Exit Waterfall Calculator: Why the Math Breaks Before You Even Open One (For Founders Outside Delaware)

An exit waterfall calculator takes a sale price and runs it down through your cap table in a fixed order: deal costs and debt first, then liquidation preferences by seniority, then each preferred holder's choice between taking their preference or converting to common, then whatever's left split across common, converted SAFEs, and vested options. The order is standard. What breaks it, for most founders outside the US, isn't the math — it's that the calculator assumes a single Delaware entity with a clean preferred-stock vocabulary, and your structure often isn't that.

Every waterfall guide ranking today walks through the same five-layer order correctly. None of them tell you what to do when your cap table spans a Kenyan OpCo and a Delaware HoldCo, or when your entity is a UAE mainland LLC that couldn't legally issue the preferred shares the calculator assumes until last year. That's the gap this article fills.

The order, briefly — then where it actually gets used

We've written separately about how liquidation preferences work and how to negotiate 1x-vs-2x, participating-vs-non-participating terms before you sign anything. This piece assumes you know what a preference is and focuses on what happens when you plug a real exit price into a real stack.

The five layers, in the order money actually moves:

1. Transaction costs and debt. Legal fees, banking fees, any bridge loan, accrued unpaid liabilities — off the top before any shareholder sees a cent.
2. Liquidation preferences, by seniority. Each preferred series gets its preference amount back, typically last-money-in-first-money-out, unless the term sheet stacked it pari passu with an earlier series.
3. The convert-or-take-the-preference decision. Non-participating holders compare their flat preference against what they'd get converting to common at their fully diluted percentage, and take whichever is bigger — automatically, not as a negotiation.
4. Participating preferred's double-dip, if any exists. A fully participating series takes its preference and a pro rata share of what's left, as if it also held common. Cooley's deal data puts this at roughly 5% of rounds now — the exception, not the default, but check your term sheets.
5. The common pool. Founders, employees with exercised options, and any converted SAFE holder split what remains, pro rata by fully diluted share count.

A simplified run: a company sells for $20 million. $800,000 in deal costs and a bridge loan come off the top, leaving $19.2 million. A Series A with a $4 million 1x non-participating preference takes its $4 million, leaving $15.2 million, unless converting to common pays more. A seed round with a $1 million 1x preference does the same math. What's left splits across common on a fully diluted basis. The arithmetic is mechanical; step 3, run correctly for every series, is where a spreadsheet built under deal pressure gets a sign wrong.

SAFEs don't sit cleanly inside this order — they sit above or beside it

A SAFE isn't preferred stock and isn't common stock until something triggers conversion, which is why every generic waterfall guide handles it vaguely. Y Combinator's standard post-money SAFE has an explicit change-of-control clause: at acquisition, the holder chooses between a cash payout equal to the greater of their original investment or their as-converted value at the valuation cap, or actually converting into shares. Most SAFE holders take the cash option when the deal price clears their cap comfortably — it settles faster and skips a last-minute share issuance.

If you've stacked multiple SAFEs at different caps — something we've walked through the dilution math for separately — each one runs this choice independently. A $200,000 SAFE at a $4 million cap and a $300,000 SAFE at a $6 million cap don't convert the same way at a $20 million exit; each investor compares their own cap against the deal terms. A waterfall that treats "the SAFEs" as one line item instead of running this per instrument will misstate who gets what once caps are far apart.

The option pool: the layer that isn't a shareholder yet

The unallocated portion of your option pool — shares reserved but not yet granted to anyone — isn't a shareholder and doesn't collect exit proceeds. Only vested, exercised (or cashlessly exercised at close) options sit in the common pool and get paid. That matters twice over. If your pool is 10% allocated on paper but only 6% is actually vested and granted at exit, the real common denominator is smaller than your summary suggests, so everyone else's share of the remainder is bigger. And what happens to unvested grants depends entirely on the plan's acceleration terms and the acquisition agreement — full acceleration, partial acceleration, assumption by the acquirer, or cancellation with no payout all show up in real deals, with no default you can assume without reading the documents.

Where the generic calculator actually breaks

Every waterfall calculator ranking today is built around a single entity with a Delaware-style share taxonomy: common, a clean stack of preferred series, options, done. That holds for a startup that's been a Delaware C-corp since incorporation. It doesn't hold for most of the founders reading this.

