What NIRF Placement Data Actually Tells You About Jobs
The Graduation Outcomes score is not a placement rate, and it does not behave like one. Across 96 ranked engineering institutes it correlates with median salary at 0.81 but with placement rate at only 0.51. A high GO tells you graduates were paid well, not that most of them were placed.
Graduation Outcomes is the parameter everyone reaches for when they want a jobs signal out of NIRF, and it is the right instinct. But almost nobody knows what is inside it, and the contents make it behave in a way that surprises people.
What GO is actually made of
| Sub-metric | Marks (of 100) | What it counts | Share of the total NIRF score |
|---|---|---|---|
| GPH | 40 | Placement and higher studies, combined into one figure | 8% |
| GMS | 25 | Median salary of placed graduates | 5% |
| GPHD | 20 | PhD students graduated | 4% |
| GUE | 15 | Students graduating in the minimum stipulated time | 3% |
Four things, and only two of them touch employment. Note the last column: because GO carries 20% of the total NIRF score, placement and higher studies combined are worth 8% of an institute’s rank and median salary 5%. The full weighting is in how NIRF scores are calculated.
The trick is in GPH: placement and higher study are one number
GPH is the largest sub-metric at 40 marks, and it does not measure placement. It measures progression — students who were placed or went on to higher study, added together and treated as the same outcome.
That single design choice changes everything, because the two rates are far apart. Across the ranked institutes the median placement rate is 73.8%, while median progression is 91.4%.
| Institute | Placement rate | Higher study | Progression (what GPH counts) |
|---|---|---|---|
| Rajiv Gandhi Institute of Petroleum Tech | 52.2% | 47.8% | 100.0% |
| Aligarh Muslim University | 55.7% | 44.3% | 100.0% |
| Vignan University | 43.4% | 38.0% | 81.3% |
| Pandit Deendayal Energy University | 55.6% | 37.6% | 93.2% |
| IIT Tirupati | 55.6% | 35.1% | 90.7% |
| NIT Srinagar | 57.9% | 34.2% | 92.1% |
| Jawaharlal Nehru Technological University | 44.4% | 30.5% | 74.9% |
Rajiv Gandhi Institute of Petroleum Tech placed 52.2% of its cohort and reports 100.0% progression, because 47.8% went to graduate school. On GPH those two institutes are indistinguishable from one placing everybody.
This is not a flaw. A student going to a master’s programme has not failed, and a framework that counted them as unplaced would be worse. But it does mean GPH saturates: once most institutes cluster above 90% progression, the sub-metric stops separating them.
So what is GO really measuring?
If the biggest sub-metric has stopped discriminating, something else must be doing the work. It is median salary.
| Measure | Correlation with GO | Reading |
|---|---|---|
| Median salary | 0.81 | Strong. This is what GO mostly reflects. |
| Progression rate (placed + higher study) | 0.55 | Moderate, despite being the largest sub-metric. |
| Placement rate alone | 0.51 | Moderate. GO is not a placement rate. |
| Share going to higher study | -0.02 | None. GO does not care which of the two it was. |
Median salary correlates with the GO score at 0.81, far above progression at 0.55 or the placement rate at 0.51. GMS carries only 25 of the 100 marks, but salary ranges from about three lakh to thirty-two lakh across these institutes, while progression is squeezed into a narrow band near the top. The sub-metric with the most spread ends up driving the score.
Note the last row too. GO correlates with the higher-study share at -0.02, which is effectively zero. That is GPH working exactly as designed: it genuinely does not care which of the two outcomes a student had.
Where GO and the placement rate disagree
| Institute | GO score | Placement rate | Median salary |
|---|---|---|---|
| High GO, modest placement rate | |||
| IIT Kanpur | 88.08 | 66.5% | Rs 1,940,000 |
| IIT Ropar | 75.18 | 68.5% | Rs 1,697,000 |
| IIT Roorkee | 85.82 | 69.8% | Rs 1,700,000 |
| NIT Durgapur | 77.92 | 69.9% | Rs 1,124,000 |
| IIT Mandi | 78.29 | 70.0% | Rs 1,850,000 |
| Low GO, high placement rate | |||
| C.V. Raman Global University | 59.54 | 91.7% | Rs 645,000 |
| UPES | 54.26 | 91.0% | Rs 650,000 |
| SR University | 55.79 | 81.1% | Rs 600,000 |
| Panjab University | 56.80 | 78.2% | Rs 825,000 |
| Manipal University Jaipur | 58.63 | 68.5% | Rs 788,000 |
IIT Kanpur is the clean example of the first pattern: 66.5% placed, 25.0% to higher study, and a GO score of 88.08. The second pattern is the mirror image. C.V. Raman Global University placed 91.7% of its cohort — an excellent rate — and scores 59.54 on GO, because the median those placements paid was Rs 645,000.
