How to Run a Cap Table: 19.5% Median Dilution, the Option Pool Shuffle, and a Founder's Playbook

19.5% is the median equity founders give up at a seed round in 2026, and by the time a company closes its Series A, the founding team's ownership has fallen to a median of just 36%, according to Carta's 2026 Founder Ownership Report. That's the short answer. The longer answer is that most of the equity founders lose isn't lost in the headline valuation number — it's lost in the option pool math nobody explains clearly until the term sheet is already signed.

Running a cap table well is a mechanical skill, not a legal one — it means knowing what "fully diluted" actually includes, catching an option pool shuffle before you agree to it, and keeping your ownership numbers current every time a SAFE converts. Get it wrong and you find out at your Series B that you own far less of your company than you thought. Here's how the math actually works, stage by stage.

How to Run a Cap Table: The Four Things That Actually Matter

Running a cap table well means tracking fully diluted ownership, not just issued shares, so every SAFE, option grant, and warrant is counted against your true percentage before a round closes. The four things that determine whether you run it well are: knowing your fully diluted share count at all times, keeping vesting schedules current, modeling dilution before you negotiate a term sheet, and catching option pool changes before you sign — miss any one of these and you find out the hard way at your next raise.

Most founders think of their cap table as a record of what happened. It's actually a forward-looking model — the number that matters isn't how many shares exist today, it's what your ownership becomes after every SAFE converts and the new option pool gets carved out. A cap table tool that only shows current issued shares without a fully diluted, post-financing view is giving you a number that's already out of date the moment a term sheet lands.

Vesting is the other half of the equation founders underweight. A standard 4-year vesting schedule with a 1-year cliff protects the company if a co-founder leaves early, but it also means your cap table's "ownership" column and your actual, vested ownership at any given moment are two different numbers — and only one of them is real if you exit before your cliff. Founders who skip vesting on their own shares, thinking it only applies to employees, routinely regret it when an investor's counsel flags the gap during Series A diligence and asks for it to be fixed retroactively, usually on terms less favorable than if it had been set up correctly on day one.

How to Run a Cap Table at Each Funding Stage

Dilution compounds differently at every stage, and treating each round as an isolated 20% hit is the single most common cap table mistake. At pre-seed, founders typically give up 10-15% through SAFEs and a small initial option pool. At seed, the median dilution is 19.5%, and the most common outcome — occurring in 28% of seed and Series A rounds — is a founder selling between 20% and 24% of the company in one round. Series A dilution lands in a similar 19-20% band, but by then the option pool has usually already been topped up once, meaning founders absorb dilution twice: once from new investor shares, once from the pool refresh.

By Series B, the gap between AI and non-AI companies becomes visible in the data: the median AI founding team holds 27.3% of fully diluted equity at that stage, compared to 21.8% for non-AI teams — evidence that AI-focused rounds are currently pricing at higher valuations relative to dilution than the broader market. None of this is visible from a single round's term sheet; it only shows up when you model the full stack of rounds together, which is exactly what a properly maintained cap table is for. For more on how round sizes and valuations move together, see our benchmarking dashboard.

Illustrative post-Series A cap table structure blended from Carta's 2026 Founder Ownership Report medians and re:cap's cap table management guide. Actual ownership splits vary by round size and negotiated terms.

The Option Pool Shuffle: Where 8-12% of Founder Equity Disappears

The single most common way founders lose more equity than the headline dilution number suggests is the option pool shuffle: an investor requires a larger employee option pool be created or topped up before the round closes, and that dilution is calculated pre-money — meaning it comes entirely out of existing shareholders' pockets, not the new investor's. It typically costs founders 8-12% of their equity, and because it's baked into the pre-money valuation, the effective price per share founders receive is lower than the headline valuation implies.

The fix isn't refusing an option pool top-up — a well-funded pool is table stakes for hiring — it's negotiating the size of the pool and who bears the dilution before agreeing to a valuation. Model the post-money option pool size as a percentage of the round before you accept a term sheet, not after, since renegotiating pool size after a valuation is agreed is far harder than negotiating it as part of the same conversation. We've covered this mechanic in more depth in our piece on the option pool shuffle.

Tools You Need to Run a Cap Table Correctly

Spreadsheets fail as cap tables the moment a company has more than one class of preferred stock, since manually tracking conversion prices, liquidation preferences, and pro-rata rights across rounds is exactly the kind of error-prone math dedicated software exists to eliminate. Carta's Launch tier starts at $149/month with Growth at $599/month, and offers a free plan for up to 25 stakeholders — though the free tier excludes pro forma modeling and 409A valuation support, which most seed companies need within their first year.

Pulley is the cheaper alternative for early-stage companies, with a Startup plan around $1,200/year and 409A valuations turned around in 3-5 days on its Growth tier, compared to roughly 10 days on Carta. Carta generally wins once a company reaches Series A and needs combined cap-table, 409A, and fund administration functionality under one roof, plus broader acceptance among law firms handling the closing paperwork — a factor that matters more than most founders expect when a deal is racing toward signature. We've ranked the full field of options in our cap table software comparison.

The Mistakes That Cost Founders the Most Equity

Roughly 47% of founders don't model dilution scenarios before raising a round, which means they're negotiating valuation and option pool size in real time instead of walking in with a target ownership number already calculated. That single gap — not modeling dilution in advance — is the root cause of most of the other cap table mistakes founders make: failing to vest founder shares from day one, over-granting equity to early hires without a standard vesting schedule, and letting SAFEs pile up without updating the fully diluted count until a priced round forces the issue.

These mistakes aren't cheap. Cap table disputes that require legal cleanup — repricing errors, missing consents, unclear conversion terms — average $2.3 million in cost once law firm fees, re-issuance, and shareholder negotiations are factored in. The fix is procedural, not legal: update the cap table the same week any SAFE, note, or option grant is issued, run a fully diluted dilution model before every term sheet conversation, and use a tool that flags pro-rata and option pool changes automatically rather than relying on a founder to catch them in a redline. For how funding round sizes and dilution interact more broadly, see our VC performance dashboard.