You've now met every piece separately: what a cap table is, common versus preferred, vesting, and option pools. This capstone shows you how to read a cap table the way an investor does: line by line, then as a story. We'll walk one realistic post-seed table in full (founders, two converted SAFEs, a preferred lead, employee options, an advisor, the pool) and finish with the red flags diligence teams look for.
What a cap table actually shows
A cap table is a list of who holds which securities in the company, and how much of the company that adds up to. Each row is a holder of one security type; the columns you'll almost always see are the security class, the share count, and an ownership percentage.
The single most important reading skill is knowing which denominator a percentage is using:
- Issued and outstanding counts only real shares that exist today: founder stock, preferred stock, exercised options. This is the denominator for votes.
- Fully diluted (FD) adds everything that can become a share: granted options (vested or not), warrants, and (by the most common convention) the entire authorized option pool, allocated or not. This is the denominator investors price rounds on and the one that tells you what your stake is worth at an exit.
Always ask: "percent of what?" The same 900,000 shares can be 8.1% of issued stock and 7.2% fully diluted. Neither number is wrong; they answer different questions. A table that shows percentages without stating the denominator, or that mixes denominators between rows, is the first thing to fix before reading anything else into it.
The table we'll read
Here's the fully diluted cap table of Meridian Devices, Inc., a fictional but realistic startup a few months post-seed. The history in one line: two founders incorporated with 8,000,000 shares, raised two pre-seed SAFEs, then closed a 1.5M seed at 0.80 per share, converting both SAFEs and creating an option pool in the same close.
| Holder | Security | Shares | % fully diluted |
|---|---|---|---|
| Founder A (CEO) | Common | 4,400,000 | 35.2% |
| Founder B (CTO) | Common | 3,600,000 | 28.8% |
| Orchard Ventures (seed lead) | Series Seed Preferred | 1,875,000 | 15.0% |
| Angel 1 (converted SAFE) | Series Seed-1 Preferred | 900,000 | 7.2% |
| Angel 2 (converted SAFE) | Series Seed-2 Preferred | 375,000 | 3.0% |
| Employees (3 grants) | Options (common) | 350,000 | 2.8% |
| Advisor | Options (NSO) | 100,000 | 0.8% |
| Unallocated option pool | Reserved | 900,000 | 7.2% |
| Total fully diluted | 12,500,000 | 100.0% |
First pass, before reading any single line: the percentages sum to 100.0%, the share counts sum to the stated total, and every row names its security type. Trivial-sounding, yet plenty of spreadsheet cap tables fail one of the three, and until the arithmetic reconciles, nothing else on the table can be trusted.
Reading it line by line
Rows 1–2: the founders. Common stock, 8,000,000 shares between them, 64.0% fully diluted, split 55/45. What the summary table doesn't show, and what you must check in the underlying records, is whether these shares carry reverse vesting. Post-seed, a standard setup has each founder about a year into a 4-year schedule, with the company holding a repurchase right on the unvested balance.
Row 3: the seed lead. Orchard Ventures holds Series Seed Preferred: real preferred stock, bought with cash at the round price of 0.80. Preferred means this row carries rights the common rows don't: a liquidation preference (1x, non-participating, in a typical seed), protective provisions requiring their consent for key actions, and usually a board seat. Their 15.0% cost 1.5M, which pins the round's post-money value at 12,500,000 shares times 0.80, i.e. 10M.
Rows 4–5: the converted SAFEs. These two angels wrote cheques long before the seed, on SAFEs, which sat off the cap table as contracts until the priced round converted them (see SAFEs and convertibles). Notice the sub-series: Seed-1 and Seed-2, not plain Series Seed. Each SAFE converted at its own cap-derived price, so each gets its own "shadow series": identical rights, but a liquidation preference equal to what that holder actually paid rather than the round price. That detail keeps preference stacks honest at exit (more in liquidation waterfalls).
Worked example: reading one line completely (Angel 1).
Angel 1 invested 360,000 on a pre-seed SAFE with a 5M post-money cap. At the seed, the cap set their conversion price at 0.40, half the 0.80 the seed investors paid. Shares: 360,000 ÷ 0.40 = 900,000. Ownership: 900,000 ÷ 12,500,000 = 7.2% fully diluted. Marked at the seed price, the stake is worth 900,000 × 0.80 = 720,000, a clean 2.0x paper markup for taking pre-seed risk. And because Seed-1 is a shadow series, its liquidation preference is 360,000 (what was paid), not the 720,000 the round price would imply. One row, four facts: cost basis, conversion mechanics, current stake, and downside protection.
Angel 2's line reads the same way: 240,000 invested at an 8M cap, converting at 0.64 into 375,000 shares, 3.0% and a 1.25x paper markup. Same round, same rights, different price: that's the SAFE cap doing its job.