The denominator is smaller than you think
One more definitional detail, and it is the one most likely to mislead. NIRF asks for students placed against students graduating in the minimum stipulated time — not against the cohort that was admitted four years earlier.
Those are different numbers. Across the ranked institutes, on-time graduates are a median of 94.2% of the admitted cohort, and as low as 75.7%.
| Institute | Admitted | Graduated on time | Share | Placed, of graduates | Placed, of admitted |
|---|---|---|---|---|---|
| Jadavpur University | 1,146 | 868 | 75.7% | 91.0% | 68.9% |
| IIT Madras | 929 | 714 | 76.9% | 76.9% | 59.1% |
| Panjab University | 809 | 625 | 77.3% | 78.2% | 60.4% |
| Aligarh Muslim University | 449 | 352 | 78.4% | 55.7% | 43.7% |
| Graphic Era University | 1,144 | 916 | 80.1% | 49.3% | 39.5% |
| SR University | 1,080 | 873 | 80.8% | 81.1% | 65.6% |
Anyone who takes an extra year to finish is out of the denominator. That is not the same as failing — most of them graduate later — but it does mean the published rate is calculated on a smaller base than the intake, and the effect is not trivial. IIT Patna placed 95.5% of its on-time graduates and 81.2% of the students it admitted.
NIRF is not hiding this: on-time graduation is scored separately as GUE, worth 15 marks. It just means the placement percentage you calculate from the sheet answers a narrower question than "what happened to the people who started here".
How to read it properly
- Do not read GO as a placement percentage. Pull the real counts from the institute’s NIRF data sheet and divide.
- Read placement rate and higher-study share together. A low placement rate at a research-heavy institute usually means graduate school, not unemployment.
- Treat GO as a salary proxy, because on this evidence that is mostly what it is. If pay is what you care about, it is the best single public signal.
- Check the median salary directly rather than inferring it. It is published per institute, and why the median matters more than the highest package covers how to read it.
- Mind the denominator. The rate is placed over on-time graduates, not over the admitted cohort.
- Remember it is self-submitted. NIRF runs consistency checks, not a full audit.
Jobs and study abroad
- If you want a job, GO plus the published median is the strongest public pair. The ownership comparison is in private vs government placements.
- If you are aiming at a master’s abroad, a high higher-study share is a positive signal rather than a warning: it means the institute routinely sends people to graduate school and the recommendation pipeline exists.
- If you are comparing India against abroad, the median salary is the input that decides it. Run it in the payback calculator or read study abroad vs India on ROI.
Frequently asked questions
Does NIRF publish placement percentage?
Not as a score. The ranking table publishes a Graduation Outcomes score, which is something else. But each institute's per-institute data sheet does publish the raw counts: students graduating, students placed, median salary and students going to higher study. The percentage is yours to calculate.
What is the GO score in NIRF?
Graduation Outcomes, marked out of 100 and weighted at 20% of the total. In engineering it is built from placement and higher studies combined (40 marks), median salary (25), PhDs graduated (20) and on-time graduation (15).
Is a high GO score the same as good placements?
Not quite. Across 96 ranked institutes the GO score correlates with median salary at 0.81 but with placement rate at only 0.51. A high GO more reliably means graduates were well paid than that a large share of them were placed.
Why does a college with 66% placement still score high on GO?
Because GPH counts placement and higher study together. IIT Kanpur placed 66.5% of its cohort and sent 25.0% to higher study, giving 91.5% progression, and it scores 88.08 on GO. The students who were not placed largely went to graduate school.
What is the denominator for NIRF placement figures?
Students graduating in the minimum stipulated time, not the cohort admitted four years earlier. Across ranked institutes the on-time graduates are a median of 94.2% of those admitted, so the published rate sits on a slightly smaller base than the intake.
Does NIRF verify placement data?
It runs consistency checks against submitted records rather than a full independent audit. The figures are self-submitted by institutes. That still makes them a declaration to the Ministry of Education, which is a stronger commitment than a brochure claim.
Sources and method
Sub-metric marks are from the official NIRF 2025 engineering framework document. Placement counts, median salaries and higher-study counts come from the per-institute data sheets NIRF publishes for every ranked institute, for example IIT Madras (IR-E-U-0456), downloaded 26 August 2026, four-year UG block, 2023-24 graduating cohort. 96 of the 100 ranked institutes parsed successfully; the rest are excluded rather than estimated. Correlations are Spearman rank correlations computed by us. Jobs to Education is not affiliated with NIRF or the Ministry of Education; see our terms.