A SAFE feels like a settled deal the day you sign it. The real terms get written by the next round’s valuation
You raised on a SAFE. An angel wired the money, you signed a two-page document with a cap and maybe a discount, and you got back to building. As far as your bank account is concerned, the deal is done.
But nothing actually happened on your cap table. Not yet. A SAFE is a promise to issue shares later, on terms that depend on a number nobody knows today: the valuation of your next priced round. You signed the agreement, but you haven’t found out what it costs you. That part is still in the future, and it moves.
This is the thing founders most often miss when they use the SAFE calculator. They type in a cap, a discount, and an amount, and they want a single answer for “how much equity did I just give away.” There isn’t one. The honest answer is a range, and where you land in that range depends almost entirely on whether your next round comes in higher or lower than you hope.
So let’s walk through what a SAFE is actually made of, how it turns into shares, and why the down-round case is the one worth understanding before you sign anything.
What a SAFE is actually made of
A SAFE has three numbers and one piece of context.
The three numbers: an investment amount (what the investor puts in), a valuation cap (a ceiling on the price they’ll convert at), and sometimes a discount (a percentage off the next round’s price). The context is your cap table at the moment you sign, your own equity and your co-founders’, because that’s the base everything gets measured against.
None of this becomes shares until a later event: a priced equity round. When that round happens, the SAFE converts. The investor’s money turns into a number of shares, and that number determines how much of the company they end up owning, and therefore how much you and your earlier shareholders get diluted.
One clarification that prevents a lot of confusion: the cap is not a valuation. It does not mean your company is “worth” that amount. It’s a ceiling on the price per share the SAFE investor will pay when they convert. A $10M cap on a Series A that prices at $20M makes the SAFE convert as if the company were worth $10M, which hands that investor roughly twice the equity the headline round price would imply. A higher cap isn’t validation that you’re worth more. It’s a promise to grow into it. Two companies that both signed “a $10M cap” can end up with very different dilution depending on where their next round actually prices.
One more thing before we go further: this is the common case now, not a niche one. SAFEs were used in 92% of pre-seed rounds as of Q3 2025, having largely displaced convertible notes. If you’re raising your first money today, you’re almost certainly dealing with SAFE mechanics, which is why they’re worth getting right.
How conversion actually works
Here’s the rule that runs the whole thing, and it’s the rule founders most often get backwards.
When a SAFE converts, the investor’s price per share is the lower of two prices: the price implied by the cap, or the price implied by the discount. Lower price means more shares for the same dollars. More shares for the investor means more dilution for you. The instrument is built to give the SAFE investor whichever outcome is better for them, which is always the one that costs you more.
A worked example makes this concrete. On a $100,000 SAFE with an $8M cap and a 15% discount converting into a round priced at $0.909 per share, the discount gives a price of $0.77265 and the cap gives $0.72727. The cap price is lower, so the cap wins, the discount doesn’t apply at all, and the investor gets 137,500 shares. The discount was in the contract, but it never mattered. The cap did all the work.
That’s typical. The most common SAFE structure today is post-money with a valuation cap only, no discount, and when a discount is used, it’s 20% in 63% of those cases. So for most founders, the cap is the term that determines the outcome. (If you do use a discount, the common range is 10% to 25%, with 20% sitting as the market standard, lower for strong traction and higher for very early or higher-risk rounds.)
The “post-money” part matters for who absorbs the pain. A post-money SAFE fixes the investor’s percentage of the post-money cap table, which means the dilution from later money lands on you and your earlier holders rather than being shared with the SAFE investor. Most SAFEs today are this kind. So when you read the rest of this, assume the dilution we’re describing is yours.
The same SAFE across a range of outcomes
This is the part the calculators and the explainer pages tend to skip. They show you one conversion, at one assumed next-round valuation, and call it a day. But you don’t know your next-round valuation. The useful exercise is to hold the SAFE fixed and slide the next round up and down, and watch what happens to your stake.
Say you raised on a SAFE with a $10M cap, and you’re modeling a next round that brings in $5M of new money. The size of a small US seed or early Series A today. Where will it price? You don’t know. So look at the spread.
If the round prices well above your cap, say at $20M, the cap price binds hard. The SAFE investor converts at the $10M cap, meaning their price per share is roughly half the round price. A $10M-cap SAFE converting in a $20M round prices at $1.00 per share against the $2.00 round price, doubling the SAFE investor’s share count from 625,000 to 1,000,000. That sounds bad, but look at it the right way around. The new investors are paying full freight at $2.00. The company grew. Your stake, measured against everything, is in good shape: founder ownership in that scenario lands around 68.2% with the cap, versus about 70.6% if a 20% discount had governed instead. The cap cost you a couple of points relative to a discount, but you raised at a strong price and you kept most of your company. When the next round prices high, you keep more.
