What Is a Cap Table? A Field Guide for VC Analysts
Summary
A cap table is the record of who owns what in a company: shares, options, warrants and convertibles, with percentages on an issued and fully diluted basis. For a VC analyst, what is a cap table is really a question about returns. It shows founder ownership, dilution from the pool and SAFEs, and the stake you will hold after the next round.
What is a cap table? It is the ledger of who owns what in a company: every share, option, warrant and convertible, mapped to a holder and a percentage. For a VC analyst it is the first document that shows whether a deal can work. It tells you what your cheque buys, who sits ahead of you, and how much of the company is left after the next round.
Most definitions stop there. The useful question is what the table does to your returns, so this guide covers the structure, the three mechanics that move the numbers, and the checks worth running before the IC memo.
What does a cap table actually contain?
Strip away the formatting and a cap table lists securities, not people. Each row is a holder. Each column is a class of security with a count, a price paid and a percentage on two bases.
The two bases matter. Issued and outstanding counts only shares that exist today. Fully diluted adds everything that could exist: unexercised options, the unallocated option pool, warrants, and convertibles as if they had converted. Investors price rounds on the fully diluted figure, so that is the number to anchor on.
A seed-stage table usually holds five blocks:
Common shares held by founders, often subject to vesting.
Preferred shares from priced rounds, carrying liquidation preferences and other rights.
Options granted to employees, plus the unallocated pool.
SAFEs and convertible notes, which are not shares yet but will be.
Warrants, rare at seed, common after venture debt.
If a founder sends you a table with only percentages and no share counts, ask for the real file. Percentages hide errors that share counts expose.
Why does the cap table decide whether you invest?
Because ownership at exit, not ownership today, drives your return. A fund that needs a 10x on a position has to model the stake it will hold after two or three further rounds. The cap table is the starting point of that model.
Three questions come out of it straight away. Is the founding team still motivated, meaning do they hold enough that a Series A will not leave them with a minority sliver? Is there dead equity, such as a departed co-founder holding 20% with no vesting? And is the structure clean enough that a Series A lead will not ask for a restructuring that costs you ownership?
Out of the 800 or so early-stage companies a typical analyst screens in a career, the cap table is the file most often found wrong. Not fraudulent. Wrong: a missed SAFE, a pool that was never formally approved, an advisor grant that nobody recorded. Coverage before conviction applies here as well.
How does dilution work across rounds?
Dilution is arithmetic, and it is brutal in the way compounding is. Each new round issues new shares, so every existing holder owns a smaller slice of a larger company. The slice is worth more only if the valuation rises faster than the slice shrinks.
A simple seed case. Two founders hold 10 million shares, 100% of the company. A fund invests $2M at an $8M pre-money valuation, so the post-money is $10M and the new investor owns 20%. The founders drop to 80%.
That is the clean version. In practice the investor also asks for an option pool, and that is where the numbers shift.
What is the option pool shuffle?
The option pool shuffle is the habit of sizing the employee pool before the new money arrives, so that the pool dilutes only the existing holders. The headline valuation stays at $8M. The effective valuation falls.
Take the same round with a 10% post-round pool created pre-money. The investor still owns 20%. The pool takes 10%. The founders keep 70%, not 80%. The price per share works out to $0.70, and the founders' 10 million shares are worth $7M, not $8M.
So the headline pre-money was $8M and the effective pre-money was $7M. Founders feel that gap. Investors on the other side of the table, including you at a later stage, should know it exists when comparing deals on valuation alone.
Skip the argument that a pre-money pool is always unfair. A pool sized to a real 12-to-18-month hiring plan is a legitimate request. A pool sized to 15% because the term sheet template said so is a negotiation point.
How do SAFEs and notes sit in the table?
A SAFE has no shares until the next priced round, so it appears as a line item with a cap and sometimes a discount. The modeling error comes from ignoring it. Convertibles are real dilution that has not been booked.
