SAFE Stacking: The Dilution Maths Founders Get Wrong
SAFEs feel cheap because you never see the dilution when you sign them. It arrives all at once, at the priced round, and it is reliably larger than founders expect. In this post we take a company that raises $3.4m across four post-money SAFEs at caps of $6m, $8m, uppercase;">TL;DR
2m and uppercase;">TL;DR
5m, then prices a $5m Ser
Author: Yanni Papoutsis · Fractional VP of Finance and Strategy for early-stage startups · Author, Raise Ready
Published: 2026-06-10 · Last updated: 2026-06-10
Reading time: ~10 min
What Is Driver-Based Revenue Forecasting?
A revenue forecast is a projection of the money your business will earn over a defined future period. There are two ways to build one:
Top-down forecasting starts with the total addressable market and works down to a market share assumption: “The UK B2B software market is worth £10 billion. If we capture 0.1%, we generate £10 million in revenue.” Useful for sizing the opportunity, useless for operational planning. Investors have heard thousands of 0.1% market share projections and are rightly sceptical.
Bottom-up, driver-based forecasting starts with the specific activities that generate revenue: “We have capacity to run 20 outbound sales conversations per week. Our conversion rate is 10%. Our average contract value is £12,000 per year. That gives us 2 new customers per week, or roughly 100 new customers per year, generating £1.2 million in new ARR.” Every assumption in that chain is testable, improvable, and explainable.
Driver-based forecasting is also the input layer for your 3-statement model — your revenue drivers feed the income statement, which integrates with the balance sheet and cash flow statement.
Why a Revenue Forecast Startup Needs a Different Approach
Established businesses forecast revenue by extrapolating historical data. Startups do not have historical data. The entire forecast must be built on forward-looking assumptions rather than trend lines. A driver-based model built on transparent assumptions is actually more useful to an early-stage investor than a statistical extrapolation, because it makes the business logic explicit and discussable.
The Core Framework: Identify Your Revenue Drivers
If you are still deciding between instruments, read SAFE vs convertible note first. This post assumes you have already raised on SAFEs and want to know what you actually gave away.
Why does SAFE dilution surprise founders so consistently?
Because the instrument is designed to defer the question, and deferring a question is not the same as answering it. When you sign a SAFE you agree an amount and a cap. You do not agree a share count, a price or a percentage. Nothing appears on your cap table. The money arrives and your ownership, on paper, is unchanged.
That is the whole appeal. It is also the trap. Every SAFE you sign is a promise to issue an unknown number of shares at an unknown future date, and you do not find out how many until the priced round, at which point it is entirely too late to negotiate.
There are three compounding effects, and each is individually modest and collectively brutal.
The first is that founders think in money rather than in percentages. Raising $500k on a $6m cap feels like a small transaction. In percentage terms it is 8.33% of the company, which is not small at all. Four such transactions, each individually reasonable, produce a total that would have been shocking as a single ask.
The second is post-money SAFE mechanics, which are more founder-hostile than the pre-money version they replaced and which almost nobody reads carefully. We will come to this in detail, but the short version is that a post-money SAFE holder's percentage is locked in and protected from dilution by later SAFEs. Someone has to absorb that dilution instead, and that someone is you.
The third is the option pool shuffle. The pool is almost always created from the pre-money, meaning existing shareholders pay for it and the new investor does not. Combined with SAFE conversion, this produces a circular calculation that founders' spreadsheets get wrong roughly half the time, always in the same direction.
How does a post-money SAFE actually convert?
A post-money SAFE with a valuation cap entitles the holder to a fixed percentage of the company, calculated as the SAFE amount divided by the cap, measured at conversion on a post-money basis that includes all other SAFEs but excludes the new priced round money.
That definition is a mouthful, so take it apart with a single SAFE.
You raise $500,000 on a post-money SAFE with a $6,000,000 cap. The holder's entitlement is $500,000 / $6,000,000 = 8.3333%. That percentage is fixed. Whatever happens between now and the priced round, that holder converts into 8.3333% of the company measured immediately before the new money comes in.
