SAFE Notes Explained: Caps, Discounts, and Dilution
For founders handed their first SAFE: how conversion, post-money math, and stacked notes quietly reshape your ownership, and what to check before you sign.

A SAFE — Simple Agreement for Future Equity — is a short contract that gives an investor the right to shares in your company later, when you raise a priced round, in exchange for money now. It is not a loan and it is not stock on the day you sign it. It is a promise to issue equity on defined terms once a real valuation exists.
The instrument was created by Y Combinator and is published as a standard set of free documents on Y Combinator's SAFE documents page. Because the paperwork is standard and short, founders often sign one without fully pricing what they gave away. The terms that matter — the valuation cap, the discount, and whether the SAFE is pre-money or post-money — do not change how much cash you receive today. They change how much of your company the investor owns after conversion.
This guide explains what a SAFE is, how it differs from a convertible note, what each term does, and how a cap and a discount actually convert into shares. It uses one clearly labeled illustrative example so you can see the mechanics. This is general education, not legal, tax, or investment advice. Every real SAFE is a binding contract whose conversion depends on its own defined terms and your specific cap table, so model your own numbers and have a qualified startup lawyer review the document before you sign.
What a SAFE is — and what it is not
A SAFE sits in a strange middle state. The investor has paid you, but owns no shares and holds no debt. They hold a contractual right to receive shares at a future event — almost always your next priced equity round, and sometimes a sale of the company or a wind-down.
When Y Combinator introduced the SAFE in December 2013, it described the instrument as "not debt, but something more like a warrant," which is why, in its words, "there is no need to fix a term or decide on an interest rate" (Announcing the Safe, a Replacement for Convertible Notes). That single design choice drives everything else:
- No interest. The amount that converts is the amount invested, not principal plus accrued interest.
- No maturity date. A SAFE does not come due. It waits for a priced round for as long as that takes.
- No repayment right. If you never raise a priced round, a SAFE does not automatically become money you owe. It converts on the events written into the document, and outside those events the investor generally cannot demand cash back.
That is the whole appeal: a SAFE lets an early investor put money in before anyone can credibly price the company, and it defers the hard valuation question to a later round when there is more information. The cost is that "defer the valuation" is not the same as "avoid dilution." The dilution is simply set later, by terms you agree to now.
SAFE vs. convertible note: where the debt line falls
Founders often use "SAFE" and "convertible note" interchangeably. They solve the same problem — raising early money without setting a valuation — but they are legally different animals, and the difference matters if a round is slow to arrive.
A convertible note is debt. It carries an interest rate and a maturity date, and until it converts it is a loan sitting on your balance sheet. A SAFE is not debt, so it has neither. The practical consequences show up exactly when things go sideways: a note that reaches maturity before you raise has to be repaid, extended, or renegotiated, while a SAFE simply keeps waiting.
| SAFE | Convertible note | |
|---|---|---|
| Legal nature | Contract for future equity | Debt (a loan) |
| Interest | None | Yes, accrues until conversion |
| Maturity date | None | Yes — can come due before you raise |
| Sits on balance sheet as | Not debt | A liability |
| Converts at | Next priced round (and defined exits) | Next priced round or maturity |
| Typical failure mode | Waits indefinitely | Must be repaid or extended at maturity |
Neither is automatically "better." A note's maturity date can be a feature for an investor who wants a deadline that forces a conversation. A SAFE's open-ended patience is a feature for a founder who cannot predict when the priced round will land. What you should never do is treat them as identical when you model your cap table — the interest on a note compounds the investor's eventual ownership, and the maturity date is a real event on a calendar.
The four terms that quietly set your dilution
Most of a SAFE's economic weight lives in a few defined terms. If terms like fully-diluted shares and the option pool are new to you, read how startup equity works first, because a SAFE converts against exactly that share count.
Valuation cap. The maximum company valuation used to convert the SAFE into shares, no matter how high your priced round is later. A lower cap means the SAFE holder pays a lower effective price per share and gets more of the company. The cap is the single most negotiated number on the page.
Discount. A percentage reduction off the price per share that new investors pay in the priced round. A 20% discount means the SAFE converts at 80% of the round price. It rewards the early investor for going first.
