Notes From a Science Park: What an Incubator Director Actually Worries About
I sat down with the deputy director of Khorasan's Science and Technology Park, the regional body responsible for turning early-stage tech ideas into actual businesses. I went in expecting a conversation about funding and facilities. What I got instead was closer to a case study on the gap between what an institution is designed to do and what it actually has the resources to do.
What the park is actually for
The stated mission is narrower than "support innovation." A founder can walk in with almost nothing, an idea, no business background, and the park will run a basic feasibility assessment on it. What it won't do is accept an idea with zero prior thought behind it. Roughly one in three applicants gets in. The rest are turned away for the same underlying reason: no real understanding of whether anyone would actually buy the thing.
That filter matters more than it sounds. His framing was blunt: a lot of technical founders assume that building the product is the hard part and selling it is an afterthought. The park's job, as he described it, is catching that assumption before it costs someone years.
The two traits that actually predict success
Asked what he screens for beyond the feasibility study, he named two things, and ranked the second one above the first. Relevant academic background, business, entrepreneurship, engineering, helps, but it's not close to decisive; most of his own team came from engineering or humanities, not business degrees. What actually predicts whether someone survives the program is something closer to appetite for self-directed growth over a five-to-ten-year horizon, someone who can develop themselves without being told to, because a resource-constrained public institution can't hold anyone's hand.
The real product isn't the building
He was direct that the park's actual value proposition today isn't office space or funding, it's density. Companies inside the park borrow equipment from each other, hire interns from the same talent pool, find investors through the same informal network, and refer work to each other without the park doing anything. A toy company sourcing a CNC machine from a neighbor in the same building, cheaper and faster than sourcing it externally, because proximity made the relationship possible in the first place. He specifically preferred a single-floor layout over a tower for this reason, people run into each other, ask what the other person is working on, and problems get solved as a byproduct of physical proximity, not because anyone scheduled it.
He was candid about the limits of that model too. Companies here, he admitted, never got a formal roadmap, nothing like a university curriculum that spells out what's expected of a company in its pre-growth phase versus its first or second year, and what support it gets in return. Resource constraints meant he never felt able to commit to something that explicit. His fix, going forward: a clear contract signed at intake, spelling out exactly what a company gives and gets, rather than leaving that relationship implicit and renegotiated informally as it goes. A related conviction he stated plainly: direct financial subsidies to early-stage companies are usually a mistake, in his view, outside a narrow set of well-defined facility-type support, money handed over too early tends to slow a young company down rather than speed it up.
Where the ecosystem is thin
Several structural gaps stood out. First, there's no formal broker function, someone whose job is to sit between companies and match unsolved technical problems with the right outside expertise. He noted this exists in more mature startup ecosystems and is almost entirely absent here. A related, more specific skill he named as scarce: knowing how to actually work within and leverage existing regulations to a company's advantage, a distinct form of expertise from brokering, and one he said the region doesn't have enough of either.
He also raised a cultural gap I hadn't expected. Outside Tehran, he said, there's a stronger local instinct toward isolated, private work, something close to an invisible wall against collaboration. He drew the line specifically by field: people with more artistic or creative temperaments tend to organize sessions and share with each other; technical and engineering people, in his experience, do it far less. That undercuts the density argument in an interesting way, put people in the same building and collaboration doesn't just happen on its own; it depends on who you've put there and what habits they walked in with.
Asked what's actually missing when a company brings him a problem, his sharpest example wasn't technical at all. He described an owner who'd had to lay off staff he'd personally trained at real cost, and simply didn't know how to make that decision or handle it well. Hiring under pressure, motivating a difficult employee, structuring incentives, knowing when and how to let someone go, these are the questions that actually show up, far more than pure engineering problems.
A gap I raised myself
The second gap is one I raised with him directly, from personal experience. I told him about my own time running a startup inside the park, where I ran into problems that felt enormous to me at the time but were almost certainly routine to anyone with more experience, the kind of thing a five-minute conversation with the right person would have resolved, if that conversation had ever had a reason to happen. What was missing, I said, was something close to a peer case-discussion format: a company with a live, unresolved problem presents it to a room of people who've dealt with something similar, gets real-time feedback, and moves on. I mentioned that Harvard Business School students run something structurally similar, and that I was raising it partly because I'd personally felt the absence of it.
He agreed immediately and said the park has nothing like it, and should. What struck him about the idea was the specific failure mode it solves, a coordination problem, really: someone in the same building solved the exact issue a founder is facing right now, a year ago, and there's no mechanism for the two of them to ever find each other. Given what he'd just said about people-management being the real recurring problem, not technical skill, the fit was almost exact, this is precisely the kind of problem a room of experienced peers could resolve in an afternoon that no engineer or broker is positioned to touch.
On being a public institution
He was unusually candid about the tension of running an innovation-support function inside a government structure. Management of these parks is legally required to come from academic faculty, which he considers a real constraint, since people with deep entrepreneurial or business-community experience are effectively locked out of leadership roles by law, not by choice. His own advice to anyone weighing public versus private sector work in this space was equally direct: if the goal is actually solving problems, go private, the public sector hits structural walls that experience alone doesn't get you past.
What I took from it
The idea I'd raised, that peer case-discussion, the same mechanic this site is built around, might have a place inside an incubator, came out of a real gap I'd felt myself, not something I expected to land as well as it did. But the bigger thing I took away wasn't that specific fix. It was his framing of what an incubator's real product actually is once you strip away the funding and the branding: whether it manufactures the conditions for peer problem-solving to happen at all. Density, informal referral networks, physical proximity that turns into collaboration without anyone scheduling it, and, just as important, an honest reckoning with where that model quietly breaks down. Put the wrong mix of people in a building and density alone won't produce collaboration. Hand a young company money too early and it can slow them down instead of helping. Everything else, the office space, the subsidies, the mandate, is infrastructure sitting on top of one real function. An institution can have all of it and still fail at its actual job if there's no mechanism for one founder's hard-won answer to reach the next founder asking the same question.