The word "post-money" refers to the cap being measured after all the SAFE money is in, not after the priced round. This is the source of endless confusion. A $6m post-money cap does not mean the company is valued at $6m after your Series A. It means the SAFE's percentage is computed against a $6m denominator that already includes the SAFEs themselves.
Contrast the older pre-money SAFE. Under a pre-money SAFE with a $6m cap, the price per share is set by dividing $6m by the pre-conversion share count, and the holder's resulting percentage depends on what other SAFEs convert alongside. More SAFEs meant everyone's percentage shrank, including the earliest holder's. The dilution was shared.
Under the post-money version, the first holder's 8.3333% is guaranteed regardless of how many SAFEs you sign afterwards. If you sign three more, they get their percentages too, and all of that dilution lands on the founders and the pool. The post-money SAFE moved the risk of the founder over-raising from the investor to the founder. That is not a criticism of the instrument, which is admirably clear about what it does. It is a criticism of founders signing it without doing this arithmetic.
What happens with four stacked SAFEs? The full worked example
Here is the company. Two founders, Sam and Delia, hold 6,000,000 shares between them, split 55/45, and own 100% of the company. There is no option pool yet.
| Founder | Shares | Ownership |
|---|---|---|
| Sam | 3,300,000 | 55.00% |
| Delia | 2,700,000 | 45.00% |
| Total | 6,000,000 | 100.00% |
Over eighteen months they raise four post-money SAFEs.
| SAFE | Amount | Post-money cap | Entitlement |
|---|---|---|---|
| SAFE 1 (pre-seed angel) | $500,000 | $6,000,000 | 8.3333% |
| SAFE 2 (angel syndicate) | $400,000 | $8,000,000 | 5.0000% |
| SAFE 3 (pre-seed fund) | $1,000,000 | $12,000,000 | 8.3333% |
| SAFE 4 (strategic) | $1,500,000 | $15,000,000 | 10.0000% |
| Total | $3,400,000 | 31.6666% |
Check each: 500/6,000 = 8.3333% ✓. 400/8,000 = 5.0000% ✓. 1,000/12,000 = 8.3333% ✓. 1,500/15,000 = 10.0000% ✓. Sum: 8.3333 + 5.0000 + 8.3333 + 10.0000 = 31.6666%.
Stop and look at that total. The founders have raised $3.4m and promised away 31.67% of the company before a single priced round. Nobody in the room ever said the words "we are selling a third of the company", and yet that is what happened, one reasonable increment at a time.
Notice also the per-dollar disparity. SAFE 1 paid $500,000 for 8.3333%. SAFE 4 paid $1,500,000 for 10.0000%. SAFE 1 paid $60,000 per percentage point. SAFE 4 paid $150,000 per point. The earliest money is 2.5 times more expensive per point than the latest, which is exactly as it should be, because the earliest money took the most risk. But founders rarely notice the magnitude until they see it laid out.
Check: $500,000 / 8.3333 = $60,000 per point ✓. $1,500,000 / 10.0000 = $150,000 per point ✓.
Now the Series A. A $5,000,000 investment at a $20,000,000 pre-money valuation, with a 10% post-money option pool required, created from the pre-money.
Post-money valuation: $20,000,000 + $5,000,000 = $25,000,000. Check: pre-money plus investment equals post-money ✓.
The Series A investor's ownership: $5,000,000 / $25,000,000 = 20.00%.
The option pool: 10.00% of the post-money company.
The SAFEs: 31.6666% of the company measured pre-new-money. This is the step that requires care, because the SAFEs' percentages are defined against the pre-money company, and the Series A and the pool are defined against the post-money company. You cannot simply add 31.6666 + 20.00 + 10.00 and subtract from 100.
How do you do the conversion arithmetic without getting it wrong?
Work in percentages of the final post-money company, and derive the SAFE percentages from their pre-money entitlements. The trap is mixing the two bases.
Here is the clean way to think about it. The Series A investor takes 20.00% of the post-money company. The pool takes 10.00% of the post-money company. Everything else, meaning the founders and all four SAFE holders, shares the remaining 70.00%.
