Build It In-House or Licence It: The Real Cost Per Seat

Every college that considers a certified skill programme asks the same question, and usually asks it last: what does this cost per student. Asked properly, it is the question that settles the whole design, so it should be asked first.

Build it yourself: the costs people count

Curriculum development. Faculty time. Some material. Maybe a lab.

Those are the visible costs and they are the small ones. A college with subject experts on staff can genuinely write a decent 60 hour AI or business analytics module without paying anyone extra. This is why in-house always looks cheap on the first pass.

Build it yourself: the costs people miss

The certificate has no external level, because your institution is not an NCVET recognised awarding body. So it does not sit at NSQF 4.5, it sits nowhere.

The credits do not enter the Academic Bank of Credits through a recognised awarding body route, so portability depends entirely on your own arrangements. Assessment is internal, which means the person who taught the course also decided it was passed, and every external reader knows it. Nobody outside your campus can verify any of it, which puts you back where this started, issuing documents employers discount.

And the recurring cost is not zero. Somebody updates the curriculum every year. Somebody sets and marks the assessment. Somebody defends it to the academic council. That is real staff time, every year, permanently.

In-house is not cheap. It is unpriced, which is different.

Licensing: what it looks like on a spreadsheet

The shape that works is a per student price against an enrolled cohort. Our bundled institute programme runs at ₹3,000 per student, and the reason it is structured that way rather than as a day rate is that a day rate rewards the vendor for showing up and a seat price rewards the vendor for the student finishing.

What sits inside that: licensed curriculum against a published qualification pack, faculty certification so your own staff deliver it, independent assessment, the certificate itself, and the documentation set for your accreditation file.

What sits outside it: your faculty's time, which you were paying anyway, and your classroom, which you already have.

The comparison run honestly

Take 300 students, three years.

The licensed route is roughly ₹9 lakh over three years, with the certification cost for faculty concentrated in year one and the per seat cost flat after that. The recurring work of curriculum updates, assessment and standards maintenance sits with the awarding body.

The in-house route has a low cash cost and a high staff cost that never ends, plus a certificate that carries no external level. The trap is that the staff cost never appears in a budget line, so year on year it looks free while quietly consuming the time of the people you most need teaching.

External trainers hired per batch are the worst of both. High recurring cash cost, no transfer of capability, and the programme stops the day the contract does.

Where the money actually comes from

This is the part principals raise immediately and vendors avoid. On the bundle model the course is folded into the fee structure at admission rather than sold to students separately. A student is not asked to make a second purchase decision in October about something they have never heard of.

That single design choice is the difference between 300 enrolments and 30. Optional paid add-ons in Indian colleges convert badly, and everyone who has run one knows it. Bundling is not a pricing trick, it is an acknowledgement that the decision maker is the institution and the beneficiary is the student, and those are two different people.

Questions to ask any vendor, including us

Is the price per seat or per day, and what happens if the batch is 180 instead of 300. Which qualification pack code, at what NSQF level, and is it inside the awarding body's recognised scope. Who assesses, and are they certified assessors independent of whoever teaches. Who registers learners on ABC and when. And in year three, can we run this without you.

That last question is the honest test of the model. If the answer is no, you did not buy a programme. You bought a dependency.

What we would say to a principal

Do not start with 300 students. Start with one programme and one cohort, insist the faculty certification happens in year one so the capability transfers, and hold the vendor to the documentation set rather than the certificate design.

If it works, the second cohort costs you less and needs us less. That is the arrangement working correctly, and any vendor uncomfortable with that sentence is worth a second look.