When a venture fund sends a term sheet, the founder's next 30 days are defined by one document: the data room. The partner who wrote the term sheet has already decided. The associate, the analyst, and the outside counsel doing the diligence have not. Their job is to find reasons to walk away.
They will not walk away over a bad product. They walked away from those months ago. They walk away over signals that the founder does not run the company with operational rigor. The data room is where those signals live.
These are the seven red flags that consistently surface during seed and Series A diligence, in rough order of how often they show up.
Red Flag 1: The Cap Table in the Pitch Does Not Match the Cap Table on Carta
The single most common diligence finding. The pitch deck shows 18 percent founder ownership. The Carta export shows 16 percent. The discrepancy is usually explained by "options we are about to issue" or "the advisor grant that is not finalized yet."
That is not an acceptable explanation for an associate running diligence. They will reconcile both versions, identify the gap, and send a clarification request that takes two days to answer. Repeat across four or five data points and the diligence process drags for weeks.
The deeper problem: it signals the founder does not have a single source of truth for the cap table. If the pitch shows one version and the platform shows another, which is used for financial modeling? Which was shown to prior investors? Which is accurate for the 409A?
How to fix it: Before sharing the data room, export the cap table from Carta (or Pulley), verify every row, and use that as the single source. Update the pitch deck to match. If you have made verbal option promises that are not documented, either document them or remove them from the narrative.
Red Flag 2: ARR Numbers That Change Between the Deck, the Data Room, and the Board Packet
Founders pitch ARR. Boards review ARR. Data rooms contain ARR. These numbers almost never match across documents. The reason is usually not fraud; it is that ARR has no single definition and the founder has not enforced one.
Common variants:
- Contracted ARR -- all signed contracts regardless of whether the service has started.
- Billed ARR -- what has actually been invoiced.
- Activated ARR -- customers who have started using the product.
- GAAP revenue annualized -- current monthly recognized revenue times 12.
- Net new ARR -- the growth number excluding renewals.
Every version is defensible. None is interchangeable. If the deck says $1.4M ARR and the data room's monthly revenue times 12 suggests $1.1M, the investor will ask, and the founder will usually answer with a nervous "well, some of those contracts start next month..." That answer is the problem.
How to fix it: Pick one ARR definition. Document it in a one-page methodology document and include it in the data room. Reconcile every number in the pitch deck to that definition before sending. When you have to use a different metric (contracted, billed, etc.), label it explicitly. Pair the ARR number with a credible runway forecast and unit economics breakdown so investors can triangulate the story.
Red Flag 3: Missing or Late IP Assignment Agreements
Every person who has written code, designed, or contributed intellectual property to the company needs a signed IP assignment. That includes:
- Co-founders (ideally on day one, signed alongside the certificate of incorporation)
- Full-time employees (usually via the offer letter or a separate PIIA -- proprietary information and inventions agreement)
- Contractors and freelancers
- Offshore engineering teams
- Anyone who contributed during the "we were not a company yet" phase
The failure modes are predictable. An early contractor wrote the first version of the product, was paid in cash, and was never asked to sign anything. A co-founder's former employer has a claim to IP developed during overlap periods. An offshore developer in a jurisdiction where work-for-hire does not apply by default.
In diligence, the legal team will request every IP assignment agreement for every contributor. Missing agreements lead to one of two outcomes: the deal pauses while you track down contractors to sign after the fact, or the investor demands a representation that all IP is clean and takes a quiet discount for the risk.
Getting a contractor to sign an IP assignment two years after the fact is awkward. Getting them to sign one after the company is about to raise $10M is either expensive or impossible.
How to fix it: Audit every person who has ever contributed to the product. Get agreements signed before fundraising, not during. For anyone you cannot reach, document the effort and classify the risk. If you have offshore contractors, confirm the assignment is enforceable in their jurisdiction; often it requires a local addendum.
Red Flag 4: 83(b) Elections That Were Never Filed
Restricted stock with vesting requires an 83(b) election to be filed with the IRS within 30 days of grant. The filing locks in the cost basis at the time of grant, which is usually near zero for a pre-seed founder.
Miss the 30-day window and the IRS treats each vesting tranche as taxable income at the then-current fair market value. For a founder who granted themselves restricted stock at $0.0001 per share and now has a company worth $20M, each vesting cliff creates ordinary income equal to the fair market value increase. The tax bill grows with the company.
The 83(b) election is a single page. It takes 15 minutes to prepare and mail. It is missed constantly.