If you've done a flip, the waterfall runs against the HoldCo cap table, and every OpCo shareholder's position has to already be correctly mapped into HoldCo shares before the exit math means anything — the same two-entity problem African startups hit flipping to Delaware shows up flipping into Cayman or Singapore parents across Latin America and Southeast Asia. Run a waterfall against the wrong entity's cap table and the output is confidently wrong.

If you're a UAE mainland LLC, your entity couldn't legally issue the multi-class preferred structure a waterfall assumes until October 2025, when Federal Decree-Law No. 20 amended the UAE Commercial Companies Law to permit it. Before that, and still today for many rounds, the preferred stack sits in an ADGM or DIFC holding company instead — so the waterfall has to run against the free-zone holdco's share register, not the mainland LLC's.

If you're a Saudi company, an LLC — the default entity there — doesn't support issuing preferred or redeemable preferred shares with distinct bylaws-defined rights. A joint stock company or simplified joint stock company does, which is why growth-stage KSA startups typically convert to an SJSC ahead of a priced round specifically to make a waterfall like this one possible.

If your market has no native "preferred stock" concept — several African and Southeast Asian companies acts don't — preference terms live in a shareholders' agreement layered on ordinary shares instead of a distinct share class. A calculator expecting a "Series A Preferred" line item on your share register won't find one; the term exists contractually, and someone has to translate it into the waterfall by hand.

None of this means the order is wrong. It means the order only produces a correct answer when run against an accurate map of who holds what, in which entity, under which instrument — and that map is usually what's stale, not the waterfall logic.

Build the waterfall on top of a ledger you'd trust in a deal room

The failure mode that actually costs founders money isn't a wrong formula — it's running the right formula against cap table numbers nobody has reconciled since the last round closed: SAFE overhang tracked across three PDFs instead of a running total, an option pool whose "allocated" number was never checked against who actually vested, a HoldCo-OpCo exchange ratio that was correct the day the flip closed and has quietly drifted since.

Govy doesn't run exit waterfall calculations or model liquidity events — that's genuinely out of scope, and a banker or lawyer building your deal model should own that math. What Govy does is keep the inputs that math depends on correct as you go: SAFE and convertible note overhang on the same ledger as issued shares, an append-only event history so a preference term or option grant can't quietly change without a record, and ownership tracked across the instruments you actually hold, not the Delaware-default set a generic tool assumes. By the time an acquisition offer is on the table, the numbers feeding the waterfall should already be settled, not something your lawyer reconstructs from scratch under a two-week deadline. See how Govy tracks your cap table at govy.tech.

FAQ

What is an exit waterfall calculator?

It's a model that takes a sale price and runs it down through your cap table in a fixed order — transaction costs and debt first, then liquidation preferences by seniority, then each preferred holder's choice between taking their preference or converting to common, then whatever's left split across common shares, converted SAFEs, and vested options. The output is a payout number per shareholder, not a percentage — percentages lie about who actually gets paid at a given price.

What order do shareholders get paid in an exit waterfall?

Deal costs and any outstanding debt or accrued liabilities come off the top first. Then preferred stock gets paid out by seniority — usually last-money-in, first-money-out — according to each series' liquidation preference. After every preference is satisfied, whatever remains splits across common stock, converted SAFEs, and exercised options on an as-converted, fully diluted basis.

What happens to SAFEs in an acquisition?

A standard SAFE's change-of-control clause gives the holder a choice at closing: take a cash payment equal to the greater of their original investment or what they'd get converting into shares at their valuation cap, or actually convert and ride the common stock price. Most take the cash option when the deal price comfortably clears their cap — it settles faster and skips a share issuance days before close.

What happens to unvested stock options when a company is acquired?

That depends on the acquisition agreement and your plan's acceleration terms, not a universal rule. Common outcomes are full or partial acceleration written into the grant, the acquirer assuming the plan and re-issuing equivalent options, or unvested grants being cancelled with no payout — read your specific plan and the deal terms rather than assume any one outcome.

Can a UAE mainland LLC or Saudi LLC run a venture-style liquidation preference waterfall?

Not cleanly as an LLC in either jurisdiction — why most venture rounds in both markets have historically routed through a different entity. Saudi Arabia requires a joint stock company or simplified joint stock company to issue the preferred and redeemable preferred share classes a waterfall needs; the UAE only gave mainland LLCs the ability to issue multiple share classes with distinct liquidation rights in October 2025 under Federal Decree-Law No. 20, which is why ADGM and DIFC holding structures were the default workaround for years before that.