Check: 20.00 + 10.00 + 70.00 = 100.00 ✓.
Now, within that 70.00%, how is it split between founders and SAFEs? The SAFE entitlements are defined as percentages of the pre-money company, which is exactly the company consisting of the founders plus the SAFEs, before the new money and before the pool. So within the pre-money company, the split is: SAFEs 31.6666%, founders 68.3334%.
Check: 31.6666 + 68.3334 = 100.0000 ✓.
That pre-money company as a whole is worth 70.00% of the post-money company. So each participant's final percentage is their pre-money percentage multiplied by 70.00%.
| Holder | Pre-money % | x 70% | Final post-money % |
|---|---|---|---|
| Sam | 37.5834% | 26.31% | |
| Delia | 30.7500% | 21.53% | |
| SAFE 1 | 8.3333% | 5.83% | |
| SAFE 2 | 5.0000% | 3.50% | |
| SAFE 3 | 8.3333% | 5.83% | |
| SAFE 4 | 10.0000% | 7.00% | |
| Option pool | - | 10.00% | |
| Series A | - | 20.00% | |
| Total | 100.0000% | 100.00% |
Derive the founder pre-money percentages. Founders together hold 68.3334% of the pre-money company, split 55/45 between them. Sam: 0.683334 x 0.55 = 37.5834%. Delia: 0.683334 x 0.45 = 30.7500%.
Check: 37.5834 + 30.7500 = 68.3334 ✓.
Now multiply through by 70%. Sam: 37.5834% x 0.70 = 26.3084%, so 26.31%. Delia: 30.7500% x 0.70 = 21.5250%, so 21.53%. SAFE 1: 8.3333% x 0.70 = 5.8333%, so 5.83%. SAFE 2: 5.0000% x 0.70 = 3.5000%, so 3.50%. SAFE 3: 5.83%. SAFE 4: 10.0000% x 0.70 = 7.0000%, so 7.00%.
Check the total: 26.31 + 21.53 + 5.83 + 3.50 + 5.83 + 7.00 + 10.00 + 20.00 = 100.00 ✓.
The founders own 26.3084 + 21.5250 = 47.8334% between them. Round to 47.83%, and carry the unrounded figure if you are doing anything else with it, because rounding at each step and then summing will drift.
The founders went from 100% to 47.83% having raised $8.4m in total across four SAFEs and a Series A. They gave up 52.17% of the company.
Where did it go? Series A 20.00%, option pool 10.00%, SAFEs 22.17%. Check: 20.00 + 10.00 + 22.17 = 52.17 ✓. And the SAFE total: 5.83 + 3.50 + 5.83 + 7.00 = 22.16, which is 22.17 carrying the unrounded values. ✓
Note that the SAFEs' 31.6666% pre-money entitlement became 22.17% post-money, because the SAFE holders are themselves diluted by the Series A and the pool. Post-money SAFEs protect their holders from dilution by later SAFEs, not from dilution by the priced round.
Now convert the percentages into actual shares
Percentages are the right way to think, but the documents issue shares, so you need the share count too. This is where the circularity bites.
The founders hold 6,000,000 shares and that number does not change. Those 6,000,000 shares must end up representing 47.8334% of the post-round company.
Total post-round shares: 6,000,000 / 0.478334 = 12,543,940 shares.
Check: 6,000,000 / 12,543,940 = 47.8334% ✓.
Now allocate everything else against that total.
| Holder | Target % | Shares | Check % |
|---|---|---|---|
| Sam | 26.3084% | 3,300,000 | 26.31% |
| Delia | 21.5250% | 2,700,000 | 21.53% |
| SAFE 1 | 5.8333% | 731,730 | 5.83% |
| SAFE 2 | 3.5000% | 439,038 | 3.50% |
| SAFE 3 | 5.8333% | 731,730 | 5.83% |
| SAFE 4 | 7.0000% | 878,076 | 7.00% |
| Option pool | 10.0000% | 1,254,394 | 10.00% |
| Series A | 20.0000% | 2,508,788 | 20.00% |
| Total | 100.00% | 12,543,756 | 100.00% |
The share total comes to 12,543,756 rather than 12,543,940, a gap of 184 shares, because each line was rounded down to a whole share. In a real financing the lawyers allocate the rounding remainder explicitly and the numbers tie exactly. The point of showing it is that a 184 share discrepancy on 12.5m shares is fine; a 184,000 share discrepancy means you have made a real error.