During diligence, the legal team will ask for a copy of the 83(b) election for every founder and every early employee who received restricted stock. If any are missing, the founder has personal tax exposure that cannot be retroactively fixed, and the investor knows it.
More importantly, the investor now knows the founder's legal operations have a gap. If 83(b) elections were missed, what else was missed?
How to fix it: For every founder and early employee, confirm the 83(b) election was filed within 30 days of grant and that you have a stamped copy or certified mail receipt. If an election was missed, consult a tax advisor immediately -- there may be partial remediation options (a fresh grant, accelerated vesting, etc.) but the window is narrow.
Red Flag 5: A Bank Account in a Founder's Name
This still happens at seed more often than it should. The founder opened the bank account in their own name during the "we are just experimenting" phase, never moved it to the corporation, and is now running company revenue through a personal account.
Consequences:
- The IRS can recharacterize the business as a sole proprietorship or disregarded entity, with very bad tax consequences.
- The liability protection of the corporation is pierced, exposing the founder's personal assets.
- Every transaction on the personal account creates commingling and a diligence cleanup project.
- The corporation cannot claim the deductions that flowed through the personal account.
In diligence, the investor will ask for bank statements on all company accounts. If the "company account" is in the founder's name, that is the end of the conversation until it is fixed.
How to fix it: Open a corporate account at Mercury, Brex, or your preferred bank. Transfer the balance. Redirect all customer payments, payroll, and vendor payments to the new account. If you have been operating on a personal account for months, get a tax advisor involved -- the remediation usually requires recharacterizing the flows retroactively.
Red Flag 6: No Board Consents for Major Actions
Corporate formalities are unglamorous and easy to skip. Almost every startup skips them until it matters.
The major actions that require board approval include:
- Option grants (individual and aggregate)
- Material agreements (leases, loans, acquisitions)
- Equity issuances
- Officer appointments
- Amendments to the certificate of incorporation
Many founders issue options, sign leases, and appoint officers without board consents. It feels like process overhead. Until diligence.
The legal team will request every board consent for the company's history. Gaps signal that the founder operates the company informally, which is acceptable at pre-seed but not after institutional money is on the cap table. More concretely, un-consented option grants may not be legally valid, which creates uncertainty about the cap table itself.
How to fix it: Use Carta, Pulley, or your corporate counsel to maintain a running board consent document. Every material action gets a consent. If you are behind, ratify past actions in a catchup board consent. The cost is a corporate counsel invoice; the benefit is clean diligence.
Red Flag 7: Financials That Do Not Tie to the Bank Statements
Every month, the general ledger should reconcile to the bank statement. Opening balance plus deposits minus withdrawals equals closing balance. If the books and the bank do not agree, one of them is wrong, and usually it is the books.
Reconciliation gaps greater than 30 days old are a red flag. Gaps older than 90 days are a serious red flag. Gaps older than a year are a fundraise-killer.
Common causes:
- Transactions that were never categorized (stuck in "ask my accountant" or equivalent limbo)
- Inter-account transfers double-counted or missed
- Customer refunds not recorded
- Expense reimbursements paid but not booked
- Bookkeeper turnover without a clean handoff
In diligence, the quality of earnings review will tie every month's financial statement to every month's bank statement. Any gap they find will require explanation. Enough gaps and they will question whether the financials are reliable at all.
How to fix it: Close every month within 10 business days. Reconcile bank and credit card accounts before closing. If you have open reconciliation items more than 60 days old, clear them before fundraising. If your bookkeeper cannot close on this cadence, get a new bookkeeper.
The 48-Hour Data Room Cleanup
If you are reading this 48 hours before diligence starts, prioritize in this order:
- Reconcile the cap table across every document.
- Lock down one ARR definition and reconcile every number to it.
- Confirm every 83(b) election is on file.
- Close the most recent month and reconcile the bank.
- Produce a list of every IP contributor and confirm agreements are signed.
Everything else can be addressed in the Q&A process. These five cannot.
The Broader Point
An investor running diligence has already decided they want to fund you. Their job now is to find reasons to walk. You cannot prevent every issue -- startups are messy and some gaps are inevitable -- but you can control whether the gaps suggest carelessness or suggest the normal state of a fast-moving company.
Clean data rooms signal that the founder treats finance as a first-class function. Messy data rooms signal that finance is an afterthought. At seed, investors forgive the second kind. At Series A, they do not.
If you are 90 days out from a raise, our Fundraising Readiness feature scores your data room against the typical diligence checklist and flags gaps before the investor does. Book a free consultation and we will do the scan whether or not you become a customer.
