Why India's Schemes Don't Reach the Businesses They're Built For

Most businesses that qualify for government funding never apply for it.

India — Central and State governments combined — runs one of the largest public support ecosystems for private enterprise in the world. Grants. Tax holidays. Subsidised credit. Production incentives. Loan guarantees. Together they represent a meaningful commitment to startups and MSMEs that did not exist twenty years ago.

And yet most eligible businesses never claim what they are entitled to. Of 150,000+ DPIIT-registered startups, only a few thousand have received Section 80-IAC tax holiday certification — a benefit nearly all of them are technically eligible to apply for. State subsidies are claimed by a fraction of qualifying small manufacturers. Founders who could file simply don't.

This piece is about why.

Why the schemes exist

Private capital in India is concentrated — by geography and by sector. Venture capital funds roughly 1,500 startups a year, almost all in five or six cities, almost all in tech-adjacent sectors. Bank credit is collateral-driven; a software startup with thirty engineers and one customer has nothing a bank can lend against.

Government schemes are designed to fill that gap. Not by replacing private capital, but by absorbing the risk private capital won't — the seed-stage equity gap (SISFS), the tax burden in the early-profit years (Section 80-IAC), the micro-credit gap (MUDRA), the cost of building a domestic manufacturing base (PLI), and a long list of state-specific gaps.

A few of these are pure transfers: grants the business does not repay. Others are tax positions: deductions, holidays, credits. Others are credit instruments: subsidised loans, guarantees, interest rebates. The mix matters, because each addresses a different failure in the market.

If the system worked the way it was designed to, a founder would incorporate a company, within weeks know which schemes they qualify for, file digitally, and receive capital in weeks not quarters. The government would track outcomes — jobs created, output produced, exports generated — and retire schemes that don't work.

That is not the system we have.

The Central scaffolding

Central schemes form the spine. They tend to be larger in value, more uniform in eligibility, and better known in absolute terms. SISFS for seed-stage equity, Section 80-IAC for the three-year tax holiday, MUDRA for micro-credit, the PLI tracks for manufacturing, DST NIDHI for deep-tech prototyping, SIDBI for growth-stage credit — collectively they form a stack that, for the businesses they target, is meaningful.

But "well known in absolute terms" is misleading. Visibility is concentrated. A handful of programmes get media coverage. The rest live on ministry websites founders rarely visit. Even within headline programmes, the procedural detail is opaque. SISFS isn't applied for directly — it's routed through a recognised incubator. Section 80-IAC requires DPIIT recognition plus a separate Inter-Ministerial Board approval, which most DPIIT-registered startups never initiate. MUDRA isn't a loan you apply to the government for; it's a refinancing line through banks, and the eligibility translation happens at the branch level.

The Central stack is real. But it is not navigable from the outside.

The State layer

State schemes are where most of the variance lives — and where the gap between policy and practice is widest.

Each state runs its own startup or industrial policy. Karnataka Elevate offers up to ₹50 lakh to tech-stage startups. Tamil Nadu's TANSEED supports early-stage product startups. Maharashtra's Startup and Innovation Policy combines fiscal incentives with subsidised co-working. Telangana's T-Hub-linked grants target deep-tech. Kerala's KSUM offers product development, marketing, and patent filing reimbursements. Gujarat has its industrial subsidy regime.

State schemes matter for three reasons. First, they often address gaps Central schemes don't: women entrepreneurs, SC/ST founders, regional businesses, MSMEs in specific corridors. Second, the cumulative quantum can rival Central schemes for a business in the right state. Third, eligibility is often easier — fewer national certifications required.

But state schemes are also where the system fragments hardest. Eligibility rules vary between scheme rounds within a state. Portals are built independently. Approval times range from six weeks to nine months. A founder in Bengaluru may know about Elevate; moving to Pune, they will likely not know what Maharashtra offers. The information asymmetry is total.

What "working" would actually look like

It is useful to be specific about the target. A well-functioning system would have five features.

Discoverability. A founder, on incorporating, would have a single place to learn what Central and State schemes they qualify for — by sector, stage, certification, and location. They would not need to know the schemes existed in the first place.

Navigability. Filing would happen digitally, with documents the business is already required to maintain. Forms would be pre-filled where possible. Rejections would come with specific, addressable reasons.

Predictability. Approval timelines would be published and held to. A founder would know within weeks whether a grant is coming, not wait nine months in silence.

Outcome tracking. The government would publish, scheme by scheme, what disbursement produced — jobs, output, exports, follow-on capital. Schemes that worked would be expanded. Schemes that did not would be redesigned or retired.

Integrity. Intermediaries who help businesses file would charge transparent fees, not take a percentage of the grant.

None of these requirements is technically hard. Several are partly in place — Udyam registration, Jan Samarth, the DPIIT recognition portal, GSTN data. But the system as a whole still falls short on every dimension.

Why it falls short — by stakeholder

The reasons are different for each stakeholder. Treating them as one problem misses the actual structure.

For founders and businesses

The first problem is discovery. Most founders learn about schemes accidentally — from another founder, a chance LinkedIn post, an investor mentioning it in a meeting. There is no canonical front door. The founders best served by the system today are those with the densest networks: alumni of well-connected incubators, founders with experienced CAs, second-time entrepreneurs.

The second problem is opportunity cost. A pre-revenue founder running a five-person company genuinely cannot afford a week of paperwork. Even if they know about a ₹20 lakh grant, the expected value — given time cost, approval probability, and timeline uncertainty — often looks worse than just shipping their product.

