You sold it. It doubled four months later. You still remember the exact number. Everyone who has held anything for long enough has this story, and the detail people get wrong when they tell it is the part about being wrong. You were not wrong. Your thesis survived. What happened is that you needed cash on a Tuesday and your portfolio was the only place it could come from. That is a completely different failure from a bad call, and it deserves a completely different fix, but almost nobody treats it that way. We talk endlessly about how to pick well and almost never about what to do when picking well is not the constraint.
Watch what people actually do when the money is needed. Rent is due, a car dies, a medical bill lands, or an opportunity appears with a deadline attached. There is no facility to draw on, no line of credit against what they already own, and no patience left. So they open the app and sell the most liquid thing they hold, which is usually also the thing they had the most conviction in, because conviction and liquidity tend to live in the same positions. That single click does three separate kinds of damage. It ends the upside, which is the loss people notice immediately. It resets a position they spent months building conviction in, which is the loss they notice a year later. And in many jurisdictions it triggers a taxable event, which is the loss they notice in April. Now look at what people with more money do in exactly the same situation. They do not sell. They borrow against what they hold and leave the position intact. This is not a secret and it is not exotic. It is a standard service, offered quietly, to people who already have enough. The mechanism has existed for decades. Access to it has not, and that gap is the entire reason Pillar exists.
Put real numbers on the difference. Say you hold ten thousand USDG worth of tokenized Apple. Selling gives you ten thousand and leaves you holding nothing, and if the position rises thirty percent over the following year, that is three thousand you were right about and did not receive. Pillar gives you four thousand USDG without selling anything, because Apple's market carries a maximum loan to value of forty percent and a ten thousand position supports a four thousand borrow while the collateral stays yours the entire time. Then something happens that does not happen with an ordinary loan. Your collateral is not sitting idle while it backs the debt. It is routed to a yield source, and that yield is applied directly to what you owe. At an illustrative eight percent annual yield on ten thousand of collateral, roughly eight hundred a year is generated, Pillar takes a ten percent cut of that yield, and the remaining seven hundred twenty goes against your four thousand debt. Run it forward and the debt falls to roughly three thousand two hundred eighty after a year, around two thousand five hundred after two, and reaches zero somewhere in the fifth or sixth year, faster if the collateral appreciates, because a larger collateral value generates more yield. You never made a payment. You never had a due date. Time paid it.
The numbers we care most about, though, are the conservative ones. Apple and Microsoft borrow at a maximum forty percent loan to value, Nvidia at thirty five, Tesla at thirty, and a broad market ETF at fifty because it is a basket rather than a single company. Those are deliberately unexciting, and the reason is a problem that lending against stocks has and lending against crypto does not. Stock markets close. Your debt does not. A tokenized equity follows a market that shuts on Friday afternoon and does not reopen until Monday morning, while your loan stays live every second in between. On Monday that stock can open fifteen percent below Friday's close because of an earnings miss, a guidance cut, or a headline that landed at nine at night, and there was no window in which anyone could have traded out of the way. There is no clever mechanism that solves this. Not a faster oracle, not a prediction model, not a dynamic curve. The gap is genuine risk manufactured by a market that is closed, and the only honest response is to not lend as much against it in the first place. That is why those loan to value numbers look boring. Boring is the feature. A thirty percent loan on a stock that gaps fifteen percent is still comfortably safe. A seventy percent loan on the same stock is a liquidation waiting for a Monday.
The same instinct governs how the protocol behaves when its information is bad. If a price feed goes stale, a system that keeps transacting is letting people act on a number nobody can defend, so Pillar blocks new borrows and collateral withdrawals whenever a price is stale, because those are precisely the actions that could exploit a wrong number. It never blocks repayment, deposits, or the yield harvest, because those actions only ever improve your position, and a user should never be locked out of making themselves safer. Refusing to act on a number you cannot defend costs convenience and buys the only thing that matters.
The product itself is simple to describe once those constraints are understood. You deposit tokenized stock, and it becomes collateral that is immediately forwarded to a yield source rather than sitting inert in a vault. You borrow USDG against it, up to that market's maximum loan to value, drawn from a treasury of USDG the protocol holds. There is no repayment schedule attached because there is no interest accruing. Pillar does not charge interest on the loan at all. The protocol's revenue is the cut of the yield your collateral produces, which means Pillar only earns when your collateral is working, and our incentive and yours point the same direction. Anyone can call harvest on a position, and that function pulls the yield the collateral has accrued, takes the protocol's ten percent, and applies the rest to your debt, emitting a record of how much was repaid and what remains. If you have no debt at that moment, the yield is credited to you as claimable USDG rather than disappearing.
Here is the honest limit, and it belongs in the middle of the pitch rather than buried at the end. If the yield rate goes to zero, your debt stops shrinking. It does not grow, because nothing is accruing against you, but it does not vanish either. A self repaying loan repays itself at the speed the collateral earns, and no faster. Anyone describing one without saying that is selling you a brochure rather than a product. The second limit is liquidation, which exists because pretending otherwise would be a lie about a lending protocol. But it is partial by construction. When a position becomes unhealthy the contract computes the smallest repayment that restores it to health and refuses anything larger, so a liquidator attempting to seize more than that has their transaction reverted. You lose a slice, not the position. That is a mathematical constraint enforced in the contract, not a policy we promise to follow, and the distinction matters on the day it is tested.
You should also never arrive at that day by surprise. Your health factor, your current loan to value, how much yield has already gone toward your debt, the estimated time until it reaches zero, and the percentage the price would have to fall before liquidation are all on the dashboard and updating continuously. A protocol that hides your distance to failure is not protecting you from anxiety. It is protecting itself from your questions.
What you get from all of this is four things worth stating plainly. You get cash without a sale, so the position stays open, the upside stays yours, and the tax event a sale would have triggered does not happen. You get a loan with no schedule, no monthly payment, no maturity date, and no refinancing conversation, where the only thing that changes month to month is that you owe less than you did. You get an asset that works while it is pledged, which is unusual, because collateral in most lending protocols is dead weight locked up and earning nothing, whereas here it is the thing paying off the loan it secures. And you get the option wealthy investors have always had, on a portfolio of whatever size you actually have, with the parameters written into a contract rather than negotiated by a private banker.
What exists today is the core: six markets, a working self repay engine, partial liquidation, and a dashboard that tells you the truth about your position. That is the foundation and not the destination. We are widening the collateral shelf with more individual names and more ETFs, so a broader thesis can be borrowed against rather than a single ticker at a time. The yield source gets deeper too, spreading collateral across several venues instead of one, which means the engine repaying your debt improves without you doing anything. We are building a spend rail, because a credit line you have to manually withdraw and bridge is a loan while a credit line you can spend directly is a financial product, and the distance between those two things is the difference between a protocol people admire and a protocol people use. Loan to value will tighten across weekends and around known earnings dates, since gap risk is not constant and treating it as constant leaves safety on the table. Fixed term peer to peer loans arrive for people who want certainty over automation, where two parties agree a rate and a deadline directly. And all of it goes into a phone, because the moment someone needs cash without selling is rarely a moment they are sitting at a desk. Underneath every one of those steps is one specific regret we are trying to eliminate: the regret of having been right and having nothing to show for it, because life needed money on a Tuesday. Your best position should not be the first thing you sacrifice. Borrow against it, let it work, and let it settle its own debt. Pillars hold things up. You do not tear one down for parts. Pillar Finance. Never sell. Never repay.