The Price Isn't Known Yet. The Money Arrives Anyway.

If this is the first time you have heard of such an arrangement, the instinctive response is understandable: how can I invest before I even know my ownership?

The companies that need capital most can be the hardest to price

I only recently realised that quite a few people in business and technology — including some with investment experience — have never come across a SAFE. The acronym stands for Simple Agreement for Future Equity. The name sounds like a piece of financial machinery. The more interesting part is the problem it was designed around.

A conventional equity investment is intuitive: decide what the company is worth, then decide how much ownership a given investment buys. For a mature company, there is usually more evidence to work with — revenue, profit, assets, customers, cash flow and years of operating history.

COMPANY A · MATURE BUSINESSFifteen years of operating historyStable revenue and profit, customers and cash flow. The valuation is debatable, but there is a track record to work from.

COMPANY B · STARTUPThree months oldTwo founders, both PhDs from top universities; one prototype, no revenue, product possibly not launched. If it works, the market could be enormous.

Is Company B worth HK$5 million, HK$20 million or HK$100 million? At such an early stage, a plausible story can be built around almost any of those numbers. Yet investor and founder may spend months — sometimes quarters — negotiating before they converge on one.

The awkward part is that capital may be most useful at exactly this moment. The company needs to hire its first engineers, finish the product and test the market. The investor may already believe in the team and the direction. Both sides can have an answer to “do we want to work together?” while having no reliable answer to “what is the company worth today?”

Ownership matters, but a percentage never exists on its own

Startup conversations often fixate on the percentage: “How much do I own?” But a percentage has no value by itself. It must always be multiplied by something else: what the company is worth.

Tencent makes the contrast extreme but useful. Public market sources differed slightly on 17 August 2026, but placed its market capitalisation at roughly HK$4 trillion; 1% is therefore on the order of HK$40 billion. By contrast, 50% of a private company worth HK$2 million is theoretically worth only HK$1 million.

For a traditional business with slower growth and cash flows that are easier to estimate, an ownership percentage can map more directly onto economic value. For a very early startup, if the company value itself is highly uncertain, arguing whether the founder should sell 8%, 10% or 15% can simply turn one unknown into a percentage that looks more precise than it really is.

If the price is unknowable today, can the pricing decision wait?

The core idea of a SAFE is straightforward. The investor puts money into the company today, but the company does not have to complete a fully priced equity financing today. When a later priced equity round provides more information about the product, revenue, market and third-party investor demand, the SAFE converts into equity according to rules agreed in advance.

BACKGROUND · Y COMBINATORY Combinator (YC), founded in 2005, is a startup accelerator known for combining early investment with an intensive founder programme. Companies such as Airbnb, Dropbox and Reddit went through YC. In 2013, YC introduced the SAFE to reduce the legal paperwork and negotiation involved in early-stage fundraising through a shorter, more standardised instrument.

The money arrives today.
The price can wait for a future with more information.

That immediately creates another question. If one investor takes the risk a year earlier, why should that investor enter at exactly the same price as someone who arrives a year later with much more information?

Valuation Cap

One common answer is a Valuation Cap. It does not promise that the company is “worth no more than this”. It sets a maximum valuation basis used for converting that SAFE into shares.

Suppose you invest HK$1,000,000 through a SAFE today with a HK$20 million valuation cap. A year later, the company raises a priced round at a HK$40 million valuation, and a new investor also invests HK$1,000,000. In a deliberately simplified comparison, the new investor enters on the HK$40M valuation basis, while the earlier SAFE investor converts using no more than HK$20M — the same HK$1M maps to a more favourable share price for the investor who took the earlier risk.

A valuation cap is not the only way to structure a SAFE. Sometimes both sides are willing to complete the investment now but do not even want to fix the cap yet. That is where another term can appear: an uncapped MFN SAFE.

How does MFN — Most Favored Nation — work here?

MFN stands for Most Favored Nation. YC’s Uncapped MFN SAFE has neither a valuation cap nor a discount. If the company later issues another SAFE with more favourable terms, the earlier MFN investor may, subject to the document, elect to take the corresponding better terms. It deals with an even earlier layer of uncertainty: the investor is ready to commit capital before even fixing a cap.

SAFEs are not limited to pre-launch startups

SAFEs are strongly associated with seed-stage companies whose products or revenues are still forming, because that is when valuation can be especially fuzzy. But a SAFE is not defined by whether a product has launched. A later-stage company can also use one when it needs capital quickly while waiting for the next priced round, using the SAFE to ease short-term funding pressure. The underlying logic is the same: complete the financing now; leave part of the pricing decision for later.

Taking money before the product is finished is not as strange as it sounds

A SAFE can feel uncomfortable at first because we tend to imagine “investing in a company” as buying an asset that can already be priced. Step back and look instead at what is being rearranged — when capital arrives and when value becomes clearer — and the timing structure becomes much more familiar.

Kickstarter: sometimes the money is what makes the unfinished thing finishable

Kickstarter is a crowdfunding platform. Creators can present a project before the work or product is finished; supporters pledge first, and that capital helps turn the proposal into something real. Kickstarter explicitly describes itself as not being a store: backers are supporting a creative process rather than buying an existing product with guaranteed delivery.

