A term sheet feels like the finish line. It is actually the starting gun for the part of a deal where the company gets read, line by line, by people whose job is to find the reason to pay less or walk away. Legal due diligence is that read. It is not hostile, but it is thorough, and it rewards the founders who built the company as if this day was always coming, while it punishes the ones who assumed they would clean it up later.
This is a founder-level walkthrough of what a diligence team actually checks, the specific gaps that stall or shrink a deal, and how to assemble a company that clears the review in weeks rather than months. Treat it as the hub that ties together the individual decisions, ownership, equity, contracts, compliance, that each get their own attention below. For the practice behind this work, see startup contracts and founder equity strategy.
What legal due diligence is, and when it happens
Legal due diligence is the verification step: the investor or acquirer confirms the company is what it represents itself to be. Does it own its technology? Is the cap table accurate? Are its contracts valid and transferable? Does it carry liabilities no one mentioned? The review usually runs three to six weeks from the signed term sheet to funding. A company with organized records and a clean chain of title can close in three to four; a company with gaps drags longer, and time is the enemy of a deal. Every extra week is another chance for markets to move, priorities to shift, or a valuation to soften.
The important reframe: diligence is not a test you cram for. It reads whatever state the company is already in. The work that determines the outcome happened months or years earlier, in the decisions to sign or skip the paperwork as the company was built.
The checklist: six areas a diligence team works through
Diligence questionnaires vary, but they cluster into the same six domains. Here is what sits inside each, and what the reviewer is really trying to confirm.
1. Corporate structure and the cap table
The foundation. Reviewers confirm the company is properly formed and in good standing, that governance documents exist and are consistent, and that the capitalization table is accurate to the share. They look at the certificate of incorporation and bylaws, board and stockholder consents, the stock ledger, option grants and the plan, and any prior financing documents. The most common problem here is a cap table that does not reconcile, promised equity that was never documented, option grants approved verbally, or a SAFE the founder forgot converts on this very round. If you are still deciding how the company is built, that starts upstream, at choosing the right entity.
2. Intellectual property and chain of title
For most startups this is the asset being bought, so it draws the hardest look. Reviewers trace ownership from every contributor to the company: founders (including pre-incorporation work), employees, contractors, advisors, and agencies. Missing contractor IP assignments kill more deals than any other single issue, because the default rule is that the creator, not the company, owns the work. This is worth its own read on who owns the code a contractor writes. Reviewers also check trademarks, any patents, open-source usage and its license obligations, and inbound and outbound licenses.
3. Material contracts
The agreements that carry revenue, obligations, and risk: customer and vendor contracts, MSAs and SOWs, key partnerships, and leases. Reviewers care about two things beyond the terms. First, are the contracts valid and enforceable. Second, and often overlooked, are they assignable, because a change-of-control or anti-assignment clause can mean a key customer contract does not survive an acquisition without consent. A revenue base that cannot transfer is a revenue base a buyer discounts.
4. Employment and equity
Who works for the company, on what terms, and whether the equity paperwork is clean. Reviewers check offer letters and employment agreements, contractor agreements, proprietary information and inventions assignments (PIIAs), worker classification, and the equity grants themselves, including whether founders filed their 83(b) elections on time. Misclassified workers and missing or late equity elections are frequent friction points that are painful to fix retroactively.
5. Data, privacy, and regulatory compliance
Increasingly the section that surprises founders. Reviewers ask how the company handles personal data, what its privacy posture is, whether it meets the obligations of its sector, and, for AI companies, how it governs its models and whether it has considered rules like the EU AI Act. A short, credible AI governance file and a coherent data story move this section quickly; their absence turns it into a series of follow-up questions that slow everything down. This connects directly to AI governance and risk readiness.
6. Financial and tax
Financial statements, material tax filings and obligations, and any outstanding liabilities. For early companies the legal and financial overlap most on equity: the 83(b) elections above, option pricing and any 409A valuation, and payroll tax on any misclassified contractors. The reviewer is confirming there is no tax surprise waiting to land on the buyer after close.
Diligence does not test how hard you can work under pressure. It tests what you already wrote down. The company that built clean is the company that closes fast.
The red flags that delay, re-price, or kill a deal
Most diligence findings are not fatal. But a handful recur, and each one costs time, leverage, or price. The ones worth pre-empting:
- Unclear IP ownership. Missing founder or contractor assignments. The single most common and most damaging finding.
- Cap table errors. Undocumented promises, option grants that were never formally approved, or shares that do not reconcile.
- Missing or late 83(b) elections. Preventable, unforgiving, and a recurring flag in founder equity.
- Worker misclassification. Contractors treated like employees, creating tax and IP exposure at once.
- Non-assignable contracts. Key revenue that cannot transfer to a buyer without a third party's consent.
- Data and AI gaps. No privacy posture, no data processing terms, no governance story for the model stack.
When a reviewer finds one of these mid-deal, the usual outcomes are a delay while it gets fixed, a lower valuation to price the uncertainty, or a holdback that keeps part of the proceeds in escrow until it is resolved. None of those are catastrophic on their own. Together, and discovered late, they are how a strong company negotiates from a weak position.
Build the data room before you need it
The practical antidote is unglamorous: keep an organized data room, and keep it current. Create folders for incorporation and governance, the cap table and equity, IP assignments, material contracts, employment documents, and compliance, and file each document as it is signed rather than reconstructing everything under deadline. A founder who can hand over an organized room the day a term sheet lands signals competence and shortens the review. A founder who spends the first two weeks of diligence hunting for signatures signals the opposite, and invites a closer look.
The founder mindset: build for the diligence you have not scheduled
The through-line of every section above is the same principle: the structure you set today is read later by someone on the other side of a raise or a sale, and the work should be done so those moments are confirmations, not cleanup. That is not a reason to over-lawyer an early company. It is a reason to get a small number of things right as you go, ownership, equity, classification, key contracts, so the paper trail is already clean when it finally matters. The founders who pass diligence quickly are rarely the ones who worked hardest during the deal. They are the ones who made a handful of unglamorous decisions early and wrote them down.
- Diligence reads the company's existing state; the outcome is set by decisions made long before the term sheet.
- Six domains: corporate and cap table, IP chain of title, contracts, employment and equity, data and compliance, financial and tax.
- The recurring deal-killers are unclear IP, cap table errors, missing 83(b)s, misclassified workers, and non-assignable contracts.
- Late findings cost time, valuation, or a holdback, even when they are fixable.
- Keep an organized, current data room; the company that built clean closes in weeks, not months.
If you are heading toward a raise or an acquisition, a diligence-readiness review now, while there is no clock running, is exactly the kind of upstream work where an hour of judgment saves a quarter of cleanup. The goal is simple: make the day someone finally checks a formality, not an event.
Go deeper on the pieces this checklist depends on: code and IP ownership, the 83(b) election, the equity split, how the money converts, and AI compliance. Or start a conversation about getting diligence-ready.