Last reviewed: 25 September 2026. DPIIT rejections cluster around one theme: the application did not prove the definition. Either the entity fails a hard gate (age, turnover, structure) or — far more often — the innovation write-up reads like a brochure instead of evidence. Both are fixable; only one deserves a refile. Eligibility basics live in the recognition guide, and you can test the hard gates in two minutes with the DPIIT eligibility checker.
The current limits (G.S.R. 108(E), 4 February 2026)
| Test | Standard startup | Deep-tech startup |
|---|---|---|
| Age from incorporation / registration | Up to 10 years | Up to 20 years |
| Turnover in any financial year since incorporation | Not above ₹200 crore | Not above ₹300 crore |
| Eligible forms | Private limited company, LLP, registered partnership firm, cooperative society, multi-State cooperative society | |
| Formed by splitting up or reconstructing an existing business | Not eligible | |
The February 2026 notification replaced the 2019 framework, which capped turnover at ₹100 crore and did not cover cooperative societies. A company rejected earlier on the old turnover limit should re-test itself against the revised figures.
Hard gates vs soft failures
| Type | Examples | What to do |
|---|---|---|
| Hard gate (no refile can cure) | Entity over 10 years (20 for deep-tech); turnover in any year above ₹200 crore (₹300 crore deep-tech); ineligible form (proprietorship); formed by splitting/reconstructing an existing business | Accept the gate — energy goes to other benefit routes, not repeat filings |
| Soft failure (rebuild and refile) | Generic write-up (“we provide quality services using technology”); no evidence links; scalability asserted, not shown; wrong documents | Rewrite with specifics + attach proof — this is the majority of rejections and the fixable one |
What a winning write-up contains
- The PROBLEM stated with numbers someone can check
- What exactly is new/improved versus what exists (name the alternatives honestly)
- Evidence links: product demo, users/pilots, traction, IP filings if any
- Employment/wealth-creation potential shown through the model, not adjectives
Before and after: one rejected write-up, rebuilt
| Rejected version | Rebuilt version |
|---|---|
| “We are a technology-driven logistics company providing efficient and reliable delivery services to our valued clients using modern tools.” | “Small pharmacies in Thane wait 2–3 days for restocking because distributors batch orders by route. Our routing software pools orders from 140 pharmacies across 6 distributors and cuts the average wait to under 24 hours (pilot data, March–June 2026, link). The routing logic is ours; a patent application is filed (application number attached). Revenue comes from a per-drop fee, so each new distributor adds routes without adding head-office staff.” |
| No evidence attached | Pilot dashboard export, two distributor letters, product demo video, patent filing receipt |
| Scalability asserted: “huge market potential” | Scalability shown: cost per drop falls as route density rises, with the numbers from the pilot |
The example is illustrative. What changed is not the business but the proof: a reviewer can check every claim in the second column.
The split/reconstruction trap
- Founders migrating an existing business into a “new” startup entity trip this test
- Related-party continuity (same clients, assets, contracts moved over) is what reviewers look for
- Where a genuine new venture shares founders with an old firm, the write-up should address the difference head-on rather than hope nobody notices
Deep-tech applications: prove the attributes
The higher deep-tech limits are not a sector label. The notification describes a deep-tech startup by attributes — new scientific or engineering knowledge, high R&D spending as a share of costs, significant novel IP with a commercialisation plan, and real technical uncertainty. Detailed benchmarks have not been prescribed, so an application relying on the 20-year or ₹300 crore limits should carry the evidence: R&D expense schedules from audited books, IP filings, technical reports and a development roadmap. See our deep-tech recognition service for how we assemble that file.
Frequently asked questions
How do I find out WHY we were rejected?
The communication usually indicates the ground generically — read it against the hard-gate list first; if no gate applies, assume the write-up failed and rebuild it with evidence.
How soon can we reapply?
Fresh applications after curing defects are the normal course — the calendar cost is preparation time, so invest it in the rewrite rather than speed.
Our services company was rejected for “no innovation”. Hopeless?
Services CAN qualify — process/technology-led improvement with evidence clears; “we serve clients well” never will. The rebuild is about showing the mechanism, not relabelling it.
Does rejection affect our other registrations or taxes?
No — recognition is a benefits gateway, not a compliance status. Its absence just keeps the section 140 (old 80-IAC) tax holiday under the Income-tax Act, 2025 and startup-scheme doors shut until granted.
We crossed ₹200 crore once, years ago. Still fatal?
Under G.S.R. 108(E) of 4 February 2026 the turnover test looks at every financial year since incorporation — turnover above ₹200 crore in any year (₹300 crore for a deep-tech startup) is a hard gate. Check the actual audited numbers before concluding either way; a company that crossed only the old ₹100 crore limit may now be within the revised limit.
Can a proprietorship get recognised?
No — the eligible forms are private limited company, LLP, registered partnership firm, and (since February 2026) a cooperative society or multi-State cooperative society; conversion first, recognition after, where the business genuinely continues as new.
Can a cooperative society now get DPIIT recognition?
Yes. G.S.R. 108(E) of 4 February 2026 added cooperative societies registered under State or Union territory law and multi-State cooperative societies to the eligible forms. The same age, turnover, innovation and no-reconstruction tests apply, so the write-up still has to show what is new or improved.
What evidence supports a deep-tech application?
The notification describes deep-tech startups by attributes: solutions based on new scientific or engineering knowledge, high R&D spending as a share of total costs, significant novel intellectual property with clear commercialisation plans, and substantial scientific or technical uncertainty. Useful evidence includes R&D expense schedules from the books, patent filings or other IP records, technical papers or lab results, and a development roadmap. Detailed benchmarks have not been prescribed, so treat the deep-tech limits as available only where these attributes are documented.
Do we need a patent to prove innovation?
No — IP helps but users, pilots, metrics and a demonstrable product speak just as loudly. Evidence beats certificates of every kind.
Is professional drafting of the write-up allowed?
Of course — what matters is that every claim is yours and checkable. Drafting sharpens truth; it cannot substitute for it.
We test the hard gates honestly, rebuild the innovation case with evidence, and file recognition — then sequence the section 140 (old 80-IAC) tax holiday and schemes after it.
DPIIT Recognition80-IAC AdvisoryRequest a Scope DiscussionThis article is a general educational summary as on 25 September 2026 and is not professional advice or an assurance of any approval, registration or outcome — departmental decisions rest with the authorities on each case’s facts. Requirements change; verify current rules or discuss your specific case before acting.