Risks and what’s actually built
Publishing this is the point. Read it before you put money in.
Liquidity, the one that actually kills cooperatives
Your capital is refundable, but it’s lent out to farms on one-year terms. If many members leave at once, the co-op cannot pay everybody. That is not a hypothetical; it is how co-ops fail.
The defences are a 20% liquidity floor, a 90-day notice period, and a redemption queue. Right now the co-op holds {cash} in cash against {capital} of member capital, {ratio}%, against a floor of $4,200.00. Above the floor. That number is on the fund page and it updates itself.
Your capital is not guaranteed and not insured
Stations can default. Equipment can be stolen or break. A loss reserve, target 10% of the loan book, funded from coffee margin, absorbs the first losses so member capital doesn’t. Beyond the reserve, your capital is at risk.
It is also nominal. $1,000 back in ten years buys what about $744 buys today. Capital at par is not capital preserved in real terms.
You are not investing, and you must not treat this as an investment
Member capital earns no interest, no dividend and no appreciation, and it is not transferable. Your return is coffee, which means it is worth nothing at all if you stop drinking coffee. If you want a financial return, buy an index fund, honestly, that is the correct advice.
The legal position is strong, not certain
Consumer cooperatives that redeem member shares at par are a well-worn non-security pattern, and this one is designed for it: par in, par out, non-transferable, one member one vote, benefit delivered as goods at cost rather than a distribution.
But co-op membership shares are not automatically outside securities law, several state co-op exemptions cap member capital, and the analysis turns on details, including the marketing copy. This build has not been reviewed by counsel.
Governance is a real obligation
One member, one vote means actual elections, actual meetings and actual minutes. Part of what makes this structure defensible is members genuinely controlling it. You do not get the benefit without doing the work, and none of that machinery exists yet.
Coffee supply is seasonal and can fail
Most origins harvest once a year, so there are gaps between batches. A lot can cup below spec, rain can ruin a drying week, and freight from a landlocked country is slow. When there is no coffee we say so rather than quietly substituting something else.
What is real and what is simulated
| Real | Not real |
|---|---|
| One real farm, Kelagur Estates, Chikmagalur, with a vendor quote on file | The other origins are illustrative |
| Postgres database, all business logic | The cooperative is not incorporated |
| Double-entry ledger, append-only, balance-enforced | No station has signed a contract yet |
| Spending is locked in code, a vendor cannot be paid without a deliberate unlock | No money has moved, to anyone |
| Liquidity floor and reserve, enforced in code | Coffee is not yet roasted or shipped |
| Stripe test-mode payments and idempotent webhooks | Mobile-money disbursement, logged, not sent |
| USSD endpoint on the real gateway contract | Farmer identities and delivery history |
| Landed-cost model with real duty and VAT mechanics | Vendor prices are researched estimates, not quotes |
| Sign-in by one-time emailed code, single-use, 10-minute expiry | Resend only delivers to a verified domain, other addresses fall back to showing the code |
The mobile-money payload we would post is constructed and stored in full. You can read it on the farmers page. A fake success animation would demo better and tell you less.