Most-favored-nation (MFN). A clause that lets the SAFE holder adopt the more favorable terms of any later SAFE you issue before the priced round. If you give a subsequent investor a lower cap, an MFN holder can elect to take it too. Y Combinator publishes an "uncapped MFN" SAFE that has no cap and no discount at all — just this ratchet.
Pro rata rights. A separate right, often in a side letter, letting the investor put more money in at the priced round to maintain their percentage. It does not affect the conversion itself, but it affects how much of the next round they can claim.
A single SAFE usually carries a cap, a discount, or both. Y Combinator's current US templates come in three flavors: valuation cap with no discount, discount with no cap, and the uncapped MFN version (SAFE documents). When a SAFE has both a cap and a discount, the investor converts at whichever gives them more shares — never both stacked on top of each other.
Pre-money vs. post-money SAFEs: why YC changed the default in 2018
This is the distinction most founders miss, and it is the one that most affects how predictable your dilution is.
The original 2013 SAFE was a pre-money SAFE: the cap referred to your valuation before the SAFE money went in. The problem was that you could not know the SAFE holder's final ownership percentage until the priced round, because it depended on how much other SAFE money arrived in the meantime and on the eventual round price. Founders raising on several pre-money SAFEs were often genuinely surprised by their combined dilution.
In 2018, Y Combinator replaced it with the post-money SAFE, and that is the version it publishes today. "Post-money" means the SAFE holder's ownership is measured after all the SAFE money is counted — the SAFEs are treated as their own round — but before the new money in the priced round that converts them. The point of the change was certainty: both you and the investor can calculate, the moment you sign, how much of the company that SAFE will become.
On a post-money SAFE with a cap, that calculation is simple:
Ownership sold = investment ÷ post-money valuation cap
So a $500,000 SAFE at a $5,000,000 post-money cap converts into roughly 10% of the company, measured on a fully-diluted basis before the priced-round investors buy in. You know that percentage on day one, which you never truly did with the pre-money form.
The certainty cuts one way, though, and this is the trade-off founders underestimate. Because each post-money SAFE holder's percentage is locked, the dilution from everything that comes after — additional SAFEs and the priced round itself — falls on the founders and the option pool, not on the earlier SAFE holders. The investor's floor is fixed; yours is not. Understanding what actually happens in a seed round makes this concrete, because the priced round is the moment all of these fixed percentages finally come out of your ownership at once.
A worked example: how a cap and a discount actually convert
Here is a single illustrative example — the numbers are invented to show the mechanics, not a benchmark. Assume an angel invests $100,000 on a SAFE with a $5,000,000 valuation cap and a 20% discount.
Later you raise a priced Series A. To convert the SAFE, you compare two candidate prices per share and use the one that gives the investor more shares (the lower price):
- Discount price = round price per share × (1 − discount)
- Cap price = valuation cap ÷ the share count the SAFE converts against
Say the Series A prices new shares at $2.00 each, and the company has 10,000,000 shares on the fully-diluted basis the SAFE uses just before conversion.
| Scenario | Series A price/share | Discount price (−20%) | Cap price ($5M ÷ 10M shares) | Converts at | Shares for $100k |
|---|---|---|---|---|---|
| Big step-up round | $2.00 | $1.60 | $0.50 | $0.50 (cap wins) | 200,000 |
| Modest step-up round | $0.60 | $0.48 | $0.50 | $0.48 (discount wins) | 208,333 |
In the big step-up row, the cap wins because your company grew far past $5M, so the SAFE converts at $0.50 and the angel receives 200,000 shares. Those shares are worth $400,000 at the $2.00 round price — a 4× paper markup for the $100,000 they risked early. Had they instead waited and put the same $100,000 into the Series A at $2.00, they would have received only 50,000 shares. That gap is the cap doing its job: rewarding early risk.
In the modest step-up row, the round is priced close to the cap, so the discount produces the lower price and wins instead. The investor always converts at whichever is cheaper for them.
One honest caveat: the exact share count a SAFE converts against is defined inside the document, and the post-money SAFE's definition of company capitalization includes converting securities and the unissued option pool. The clean $0.50 above is a simplification to show direction. Your real conversion price depends on those definitions and your actual cap table, which is why you model it rather than eyeball it.
The stacking trap: how three SAFEs shrink the founders' slice
The post-money SAFE is easy to reason about one at a time, which is exactly what makes stacking several of them dangerous. Because each holder's percentage is fixed by the investment-over-cap formula, the percentages simply add up — and they all come out of your side of the table.
