attempts” formula for 99+ percentile: correct answers, wrong-MCQ penalties, slot scaling and the year’s score distribution all matter. Find out whether you are losing marks through accuracy, selection, coverage or time before changing your attempt budget. Our earlier article explains the attempts–accuracy tradeoff.

The real takeaway

Kamath’s comment made headlines because it is provocative, not because it settles the decision. The useful question was never simply “MBA or no MBA?” It is: which programme, for which goal, at what full cost—and what evidence supports the likely outcome?

That is the preparation gap Marg aims to help with. Instead of jumping to “study harder,” it helps you examine actual mock decisions and question attempts, form a working diagnosis, then test it through focused practice. A diagnosis needs evidence over time; Marg cannot guarantee an admission, percentile or salary. If you are deciding whether an MBA is worth it, begin with realistic college options, published programme data and your own career goal. You can start exploring your preparation with Marg for free.