Diligence readiness: what we check, and what we find
The business is usually fine and the numbers mostly hold up. What delays a deal is the record of the company: the documents, the filings, the registers, the contracts.
Summary
- What delays, reprices and occasionally kills a round is rarely the company's performance. It is the record: the share register, the minute book, the option records, the key contracts and the filings.
- The same findings recur: a cap table that does not reconcile, an option pool never properly created, missing IP assignments, undisclosed related-party dealings, missed foreign investment filings and charges still shown as open.
- Fix it in the 90 days before you raise, not after a term sheet lands, when every fix happens in front of the investor and every delay is attributed to you.
Due diligence is the review an investor or buyer does before putting money in. Across the diligence exercises our team has run, the pattern is remarkably consistent.
The business is usually fine. The numbers mostly hold up. What causes delay, renegotiation and occasionally a dead deal is almost never the company's performance. It's the record of the company: the documents, the filings, the registers, the contracts.
Founders prepare the pitch. Almost nobody prepares the record. Here's what gets looked at, what we keep finding, and what to do about it before you're under a deadline.
Two companion pieces cover the ground on either side. Pre-due diligence sets out the nine areas a self-review should cover, and what due diligence actually evaluates is about the judgement an investor forms beyond the documents. This one is narrower and earlier: the record itself, what gets opened first, the findings we keep making, and a 90-day plan to put them right.
What we open first
Before the financial model, before the deck, four things:
The share register and the filings at the Ministry of Corporate Affairs (MCA). Who owns the company according to the official record, and does it match what the founder just showed us? This takes twenty minutes and sets the tone for everything after it.
The board minutes. Not for the decisions, but for whether the book exists, whether it's been kept contemporaneously, and whether the big decisions were actually approved.
The employee stock option records. Was the pool properly approved, is there a written scheme, do grant letters exist and match the ?
The top five customer contracts. Payment terms, termination rights, exclusivity, and whether anyone can walk away if the company changes hands.
If those four are in order, the rest of the exercise proceeds as a review. If two or more are not, the tone shifts from verification to investigation, and the questions get harder for the rest of the process.
The findings we keep making
Across engagements, the same issues recur. In rough order of frequency:
The cap table doesn't reconcile. The founder's spreadsheet, the share register and the MCA record are three different documents, usually because registers were never updated after a round. This is covered properly in our guide to your cap table and share register.
The option pool was never properly created. It appears in the cap table and the offer letters, but the shareholder approval was never passed or the written scheme doesn't exist. Employees hold promises, not options.
IP isn't assigned. The first engineer was a contractor. The design agency built the app. Neither signed an assignment. See contractor IP assignment: this one has no easy fix after the fact, because it requires the cooperation of someone who no longer works for you.
Related-party transactions nobody disclosed. A vendor owned by a founder's family, an office leased from a relative, consulting fees to a co-founder's other company. Rarely improper. Almost always unapproved and undisclosed, which makes it look worse than it is.
Foreign investment filings missed. A round closed and the filing that follows it, Form FC-GPR within 30 days of allotting shares to a foreign investor, never happened. Straightforward to regularise with a late submission fee, expensive in timing.
Charges still showing as open. A loan repaid in 2022, with the lender's security still recorded against the company because the satisfaction form, Form CHG-4, was never filed.
Money that isn't share capital and isn't documented. Founder loans, family money and customer advances sitting on the balance sheet with no agreement, no declaration, and no Form DPT-3 return filed.
Change-of-control clauses in key contracts. Customers who can terminate or must consent if the company is acquired, scattered across the agreements that matter most. See change-of-control consents.
An old entity with unfiled returns. The founder's previous venture, wound down informally, still accruing defaults. Three years of unfiled returns there disqualifies its directors for five years, and they must then vacate their seats on every other board, including the company being invested in.
None of these are exotic. All of them are found in a first pass.
Why this costs real money
Founders assume paperwork problems are administrative. They're not, for three reasons.
Timing. Each finding becomes a condition precedent: something to be satisfied before closing. Each condition needs documents, approvals and sometimes third-party cooperation. A round that should close in six weeks closes in twelve, and you you'd planned to extend.
