Getting Better Applications, Not Just More

More applications is usually the wrong goal

The reflex when a cohort looks thin is to widen the funnel: more listings, more posts, more outreach. It works, in the sense that the number goes up.

It also lowers the average, adds review load, and burns panel goodwill on applications that were never going to be admitted. A programme that received four hundred applications and admitted eight has not demonstrated selectivity; it has demonstrated that three hundred people misunderstood what it was for.

The goal is a funnel where a high proportion of applicants are plausible. Five to ten times your places, mostly relevant, beats fifty times, mostly not.

The three things applicants actually judge

Founders decide whether to apply quickly, on very little information:

1. Who else went through it. The single strongest signal. Not logos of partner organisations - names of companies that were in the programme and what happened to them.

2. What they specifically get. "Mentorship, networking and support" describes every programme ever run and therefore describes none. "Twelve weeks, a desk, a weekly session with an operator in your sector, and ₹X against milestones" is a decision someone can make.

3. Whether the process looks professional. Deadlines that are stated and kept. A form that works. A decision date that is honoured. This is entirely within your control and it is where most programmes lose credibility, because a founder who is treated casually during applications assumes the programme will be casual too.

Referral is the channel that works

Across almost every programme, the strongest applicants arrive through someone who already knows the programme.

Sources worth working, roughly in order of quality:

• Alumni founders. They know what it is like and their recommendation carries weight. Ask them directly, by name, each round - a general "please share" does nothing.
• Current cohort. Founders know other founders at their stage.
• Mentors. They see companies constantly and rarely think to refer unless asked.
• Other programmes. The ones you are not competing with - a different stage, a different geography, a different sector - refer usefully and reciprocally.
• Your institution, for university centres. Faculty know who is building something.

The mechanism matters as much as the list. A named ask - "we have eight places, we are looking for companies at this stage in these sectors, do you know two?" - produces referrals. A broadcast does not.

Say who you are not for

Counter-intuitive and highly effective.

A page that says "this programme is not right for you if you are pre-idea, if you need capital rather than support, or if you cannot commit to being on campus two days a week" does two useful things. It stops applications you would reject anyway, which saves everyone time. And it substantially raises the credibility of everything else you claim, because a programme willing to turn people away sounds like one that selects.

Publish the process and then keep to it

Before applications open, publish:

• The deadline
• When shortlisted applicants will hear
• Whether there is an interview, and roughly when
• The final decision date
• How many places there are

Then keep to all of it. If a date slips, tell everyone before it slips rather than after.

This is the cheapest reputational win available to an incubator, and it compounds: applicants talk to each other, and a programme known for running a clean process gets better referrals next round.

Reduce the cost of applying

Every unnecessary field costs you applicants, and it costs you the ones with options first - because they are the ones who can afford to abandon your form and finish somebody else's.

The application form article covers this in detail, but the short version: ask only what you will score, gate eligibility in the first three questions, allow save-and-resume, and show progress.

Reach out to specific companies

Most incubators broadcast. The programmes with the strongest cohorts also do direct outreach.

Build a list, each round, of ten to twenty companies you would genuinely like to have. Local, at the right stage, in sectors where you can actually help. Write to them individually - not a campaign, a short note saying why this programme in particular fits what they are doing.

The hit rate is low and the quality is high, and it is the only channel that reliably reaches founders who were not already looking for a programme.

Measure the funnel, by source

Record where each application came from - referral, direct outreach, listing, social, event - and carry that through to admission.

Within two rounds you will know which channels produce applicants who actually get admitted, which is a completely different ranking from which channels produce volume. Most programmes discover that the channel producing the most applications produces almost none of the cohort, and that a channel they nearly stopped is where half the admitted companies came from.

That single measurement changes where the outreach effort goes, and it is why source should be a field on the form rather than a guess afterwards.

What to do if the cohort is genuinely thin

Sometimes the funnel is small and the deadline is close. In order of preference:

1. Extend the deadline and use the time for direct outreach, not more broadcasting.
2. Admit fewer and run a smaller, better cohort. Filling places with companies you are unsure about damages the programme's reputation with the ones you were sure about.
3. Do not run this round. Rare, painful, and occasionally correct - particularly for a new centre, where the first cohort sets the reputation for years.

Filling a cohort with weak applicants to hit a number is the option that feels safest and costs the most, because the first cohort's outcomes are what the next round's applicants will judge you on.