This is not the same as a SAFE. Kickstarter backers do not receive equity. They typically receive a reward: perhaps early delivery of the product, an early-bird price, a special edition, access to an event or another project-specific promise. An “early-bird discount” also does not guarantee that the final retail price will turn out to be higher, because the product’s eventual pricing can still change, and delivery itself carries risk. A SAFE, by contrast, is a contractual right relating to future equity. The legal relationship and risk structure are very different.

But at the structural level they share something important: the thing may need resources today in order to have a better chance of existing tomorrow.

Presale flats: how Hong Kong turned “trade before completion” into a business method

Hong Kong’s population and housing demand rose quickly in the 1950s. Multi-storey housing expanded, yet home ownership still required a large amount of cash upfront. Developers faced the opposite timing problem: construction capital had to be committed early while sale proceeds arrived later. Selling units separately, allowing instalments and moving part of the transaction to before completion changed the cash-flow timing on both sides.

The key innovation behind the modern Hong Kong practice of selling uncompleted residential flats is widely associated with Henry Fok. The University of Hong Kong’s honorary-degree citation describes him as the first developer to sell residential blocks by instalments in the 1950s; the Chinese University of Hong Kong records that he introduced pre-sale of uncompleted residential flats. Hong Kong’s Land Registry also documents the rapid spread of separate-unit sales and instalment payment in the mid-1950s.

1953–54A deposit first, the rest in stages

Historical accounts place Fok’s innovation around this period: buyers paid a deposit before completion and the remaining amount in stages. Academic research places the earliest formal record of private residential presales in Hong Kong in 1954.

AFTERBoth sides’ cash flow was rewritten

For buyers, a single large payment was broken into several points in time. For developers, cash could begin returning before the whole building was completed. Broader forms of property presale may have earlier precedents elsewhere; the point here is the residential presale model that later became deeply embedded in Hong Kong.

That is why the comparison is useful. It is not simply that both arrangements “take money early”. Risk remains; the transaction rearranges when funding, delivery and pricing happen.

Sometimes the expensive part is not finding an investor. It is repricing the deal every single time.

A SAFE delays pricing. Another approach is to standardise part of the pricing instead.

BACKGROUND · 500 GLOBAL500 Global, formerly 500 Startups, is an early-stage venture firm and accelerator known for making many small early bets and running intensive founder programmes. Its published Flagship Accelerator terms have used a standard offer of US$150,000 for 6% equity, subject to the programme's terms and due diligence.

This is not a SAFE. One standardises part of the equity deal; the other postpones the equity price. Yet both reflect a similar design judgement:

If chasing the “most precise” valuation makes the company miss the moment,
transaction friction is itself a cost.

For a startup with only four months of cash left, spending eight weeks negotiating a valuation from HK$18M to HK$22M is not automatically better than closing an acceptable standardised financing in a few days. There is no universal answer; negotiation itself can create value. The less obvious point is that we rarely treat the time consumed by negotiation as part of the deal price.

Sometimes the answer exists. We are simply looking only inside our own industry.

Spend long enough inside one industry and it becomes easy to get locked into its habits. A familiar workflow slowly starts to look like the natural shape of the problem itself.

Cross-domain thinking does not mean copying another industry’s method wholesale. Finance, property and crowdfunding have different laws, ethics, risks and market structures. Many practices cannot and should not be transplanted. What is portable is often not the shell of the answer, but the abstract problem the answer was built to handle.

A SAFE is not simply a property presale; a property presale is not a Kickstarter project. But at this level, the connections become visible. Some things are unfinished. Some prices are still unclear. In some transactions, waiting until every uncertainty disappears means the resources arrive too late.

This is why HtPi places so much emphasis on cross-domain understanding. Not because knowing more fields makes a person look more learned, but because a problem stuck in one field may already have an analogue — and perhaps a useful structure — in a field that appears unrelated.

The next time you encounter a problem defended with “this is simply how our industry works”, it may be useful to step back and ask again:

Is this really a finance problem, a product problem or a legal problem — or is the deeper problem about time, information, risk and transaction cost?You do not need to import another industry's answer. First abstract the structure and ask whether these fields are dealing with the same kind of problem. The answer to one field's problem is often already hiding in a field we assumed had nothing to do with it.

Sources & notes

1. Y Combinator, “The SAFE — the open standard for startup fundraising” — SAFE definition, 2013 introduction, valuation caps and MFN terms.
2. Y Combinator, Press / About — YC background and accelerator model.
3. 500 Global, Flagship Accelerator — published accelerator investment terms and programme background.
4. Kickstarter Support, “What is Kickstarter?” — Kickstarter is not a store; backing involves creation and delivery risk.
5. The Chinese University of Hong Kong, honorary degree citation for Henry Fok — records his introduction of pre-sale of uncompleted residential flats in the 1950s.
6. The University of Hong Kong, honorary degree citation for Henry Fok — records his instalment-sale innovation.
7. Hong Kong Land Registry, history: the 1950s — documents the spread of separate-unit sales and instalment payment in the mid-1950s.
8. The Journal of Real Estate Finance and Economics, “What Motivates a Developer to Sell before Completion?” — early Hong Kong presale records and research on presales, financing and risk.
9. MarketCapWatch · StockAnalysis — Tencent’s market capitalisation on 17 Aug 2026 was roughly HK$3.95–4.01T; the article uses only an approximate HK$4T order of magnitude.

This article is for conceptual explanation only and does not constitute investment, legal, tax or fundraising advice. SAFE terms and conversion outcomes vary by jurisdiction, document version and actual capitalisation. Seek qualified professional advice for real transactions.