A few months ago, I sat in a meeting with a state innovation department where an official mentioned, almost in passing, that a manufacturing unit we’d both been tracking had received a capital subsidy from the state and, weeks later, a nearly identical benefit from a central scheme for the same machinery. Nobody had done anything wrong. Two departments had simply processed two applications without either one knowing the other existed. I’ve been in this ecosystem long enough that I should have stopped being surprised by this kind of thing. I wasn’t.
That conversation stayed with me, partly because of what I do for a living. At SanchiConnect, we’ve spent the better part of the last four years building India’s DeepTech enablement network, connecting founders, investors, corporates, universities, mentors, and government bodies. A good part of how we hold that ecosystem together day to day runs through SanchiAPP, the platform our team built to manage these programmes, and I’ve watched, over several years of running it, how much of the friction in any multi stakeholder programme comes down to one thing: different parts of the system not being able to see what other parts already know. So when I look at the MSME scheme duplication problem, it doesn’t feel unfamiliar. It feels like a larger version of something I already spend most of my working life on. The MSME scheme landscape has the same underlying pattern, just at a much bigger scale and with public money involved.
An abundance problem, not a scarcity problem
Every state runs its own set of MSME schemes. Subsidised loans, capital investment support, interest subvention, technology upgradation grants, marketing assistance, cluster development funds. Add the central schemes on top, Credit Guarantee, PMEGP, CGTMSE, and a small manufacturer in Rajasthan or Maharashtra can, in principle, qualify for half a dozen benefits at once.
Also read: India’s Series A Bottleneck, The DeepTech Funding Gap
What strikes me every time I look closely at this is that the problem isn’t a shortage of support. It’s abundance without shared memory. No single scheme administrator has visibility into what other schemes have already given, through no fault of their own, since the systems simply weren’t built to talk to each other. That produces two outcomes that look opposite but share the same root cause: some units end up receiving support twice for the same asset, while thousands of genuinely eligible units never discover the scheme meant for them in the first place. I’ve seen both sides of this, sitting across the table from state governments on one end and from thousands of founders and MSME owners in our own network on the other. Well networked units tend to find their way to every scheme available. Others, often the ones who need it most, simply don’t know where to look, and that’s a gap in awareness and infrastructure, not a reflection on anyone’s effort.
Why this happens, and why it isn’t anyone’s fault
Each scheme was built as its own product, with its own portal, eligibility engine, disbursal workflow, and MIS. That made sense at the time. Schemes were designed one at a time, by different departments, often years apart, using whatever systems were available. Nobody sat down and asked how this would all fit together later, because “later” wasn’t the job of any single department, and I say that with real respect for how much these teams manage to get done within that constraint.
Udyam registration helped. It gave every MSME a single identity number that most schemes now ask for. But registration tells you a unit exists. It doesn’t tell you what that unit has already received, from whom, for what, or against which asset. Verification, where it happens, is still mostly manual: an official cross checking PAN or Udyam numbers across spreadsheets, or relying on a self declaration. That works reasonably well at the scale of a few hundred applications. It naturally becomes harder to sustain at the scale of tens of thousands, which is part of why overlap tends to surface in audit reports rather than being caught while it’s still preventable.
What I mean by a unified ledger
I want to be precise about this, because it’s easy to hear “ledger” and imagine something far more disruptive than what I’m actually proposing. This isn’t about merging schemes or forcing every state into identical eligibility criteria. Those decisions should rightly stay with the departments and legislatures that own them.
What I’m describing is a shared, append only record, anchored on the Udyam number, that captures every MSME benefit disbursed anywhere in the country: which scheme, which authority, what amount or instrument, against what asset or purpose, and when. Think of it as infrastructure sitting underneath existing schemes, the way GSTN sits underneath state and central tax administration without dictating tax policy, or the way the account aggregator framework lets a lender see a borrower’s financial footprint across institutions without owning any of the underlying accounts. Before sanctioning a benefit, any scheme administrator could check the ledger and see the applicant’s history across the country, verified rather than self declared.
I find myself thinking about this in fairly concrete terms, because it echoes a problem we’ve had to work through at a smaller scale while running SanchiAPP across dozens of programmes. The pattern repeats itself: a central body needs oversight, individual states or partners need to keep autonomy over how they run their own schemes, and reporting that used to be stitched together by hand at the end of every quarter needs to happen automatically instead. Solving that at the programme level taught me that centralised visibility and distributed ownership aren’t actually in tension, they just need the right architecture underneath them. A national MSME ledger is that same idea at a much bigger scale.