Walk through three sequential post-money SAFEs:
| SAFE | Investment | Post-money cap | Ownership locked in |
|---|---|---|---|
| SAFE 1 | $500,000 | $5,000,000 | 10.00% |
| SAFE 2 | $500,000 | $6,000,000 | 8.33% |
| SAFE 3 | $1,000,000 | $8,000,000 | 12.50% |
Those three notes sell roughly 30.8% of the company before you have raised a priced round at all, leaving founders and team with about 69.2% — and that is before the option pool expands and before the Series A investors take their slice. As the startup lawyers at Avisen Legal put it, "every additional SAFE issued dilutes the founders, not the existing SAFE holders," and "stacking three or four post-money SAFEs without a model is one of the more common ways founders arrive at a Series A with less ownership than they expected" (the dilution math founders miss).
The trap is not that any single SAFE is unfair. It is that each one looks small in isolation, the caps drift upward so each feels like a "better" deal than the last, and nobody keeps a running total. By the time a real round prices, the founders discover the combined bite. The defense is boring and effective: maintain one live model that shows total percentage sold across every outstanding SAFE, updated the day you sign each one.
When a SAFE fits — and when to push back
A SAFE is well suited to the earliest money, where speed matters and a formal valuation would be guesswork. It is the default instrument for accelerators, most angels, and a lot of pre-seed funding for startups, precisely because it lets a small check close in days without lawyers arguing over a price nobody can defend yet.
Reasons it works in your favor:
- Speed and cost. Standard document, few negotiated fields, low legal spend.
- No debt overhang. No interest accruing and no maturity date forcing an awkward conversation.
- Deferred valuation. You are not locked into a low price set before you have traction.
Reasons to slow down:
- You are raising a large or lead-driven round. Past a certain size, a priced equity round with a full set of investor protections is often cleaner than a pile of SAFEs, and sophisticated leads may prefer it.
- You have already stacked several SAFEs. Each new one adds directly to founder dilution. At some point a priced round resets the picture more honestly than a fourth note.
- The cap is low relative to your real progress. A cap negotiated when you had nothing can convert into far more ownership than intended once you have grown into it.
Treat the SAFE decision as one node inside the wider seed fundraising process, not a standalone form to sign. Fundraising is a distinct stage of building a company, and running it deliberately is part of what an operating system like 100 Tasks AI is for — its Pitch Coach helps you prepare for the investor conversations that lead to these terms. The terms of any specific SAFE, though, still belong in your own dilution model and in front of your own lawyer.
Model your dilution before you sign
A SAFE is not a formality. It is a priced claim on your company with the price expressed as caps and discounts instead of a share price. My rule for closing any round is to stay diligent all the way to the signature: the thrill of hearing yes is exactly the moment founders sign things they later regret. Applied to a SAFE, that means the moment you're most tempted to just sign and celebrate is precisely when you slow down and read the specific terms in front of you — the valuation cap, the discount, the MFN clause, pro rata rights — and keep control of your own diligence timeline instead of letting the excitement set the pace. Before you sign one, do the arithmetic that the standard document makes easy to skip:
- Confirm the form. Is it a post-money SAFE? Cap, discount, MFN, or a combination? Read the actual document, not the term-sheet summary.
- Compute the ownership sold. For a post-money SAFE with a cap, divide the investment by the cap to get the percentage this note alone will convert into.
- Add it to a running total. Sum the percentage across every outstanding SAFE so you can see the combined dilution, not one note at a time.
- Stress-test the priced round. Model conversion at a high round price and a low one, then layer in the option pool and the new investors, and look at the founders' final percentage.
- Have a lawyer review it. Confirm the exact capitalization definition, the MFN and pro rata language, and the conversion mechanics before you countersign.
The founders who are unpleasantly surprised at their Series A are almost never surprised by any single SAFE. They are surprised by the total they never summed. Build the model first, sign second.
Inside 100 Tasks AI, closing your seed round on a post-money SAFE is Task 67 in the SCALE stage, and the Pitch Coach Fundraising Skill is built to work through exactly these terms — cap, discount, MFN, pro rata — right alongside you.

Martin Bell
Founder of 100 Tasks. Martin Bell has launched or supported 120+ startups and turned Rocket Internet venture-building discipline into a step-by-step system used by 25,000+ founders and startups.


