Your cap table will be read by a stranger, and it will not flatter you

At some point a person you have never met will open a spreadsheet and a folder of documents and decide whether your company is financeable. They will not be hostile. They will be in a hurry, and they will be looking for reasons to stop, because stopping is cheap and being wrong is not.

Almost everything that stalls a round at this stage is boring. It is rarely a bad business. It is something that was true informally for two years and was never written down.

The defects that actually show up

A founder who left and kept everything. Three people start a company. One leaves after seven months. Nobody wants the conversation, so the shares stay where they are. Two years later a third of the company belongs to someone who is not in it, and every investor has to price the risk that this person says no to something important later. This is the single most common serious defect, and the only cheap moment to fix it was the week they left.

Founder shares with no vesting. Investors want founders to earn their equity over time, and if that was never set up, it usually gets imposed as a condition of the round — from zero, on the full stake, four years from closing. Founders who have already worked three years are then vesting from scratch on a company they built. Vesting agreed at incorporation, with credit for time already served, is a far better position than vesting negotiated under a term sheet.

Promises that live in Telegram. “You will get 2%.” “There is a bit of equity in this for you.” These are remembered precisely by the person who was promised and vaguely by the person who promised. An undocumented claim on the company is not a small problem, because it cannot be quantified and it cannot be closed without finding the person and asking them to sign something at the worst possible moment.

Advisors on standard grants that were never standard. Someone helped early, was given equity, and has not been heard from in a year. There is no vesting schedule and no definition of what they were supposed to do. Every advisor grant should have vesting and a scope. One that has neither is a gift the company cannot take back.

An option pool that exists in conversation only. Everyone says “we have set aside 10% for the team”. Nothing has been authorised, no plan document exists, and the people who were told they have options have nothing. Worse, the pool will be sized in the round and taken out of the pre-money — so it dilutes the existing holders and not the incoming investor. If it is coming out of your side anyway, know the number before someone else chooses it.

A share register that does not match the documents. The spreadsheet says one thing, the corporate records say another, and the last three changes were never filed. Whichever is right, reconciling it takes weeks of somebody’s counsel, and the round waits.

A dormant entity nobody mentions. An old company from a previous attempt, a second entity opened for a grant, a subsidiary that was never closed. If it holds anything — a contract, an account, a trademark, code written by a founder before the real company existed — it belongs in the diagram rather than in a footnote discovered in week three.

Intellectual property that belongs to a person. Code written before incorporation or by a contractor on an invoice with no assignment clause does not automatically belong to the company. In a software business this is the one that can stop a deal outright, because the asset being bought turns out not to be owned.

Why these get expensive rather than just annoying

Each one on its own is fixable. What makes them costly is when they are found.

Found before you raise, a departed co-founder is a conversation and a document. Found during diligence, it is leverage: the company needs the round, the investor knows it, and the person holding the shares now has an audience. The same facts, a different week, a different price.

That is the whole mechanism, and it is identical to the one that makes a late corporate reorganisation expensive — the same cost curve, for the same reason. Nothing changes except who is waiting.

The check to run this month

Not a diligence exercise. An afternoon.

1. List every person who has ever been promised equity. Everyone. Including the ones who left and the ones who only helped. Write the number next to each name.
2. For each one, find the document. If there is no document, that is the finding.
3. Check the register against reality. Does the official record match your spreadsheet, and does your spreadsheet match what people believe?
4. Confirm the company owns its code and its brand. Every contributor, including founders, on an assignment. Trademarks and domains in the company name, not a personal one.
5. Draw the entities. Every company that exists, what it holds, who owns it. If the picture needs explaining, it needs fixing.
6. Decide the option pool number before an investor decides it for you.

Anything you find now is administration. The same thing found in eleven weeks is a negotiation.

The honest summary

Investors are not looking for a perfect cap table. They are looking for one that holds together when a stranger reads it, where the surprises are small enough to absorb, and where the founders clearly know their own company’s history.

The companies that sail through this are not the tidy ones by temperament. They are the ones that did this check once, early, when it was embarrassing and free.

If you want a second pair of eyes before an investor is the first to look, send it over. We have read a lot of these, including our own mistakes.

General guidance, not legal advice. What is required to fix any of this depends on your jurisdiction and your documents — take professional advice before acting.