Negotiating position. Every finding is information asymmetry resolving in the investor's favour. You are now negotiating from a position where you have been shown not to know your own record. Valuation reductions rarely get attributed to paperwork, but the conversations start there.
Personal exposure. Where something can't be cleanly fixed, the resolution is usually a warranty and an indemnity from the founders, capped at some amount and surviving for some years. You carry that personally, long after closing.
The cheapest of these three is the first, and it's still expensive.
The three workstreams, and who checks what
Diligence isn't one review. It's usually three, running at the same time with three different lists, all landing on you.
- Financial. Quality of earnings, revenue recognition, working capital, related-party flows, tax positions.
- Legal. Contracts, intellectual property, litigation, employment, regulatory permissions.
- Secretarial and corporate. The share register, statutory registers, board and shareholder approvals, filings, foreign investment compliance, the option pool.
Founders route everything to their accountant, who owns the first list and has no visibility on the third. So the secretarial workstream goes unowned until someone asks for a document that was never created. Assign an owner to each of the three before you start, not after the requests arrive.
Building a data room that doesn't create questions
A data room is simply the organised set of documents you give the diligence team access to. Two principles:
Structure signals competence. A clear folder index that mirrors the standard request list tells the reviewer this company is run properly, before they've read anything. A folder called "Documents" containing 340 unsorted files tells them the opposite.
Only final, executed versions. The most common own goal is uploading three variants of the same agreement, two unsigned. Now the reviewer has to ask which one governs, and you've created a question where there wasn't one.
Also: disclose known problems rather than hoping they aren't found. A disclosed issue is a discussion. The same issue discovered by the investor's lawyer is a warranty breach waiting to happen, and it changes how everything else you've said is treated.
The 90-day clean-up
The right time to do this is before you go out to raise, not after a lands. Once a term sheet exists, every fix is visible to the investor and every delay is attributed to you.
Days 1 to 30: find out where you stand. Reconcile the cap table against the register and the MCA record. Confirm the option pool approval and scheme exist. List every promise made in writing. Pull every IP assignment and identify who's missing. Check filings for the last three years. Search for open charges. List any dormant entity connected to the founders.
Days 31 to 60: fix what can be fixed. File the backlog. Close out satisfied charges. Obtain missing IP assignments while people still take your call. Document founder and family loans properly. Identify and approve related-party transactions. Update registers.
Days 61 to 90: assemble and pressure-test. Build the data room against a standard index. Write the disclosure list: the things you'll tell them rather than have them find. Have someone who wasn't involved try to break it.
Three months, mostly part-time. Against six weeks of closing delay and a renegotiated valuation, it's the best return available on that time.
An honest note on what clean records don't do
A clean record won't get you a better valuation. Nobody has ever paid a premium for well-maintained registers.
What it does is remove the reasons someone has to pay you less, or to take longer, or to ask you to personally guarantee something.
In a process where the other side is looking for reasons to adjust, giving them none is the whole game.
Frequently asked questions
What does a due diligence team check first in a startup?
Typically four things: the share register and MCA filings, the board minute book, the employee stock option records, and the largest customer contracts. Together they show whether the rest of the exercise will be a verification or an investigation, and the financial review follows them.
What are the most common due diligence findings in Indian startups?
A cap table that doesn't reconcile with the register and MCA filings, an option pool never formally approved, and missing IP assignments from contractors and founders. Close behind are undisclosed related-party dealings, missed foreign investment filings, charges left open after repayment, and undocumented founder or family loans.
How long does pre-diligence clean-up take?
About 90 days if started before a term sheet: a month to establish where you stand, a month to fix the backlog, and a month to assemble the data room and disclosure list. Started after a term sheet, every fix happens in full view of the investor.
If you want to know which of these would surface in a first pass on your company, our pre-due diligence work runs that review before an investor does.
Current as at September 2026. General guidance drawn from our diligence work, not legal advice: take advice on your specific situation.
How useful was this article?
One tap. It tells us what to write more of.
The next one
Get what we publish next, by email.
Working notes on raising, borrowing, protecting, growing and structuring capital in India. One email a week at most, and you can leave any time.
We use your address only to send this. See our privacy policy.
We store your address to send you these emails and nothing else. See our privacy policy.
Related reading