The pool, the employees, and the advisor
The bottom three rows are all one mechanism seen from different angles. The seed term sheet required an option pool of 1,350,000 shares (10.8% of the post-round total). Since the close, 450,000 of those have been granted: 350,000 across three employee option grants and 100,000 to an advisor as NSOs. The remaining 900,000 sit unallocated.
Three reading notes. First, none of these 1,350,000 are issued shares (options only become stock when exercised), yet all count in the fully diluted total, which is why the pool dilutes everyone from day one (the option pools lesson covers who paid for it). Second, "Employees (3 grants)" is a summary row; the real table behind it lists each grant with its own schedule, cliff, and strike price. Third, the advisor's 0.8% sits inside the normal 0.25–1% advisor range; an advisor at, say, 4% would be a question worth asking.
Reconstructing the dilution story
A cap table is a photograph, but the negatives are all still in the file. Reading the history out of the current numbers is the skill that separates skimming a table from understanding it.
Worked example: where the founders' 36 points went.
At incorporation, the founders held 8,000,000 of 8,000,000 shares: 100%. The two SAFEs raised 600,000 of cash but changed nothing on the table, since SAFEs are invisible until conversion. Then one seed close did all of the dilution at once, growing the denominator from 8,000,000 to 12,500,000:
SAFE conversions: 900,000 + 375,000 = 1,275,000 shares → 1,275,000 ÷ 12,500,000 = 10.2 points. New money: 1,875,000 shares → 15.0 points. Pool creation: 1,350,000 shares → 10.8 points. Total: 36.0 points, taking founders from 100% to 8,000,000 ÷ 12,500,000 = 64.0%.
Notice that only 15 of the 36 points bought new cash at the round price. The other 21 were commitments made earlier (the SAFEs) or negotiated into the pre-money (the pool). That decomposition, cash versus conversion versus pool, is the first thing to compute on any table you're handed. Dilution math generalizes it.
On a spreadsheet, reconstructing this means digging through old versions and emails. If the table lives in software with point-in-time snapshots (Vquity, for instance, writes an automatic pre-close snapshot at every round and can diff any two snapshots field by field), the same history is a two-click comparison instead of an archaeology project.
Who controls what
Ownership percentages and control are related but not the same, and the table gives you the raw material for both.
Votes run on issued shares, not fully diluted ones. Meridian's issued count is 8,000,000 common + 3,150,000 preferred (all three preferred rows) = 11,150,000. The founders control 8,000,000 ÷ 11,150,000 ≈ 71.7% of the vote, comfortably above a majority, while the investor bloc holds about 28.3%. Option holders don't vote at all until they exercise.
But raw voting share overstates founder control. Protective provisions give the preferred a class veto over specific actions (selling the company, raising a new round, changing the charter) regardless of their percentage. And board control is set by contract: a typical post-seed board here is two founders plus one Orchard seat. Reading a cap table for control therefore means reading three layers: share votes, class vetoes, and board seats.
Worth noticing on the other side: no single non-founder holds more than 15%, and the founders keep majority control even on a fully diluted basis (64%). At post-seed, that's what healthy looks like.
Red flags investors look for
Diligence teams read hundreds of tables, and the same problems recur. When you know how to read a cap table, you can spot them in minutes:
- Dead equity. A departed co-founder still holding a large fully vested stake, say 20% held by someone who left in year one, is capital the company gets no work for, and it crowds out the pool and future rounds. Investors often make cleaning it up (a buyback or restructuring) a condition of the round.
- Missing founder vesting. If the founders' 64% has no reverse vesting, nothing keeps either founder at the company post-investment. Expect any lead to require a vesting reset before wiring money.
- An oversized, idle pool. A 25% pool with 3% granted signals dilution taken far too early and no real equity plan, and it quietly lowered the price of every round that included it.
- A messy SAFE stack. Six SAFEs at scattered caps and discounts, some with MFN clauses, none modeled, so nobody, including the founders, knows what the post-conversion table looks like. Meridian's two-SAFE stack converted cleanly; a deep stack often doesn't (see SAFE stacking and dilution).
- A table that doesn't reconcile. Percentages that don't sum to 100, share counts that don't match the signed documents, verbal promises of equity that appear nowhere. If the spreadsheet and the paper trail disagree, diligence stalls until they agree.
More war stories in common cap-table mistakes. Most of them are one of these five in costume.
That's the whole skill: check the arithmetic, read each line for its class and rights, reconstruct the dilution story, map control across votes, vetoes, and board seats, then scan for the five flags. And that's the end of Cap Tables 101. Ready for more? Fundraising Math starts with SAFEs and convertibles, the instruments behind rows 4 and 5 of the table you just read.