Verify the individual lines. SAFE 1: 12,543,940 x 0.058333 = 731,730 ✓. SAFE 4: 12,543,940 x 0.07 = 878,076 ✓. Pool: 12,543,940 x 0.10 = 1,254,394 ✓. Series A: 12,543,940 x 0.20 = 2,508,788 ✓.
And the price per share for the Series A: $5,000,000 / 2,508,788 = $1.9930 per share.
Sanity check that against the pre-money. The pre-money valuation was $20,000,000 and the pre-money share count, meaning everything except the Series A shares, is 12,543,940 - 2,508,788 = 10,035,152 shares. Price: $20,000,000 / 10,035,152 = $1.9930 per share ✓. The two agree, which confirms the whole structure. If your price per share computed from the pre-money does not match your price per share computed from the investment, you have made an error somewhere in the pool or the SAFE conversion.
That check is the single most valuable habit in this post. It takes ten seconds and it catches nearly everything.
Why does the option pool make this so much worse?
Because it comes out of the pre-money, which means the founders and the SAFE holders pay for it and the Series A investor does not, despite the pool existing to hire the people who will build the company the Series A investor just bought into.
Look at what happened above. The pool is 10% of the post-money company. The Series A investor holds exactly 20.00%, which is precisely what they paid for: $5m of a $25m post-money. The pool did not come out of their 20%. It came out of the 80% that everyone else was sharing, reducing what would have been an 80% share for founders-plus-SAFEs down to 70%.
Run the counterfactual. With no pool at all, the Series A takes 20.00% and the pre-money company keeps 80.00%. Founders' share: 68.3334% x 0.80 = 54.6667%. With the 10% pool: 68.3334% x 0.70 = 47.8334%.
Check: 54.6667 - 47.8334 = 6.8333 percentage points. The pool cost the founders 6.83pp, and it cost the SAFE holders 31.6666% x 0.10 = 3.17pp. Together: 6.83 + 3.17 = 10.00pp, which is the whole pool ✓. The Series A investor contributed zero to it.
This is the pool shuffle, and it is negotiable. The two things to argue for are pool size and pool timing. On size, ask the investor to justify the number against an actual hiring plan for the next 18 months rather than accepting a round figure. If you genuinely need 7% rather than 10%, the difference is worth 3% x 0.683334 of your ownership, which is 2.05pp, and at a $25m post-money that is $512,500 of value. Pools are commonly 10 to 20% post-money, and the top of that range is often lazy rather than analysed.
Check: 3.00 x 0.683334 = 2.05pp ✓. 2.05% of $25,000,000 = $512,500 ✓.
On timing, the fair position is that the pool should be sized for the hires the new money will fund, and any unallocated pool left over from the previous round should count towards it. Investors will often propose a fresh pool on top of an existing one, which double-charges you. Our term sheet explained guide covers where to push.
What should you actually do about all this?
Three habits, none of which require a finance background.
First, track your cumulative SAFE percentage as you sign, not at the priced round. Keep one cell in one spreadsheet that sums amount divided by cap for every SAFE outstanding. If that cell says 31.67%, you know exactly what you have promised before you sign the next one. This is a five minute task and it is the single highest-return thing in this post. The founders in our example would almost certainly have priced SAFE 4 differently had they been looking at 21.67% on screen when they signed it.
Second, model the priced round before you need it. Take your current SAFE stack, assume a plausible Series A at a plausible pre-money with a 10% pool, and run the arithmetic above. If the answer is that you end up below 50%, that is not necessarily a disaster, but it should be a decision rather than a discovery. It also tells you something about your next raise: your financial model should carry the SAFE stack as a live input, and you should be able to see how much runway the money buys you against how much of the company it costs, which is the trade-off the runway and burn calculator helps you frame.