The third problem is the consultant economy. Founders who do pursue grants frequently encounter consultants who promise help in exchange for 10–25% of the grant amount contingent on approval. This arrangement siphons capital that was meant for the business, and teaches founders that the system is gameable — which is corrosive.

The fourth problem is fear. Many founders are reluctant to engage with government processes at all, on the assumption that triggering one filing invites scrutiny in others. Whether or not that fear is well-founded — and it usually isn't — it is real and it is widespread.

For chartered accountants

CAs are the natural last-mile professionals for grant filing. They already know the company's books, hold the documents, and have the credentialing. Yet most CAs do not offer scheme filing as a structured service.

The reason isn't competence. It is economics. Each scheme has unique documentation, unique forms, and unique idiosyncrasies. The first filing a CA does of any scheme is loss-making — the learning curve eats the fee. Without volume, no CA can specialise.

There is also a liability problem. A CA who files a grant application that's later denied for a reason the CA could have flagged is exposed. Without a structured process, many CAs simply tell their clients they don't handle scheme work.

For incubators

Incubators sit closest to startups and have the strongest theoretical incentive to ensure their cohorts claim what they qualify for. The constraints are bandwidth and standardisation. A 50-startup cohort has 50 different eligibility profiles. A programme manager who tries to track each one across Central and State schemes ends up doing the work themselves — and most programmes are not staffed for that.

The default becomes a one-time orientation session, followed by leaving each startup to figure it out individually.

For state governments

State governments face the hardest stakeholder problem: they are simultaneously designing schemes, administering them, measuring them, and competing with neighbouring states — usually with resource-constrained departments.

Good scheme design and weak implementation coexist routinely. A well-funded ₹50 lakh programme may have lower uptake than a poorly-funded ₹5 lakh programme in a neighbouring state, because the latter has a working portal and the former does not. The political incentive is often to demonstrate beneficiary counts rather than outcome quality, because counts are easier to report.

For the Central government

Central schemes depend on state implementation — state departments, banks, incubators, district offices. A scheme that looks well designed from Delhi can be invisible in Bhubaneswar.

There is also a data fragmentation problem. The Central government cannot easily answer "of all businesses that received a Central grant last year, how many also received a state subsidy, and how many filed for tax holidays?" The data exists in fragments across MeitY, MoF, MoMSME, MCA, DPIIT, GSTN, EPFO, and state portals. Without connected data, outcome measurement is partial and policy iteration is slower than it should be.

What can be fixed

Some of these problems are policy problems. Most are not. The biggest gap is not in scheme design but in scheme delivery — the layer between a business and the schemes it qualifies for.

The most leveraged fixes are infrastructural.

A single business identifier across schemes. Udyam registration is a partial version of this. If a single identifier connected GST, MCA, Udyam, DPIIT, EPFO, and state portals, eligibility could be precomputed and filing pre-filled. The technology exists; inter-ministerial coordination is the bottleneck.

A standardised eligibility taxonomy. Each scheme today defines eligibility in its own language. A taxonomy mapping every scheme's eligibility criteria to a common vocabulary — sector codes, employee count bands, turnover bands, certification statuses — would make automated matching trivial.

Outcome dashboards by scheme. A public dashboard showing applications received, approved, disbursed, and outcomes generated would change the political economy of scheme design. Schemes that work would attract more funding. Schemes that don't would face pressure to improve.

A regulated intermediary layer. The 10–25% commission consultants are the natural product of an unstructured market. A more mature market would have credentialed intermediaries operating on transparent fixed fees, with public approval-rate data.

Cohort-level tools for incubators. Programme managers running 50-startup cohorts shouldn't be doing eligibility tracking by hand. A portfolio view of cohort eligibility, filing status, and disbursement is not complex to build — widespread adoption would meaningfully reduce leakage at the incubator layer.

What is already working

It is easy to be pessimistic about all of this. It would also be inaccurate.

Over the last decade India has built infrastructure that did not previously exist: DPIIT recognition, Udyam registration, Jan Samarth, digital tax filing, and a tradition of state startup policy that is now genuinely competitive. The PLI scheme, whatever its limitations, has been one of the most ambitious industrial policy experiments any large democracy has attempted this decade.

The system is unfinished, not broken. The design layer has run ahead of the delivery layer — and the delivery layer is where the next decade of work happens.

What founders, CAs, and incubators should do today

The systemic fixes will take years. The practical question for people in the ecosystem is: what to do now?

For founders: Find out what you qualify for — specifically, with amounts and timelines, not a generic list. Most businesses qualify for more than they realise. The BenefitStack assessment takes three minutes, shows you every Central and State scheme ranked by your eligibility, and is free. You do not need a CA or a consultant to see your report. The cost of checking is zero; the expected value is often in the lakhs.

For CAs: The gap between "I do grant filings ad hoc" and "I offer grant filing as a service line" is mostly a matter of process. The CAs who systematise this in the next two to three years will own a meaningful new revenue line and a more defensible practice. The CA partner programme at BenefitStack is one way to build that volume without building the research infrastructure from scratch.

For incubators: Most cohorts are missing capital they qualify for — quantifiable capital, often in the tens of lakhs per startup. Running a cohort-level eligibility scan is a one-time exercise. The result is almost always surprising, and almost always actionable. Partner with BenefitStack to run that scan across your portfolio.

For state and Central administrators: The schemes are already there. The question is whether the businesses they were designed for can actually reach them. Investing in the delivery layer — portals that work, timelines that are held to, data that is published — is where the next meaningful improvement is.

Nothing in this piece is legal, tax, or financial advice. Scheme eligibility shown on BenefitStack is based on publicly available criteria and is indicative — final eligibility is determined by the relevant ministry or department.