Now run it the other way. The round comes in at or below the cap. Things didn’t go to plan, the market turned, your numbers slipped, and the priced round lands at a valuation near or under $10M. Now the cap stops protecting you, because the cap was a ceiling on price and the price is already low. Conversion shifts toward the round price or the discount, the SAFE investor’s dollars buy a larger slice of a smaller company, and your relative dilution gets worse exactly when you can least afford it. This is the down-round case, and it’s the one to understand before you sign, because it’s the asymmetry nobody puts in the brochure: the cap helps your investor a lot on the upside and does little for you on the downside.
The asymmetry compounds if you’ve stacked SAFEs. Founders rarely raise just one. Adding a second SAFE with a $5M cap on top of the first drops founder ownership from around 68.2% to about 65.2%, because that lower-cap SAFE converts at an even cheaper price and takes its own bite. Each SAFE is reasonable on its own. Together, converting at once, they add up faster than you’d guess from looking at any single one.
Why a down round is uniquely dangerous for convertibles
There’s a reason we keep circling the low-valuation case. With a SAFE, there is no floor and no escape hatch. A SAFE converts on any equity financing with no minimum threshold, whereas a convertible note typically only auto-converts above a minimum raise, often $1M to $2M, and otherwise sits as interest-accruing debt to maturity. Different shape, but the same lesson: the danger zone for both instruments is the round that comes in weaker than the terms assumed. With a note you might owe money at maturity; with a SAFE you take outsized dilution. Either way, “the next round disappoints” is the scenario your paperwork is quietly betting against.
This is also why we’d push back on anyone trying to “derive” a SAFE cap from a clean discounted-cash-flow number for a pre-revenue company. The math doesn’t cooperate. Aggressive discount rates and early negative cash flows can mathematically produce a near-zero or even negative DCF value for an early-stage startup, which tells you nothing useful about what cap to set. At Equidam, the financial methods, two flavors of DCF and the Venture Capital Method, carry real weight only as a company matures and produces reliable numbers; early on, the qualitative methods that price risk do most of the work. A cap is a negotiated ceiling shaped by what you can defend, not an output you can compute from a spreadsheet.
Setting the cap: real fundraising step, or short bridge
So how do founders actually choose a cap? In our experience it comes down to which of two jobs the SAFE is doing.
The first job is the SAFE as a genuine fundraising step. You set the cap roughly at the valuation you expect to raise at next, and the SAFE is a real round in everything but format. The plan is that your next priced round prices well above that cap because you’ll have done the work to justify it. This is the standard case, and it’s where most of the dominant SAFE volume sits. It also tends to be the smaller end of the market: SAFEs rule the early, small rounds, and priced equity takes over as the cheques get bigger. Small and early leans SAFE; large and later leans priced.
The second job is the SAFE as a short bridge. Here you set the cap well below the valuation you’re targeting, usually because you’re already close to closing a priced round and just need a little runway to get there. The low cap is the price of speed: the bridge investor takes more equity per dollar in exchange for moving fast and taking on the risk that the round doesn’t close at all. That’s a deliberate trade, not a mistake, as long as you’ve named it as one.
The thing to avoid is setting a low cap because it felt easier to close, while telling yourself it’s the first kind of deal. A cap well below your eventual round valuation behaves like the bridge case whether you meant it to or not, and you’ll feel it when the SAFE converts.
If you want to go deeper on how to set the terms themselves, we’ve made the argument elsewhere that you should generally pick a cap or a discount rather than stacking both, and match the choice to your risk profile. This piece is the companion to that one: that article is about choosing terms, this one is about living with them at conversion.
What to do with this
A SAFE isn’t a settled deal the day you sign it. It’s a deal with one blank left in it, and the next round fills in that blank. Before you sign, model the conversion at three numbers, not one: the valuation you hope for, the valuation you’d accept, and the disappointing one. Look at where your ownership lands in each, and look at it again with every SAFE you’ve stacked converting at the same time. The disappointing case is the one that will surprise you, so make sure it doesn’t.
The point of doing this isn’t to arrive at a single “correct” number. It’s to walk into the conversation understanding the shape of what you’re agreeing to, so that when an investor proposes a cap, you can talk about it in terms of outcomes rather than vibes. That shared language, grounded in a defensible method rather than a gut feel, is most of what makes these negotiations go well.
If you want a structured starting point for the valuation that anchors all of this, you can run your own valuation on Equidam, built on a transparent, IPEV-aligned methodology and a base of 160,000+ valued companies and 30,000+ public-market comparables, across 90+ countries and 600+ industries. And when you start thinking past the SAFE, to how all this converts and stacks on a cap table over time, that modeling is exactly what our colleagues at Share Council work on.
One honest caveat: every number above is illustrative. The $10M cap, the $5M round, the specific cap-table percentages, those are examples to show the shape of the thing, not predictions about your company. Nobody has a precise figure for where your next round will price, including us. That uncertainty is the whole reason to model the range instead of trusting a single answer, because the range is the truth and the single answer is a guess wearing a suit.