Third, understand that raising more on SAFEs at a higher cap is not free. Founders often reason that a higher cap means less dilution, which is true per dollar, but the total is what matters. Raising an additional $1m at a $15m cap costs 6.67% of the company. Whether that is a good trade depends on what the money does, and the honest answer is that if it does not clearly accelerate you to a materially better Series A, it is not worth it.
A final structural note. SAFEs with different caps and different discounts, MFN clauses that retroactively give early holders the best subsequent terms, and pro rata side letters all interact in ways this example deliberately simplified away. If your stack includes MFN provisions, your effective entitlement may be higher than the sum of amount over cap, because the MFN holders may reprice to the lowest cap in the stack. Resolve every instrument's actual terms before you model, and note that anti-dilution protections in the priced round add another layer if a later round is down. For the difference between the pre-seed and seed conventions that produce these stacks in the first place, see pre-seed vs seed.
This is educational content, not legal or financial advice. SAFE conversion mechanics depend entirely on the specific document you signed, and the various SAFE templates in circulation behave differently. Have your counsel model the conversion against your actual instruments before you rely on any number.
Frequently Asked Questions
Do SAFEs convert at the cap or at the discount? Whichever produces a better outcome for the SAFE holder, which is whichever gives them more shares, meaning the lower effective price. If your SAFE has both a $6m cap and a 20% discount, and you price a Series A at a $20m pre-money, the cap almost certainly wins because the cap implies a much lower conversion price than a 20% discount off $20m. The discount only becomes relevant when the priced round happens at a valuation close to or below the cap. In practice, for companies that make good progress, the cap is what matters and the discount is decoration. Model both and take the lower price, because that is what the document instructs.
What is the difference between a pre-money and post-money SAFE in one sentence? A post-money SAFE fixes the holder's percentage of the company so that later SAFEs dilute the founders rather than the earlier SAFE holders, while a pre-money SAFE lets each new SAFE dilute all the previous ones alongside the founders. The post-money version is now the more common standard and it is meaningfully more expensive for founders who raise multiple times before pricing a round. If you sign four post-money SAFEs, you bear all of the dilution from SAFEs two, three and four yourself.
Does the option pool have to come out of the pre-money? No, but it usually does, and changing that is a real negotiation rather than a technicality. Putting the pool in the post-money, so that the new investor shares in paying for it, is a meaningful economic ask and most investors will resist it because the market convention runs the other way. The more winnable fight is the size of the pool: bring a genuine hiring plan showing the roles you will fill before the next round and the equity each requires, and argue the pool down to that number. An investor who wants 15% and cannot justify it against a plan will often settle at 10% when challenged with actual arithmetic.
Should I just raise a priced seed round instead of SAFEs? It depends on how much you are raising and how quickly. SAFEs are genuinely cheaper and faster for small, rolling amounts, and for a first $500k from angels they are almost always right. The problem is the stack: SAFEs are ideal for one raise and increasingly poor for four. If you can see yourself raising more than roughly $2m before a priced round, the case for pricing the round gets strong, because a priced round puts the dilution on the cap table where you can see it, sets a clean price, and stops the entitlement percentages accumulating invisibly. The instrument is not the problem. Serial use of it without tracking the total is.
How do I explain to a SAFE holder that they own less than they thought? Show them the arithmetic early rather than at conversion. Post-money SAFE holders often believe their percentage is fully protected, and it is, but only against later SAFEs. They are diluted by the priced round and by the option pool exactly like everyone else, so a holder with an 8.3333% entitlement ends up at 5.83% after a Series A with a 10% pool. That is not a breach of anything, it is how the instrument works, but the conversation is much easier when you have flagged it in an investor update six months earlier than when it appears in a closing statement they were not expecting.
Further Reading
Get the complete guide with all 16 chapters, exercises, and model templates.
Get Raise Ready - $9.99