For the official sanctioning a scheme, the change is at the moment of decision. A ledger check becomes a standard part of due diligence, similar to how a credit bureau check is now routine before a loan is sanctioned. That alone could meaningfully reduce the double disbursal I keep encountering, concentrated in the highest value categories like capital subsidy and interest subvention.
For the entrepreneur, I think this is the most interesting shift, and it’s the part closest to what drew me into building SanchiConnect in the first place. Today, a founder or MSME owner has to discover, research, and apply to each scheme separately, usually relying on a consultant or a well networked industry association to even know what exists. I meet founders every week, through our accelerator programs and our government advisory work, who found out about a relevant benefit months after it would have actually helped them, purely by word of mouth. A ledger that records what a unit has already availed can be flipped around to show what it hasn’t. A unit that took a technology upgradation grant but never claimed the marketing assistance benefit it clearly qualifies for is straightforward to flag once the data lives in one place. It’s genuinely difficult to flag today, not because anyone is falling short, but because the information simply isn’t connected. That is exactly the kind of invisible exclusion I think about often, having watched the same pattern play out inside individual programmes before we found ways to surface it.
For policymakers, it produces something that currently doesn’t exist in usable form: a real time, unit level view of whether a scheme is reaching new beneficiaries or reaching the same few hundred applicants repeatedly. Today that requires a special audit exercise. With a ledger, it becomes a standing dashboard, and scheme redesign can be grounded in where the actual gaps are rather than in anecdote.
The parts I won’t pretend are easy
I’d be doing this conversation a disservice if I presented it as simple.
Data ownership comes first. States are understandably protective of their scheme data, and that instinct is a reasonable one. A shared ledger raises real questions about who controls it and who can query it. GSTN and the account aggregator framework both show this is solvable, but it needs an explicit federal agreement built with states as partners, not a mandate handed down to them. Without that buy in, participation will likely be partial, and the ledger would end up as just another dataset with gaps, through no fault of the states themselves.
Definitions come second. “Capital subsidy,” “interest subvention,” and “grant” don’t mean quite the same thing in every state scheme. Logged as free text, a ledger becomes almost as opaque as no ledger at all. Somebody has to do the unglamorous but important work of building a disciplined taxonomy of benefit types before any of this becomes genuinely useful, and that’s detailed, patient work that deserves real credit when it happens.
Incentives come third, and this is the one I think about most. Departments understandably track their own success partly through disbursal numbers, and it’s worth acknowledging that a system revealing overlap could feel, at first glance, like it’s questioning past performance rather than helping going forward. I don’t think that’s the right way to look at it, and I’d want any rollout of this to be framed clearly and respectfully: as a way to protect a department’s own budget from unintended leakage and free up funds for genuinely new beneficiaries, not as an audit aimed at anyone. Getting that framing right, and getting it right early, matters as much as the technology itself.
Where I’d actually start
I don’t think this needs to be a big bang national rollout, and I’d gently push back on anyone proposing that it should be. I’d start with a small number of high value, high overlap categories, capital investment subsidy and interest subvention, where duplication is both most common and most costly, and build the ledger for those first, anchored on Udyam, with a handful of willing states as a pilot. If it measurably reduces overlap and surfaces genuinely eligible beneficiaries in that narrow band, the case for extending it scheme by scheme should make itself on the numbers, without anyone needing to argue for it too hard.
India’s MSME ecosystem isn’t short on schemes. If anything, we have the opposite situation: abundance without a shared memory of who has received what, which quietly produces overlap for some and invisibility for others. A ledger like this won’t fix scheme design, targeting, or disbursal delays, and it isn’t meant to. What it fixes is more basic than that. It lets every part of the system see what every other part has already done. I keep coming back to this idea partly because I’ve watched a version of it work at a smaller scale, across the programmes we’ve helped run over the years, and it’s changed how confidently I believe something similar could work here too. A unified ledger feels like the natural next size of a problem I already spend a lot of my time thinking about, and it’s one I’d be glad to help build alongside the states and departments who’d be doing the harder, more important part of the work.
The article has been written by Dr. Sunil Shekhawat, co-founder and CEO of SanchiConnect