With the standard post-money SAFE, the holder's percentage is fixed against the company capitalization just before the new round, including all converting SAFEs and the pool. A $500K SAFE at a $5M post-money cap converts into 10%. Later SAFEs dilute the founders and the earlier common, not the earlier SAFE holders. The terms are published by Y Combinator in its standard SAFE documents, which are worth reading once in full.
Three checks for any convertible on the table:
Is the cap pre-money or post-money? The same $5M cap gives a different conversion.
Does the discount or the cap apply, and which is lower at the round price?
How many SAFEs are outstanding in total? Founders often forget the small ones signed in the first months.
Which red flags show up in the table itself?
Most problems are visible in five minutes if you know where to look. What the analyst actually does on a Tuesday is read the table against the data room, line by line.
A departed co-founder with unvested or fully vested equity. Fully vested means they keep it. Ask what the leaver terms were.
Founders below roughly 50% combined before the Series A. There is no fixed threshold, but after a typical A the team can end up with a stake that no longer motivates them.
An option pool that is either empty or oversized. Empty means no hiring room. Oversized means a pre-money shuffle disguised as planning.
Advisor shares with no vesting schedule.
Mismatched totals. If the sum of holders does not equal the authorized and issued figure, something was issued off the books.
Preferred terms that stack. Multiple liquidation preferences or participating preferred from an earlier round change how proceeds split at exit.
The last one is outside the percentages entirely. A cap table shows ownership, not rights. The waterfall at exit depends on the preference terms, so ask for the shareholder agreement alongside the table.
How to build a pro forma before the term sheet
A pro forma cap table shows ownership after the proposed round. It takes about 20 minutes in a spreadsheet and it is the single most useful sheet in the IC folder.
The order of operations matters:
Start with the fully diluted share count before the round, including the current pool and every convertible at its conversion price.
Decide the target post-round pool as a percentage of the post-money fully diluted count.
Compute the new money as a percentage of post-money: investment divided by post-money valuation.
Solve for the total post-round shares, then back out new investor shares, pool top-up and the price per share.
Show each holder before and after, on the fully diluted basis only.
Run it twice: once with the pool pre-money and once post-money. The gap in founder ownership is your negotiation data. Then add the next round. A Series A at 20% dilution with a fresh 5% pool top-up takes you down by roughly a quarter. Model your own stake that way, not the headline.
For analysts who live in spreadsheets, a plain-English formula helper such as
Spreadsheet, CRM or a cap table platform?
For a first look, a spreadsheet is fine. Founders at pre-seed usually keep one, and you should be able to read it. The risk starts when the sheet is the only source of truth. Versions diverge, and nobody notices until a Series A lawyer rebuilds it.
Where does the cap table live on your side? Not in the CRM, in most funds. A relationship tool records that you hold a position and the entry details. It does not model the waterfall. Keep the pro forma attached to the deal record, so that there is no re-entry between the memo and the pipeline. Tools differ here: relationship CRMs such as Affinity store the stake and the deal context, while market-data platforms such as PitchBook supply round history and prior investors to cross-check the table against.
Where cap table basics stop being enough
Three things this guide does not settle. Cross-border structures: a Swiss AG, a German GmbH and a Delaware C-corp handle share classes, pre-emption rights and option grants differently, and the table format hides it. Secondaries: a founder selling shares at a round changes both the count and the signal. And tax: option exercise and vesting treatment varies by country, so a table that looks fine can carry a liability.
Treat the table as the entry point to diligence, not the end of it. The questions it raises, who owns what, who gets paid first, and how much is left for the team, are the ones the IC will ask anyway.
What to put in the IC memo about the cap table
Keep it to one short block. State the fully diluted ownership of founders, the pool size, outstanding convertibles with caps, and your pro forma stake after the round and after a projected next round. Flag anything that does not reconcile.
Do not paste the table. Partners want the three numbers that change the decision: founder ownership post-round, your exit-adjusted stake, and the open issues. The IC memo doesn't write itself, but this block can be drafted from the pro forma in ten minutes.
If you want deal context to sit next to the cap table without retyping, Request access to see how Accorata assembles the first-pass memo from sourced deals.