Why We Built Ample

I have been thinking about this product for nearly two years in the way where you keep running the same loop in your head because the logic is so clean and the gap is so obvious that you cannot understand why it does not already exist at scale.

Today, Ample is live. This essay is an attempt to explain why we built it, what we believe, and where it goes from here. It is long because the idea is layered and I think the layers matter.

The most durable behavioral finding in economics

Let me start with the thesis underneath everything.

One of the most replicated findings in behavioral economics is probability weighting. People do not evaluate probabilities linearly. We systematically overweight small probabilities relative to their objective likelihood. A 1% chance at $1,000 feels more valuable than a guaranteed $10, even when the expected value is identical. This is not irrational in the way economists once dismissed it. It is deeply human. It is consistent across cultures, income levels, and time periods. Kahneman and Tversky documented it formally in Prospect Theory in 1979, but the pattern is as old as recorded civilization.

This is why global lottery markets exceed $300 billion in annual revenue despite offering negative expected value. Think about that for a moment. People willingly accept a mathematically inferior return because the structure of the payoff, the shape of the experience, is more compelling than the alternative.

A small chance at a large gain often feels more valuable than a guaranteed small return, even when expected value is equivalent. This is the psychological foundation Ample is built on, except we preserve positive expected value.

Now here is the question that has consumed me: if hundreds of billions of dollars flow annually into negative expected value lotteries, what happens when you offer the same psychological experience but with zero or positive expected value? What happens when you remove the loss?

The 70 year experiment that answers the question

In 1956, Harold Macmillan, then Chancellor, introduced Premium Bonds to the United Kingdom. The structure was elegant, rather than distributing interest linearly across depositors, the government pooled yield from all deposits and distributed it as randomized prizes. Your principal remained fully backed by the Treasury. You forfeited predictable interest in exchange for participation in periodic prize drawings.

The public response was immediate and overwhelming.

That is not a rounding error. Over 134 billion pounds in deposits. Roughly a third of the UK adult population holds Premium Bonds. ERNIE, the random number generator that selects winners, distributes millions of prizes every single month, ranging from 25 pounds to a million-pound jackpot.

From a purely rational standpoint, many holders could earn higher returns elsewhere. A straightforward savings account or index fund would outperform on expected value over most time horizons. Yet Premium Bonds persist, and not just persist but grow, decade after decade. They have survived interest rate cycles, inflation spikes, financial crises, the rise of the internet, and generational turnover.

Why? Because the product does not compete on rate. It competes on how saving feels. It introduces variance without introducing loss. It makes the act of holding money genuinely exciting rather than passively boring. And it turns out that matters enormously for long-term retention of deposits.

Premium Bonds have survived every macroeconomic regime the UK has experienced over 70 years. Not because they are financially optimal, but because they changed the experience of saving.

Why this never scaled beyond the UK

If the model is so effective, why has it remained largely confined to a single country?

A few other nations attempted variations. The United States experimented with prize-linked savings accounts in the early 2010s. Some credit unions offered lottery-style incentives. But none approached the scale or durability of Premium Bonds. The reasons are structural, not psychological.

The infrastructure problem

Traditional banking infrastructure was not designed for variable distribution of pooled yield. The rails are built for linear, predictable interest payments. Attempting to replicate Premium Bonds on legacy financial infrastructure introduces cascading complexity.

Consider the stack that modern fintech savings products depend on. Deposits may be custodied at a partner bank. Distribution and customer engagement occur through a separate frontend company. Balance reconciliation depends on middleware providers sitting between them. When any layer breaks down, responsibility becomes unclear, and end customers suffer. The collapse of Synapse in 2024, which left thousands of depositors unable to access funds held at Evolve Bank, was a stark illustration of this fragility.

Beyond custody fragmentation, the economics are punishing. Multiple intermediaries extract fees at every layer: card networks, issuing banks, middleware providers, payment processors. In a low-rate environment, these fees compress the available prize budget to the point where the product loses its appeal. Companies end up subsidizing prizes from venture capital rather than funding them from organic yield, which is not sustainable.

And then there is the trust problem. Prize allocation logic in centralized systems is inherently opaque. Users have to trust that the draw is fair, that the odds are as advertised, and that the pool is funded as described. Verification depends on audits and regulation rather than direct auditability.

The Yotta case study

Yotta is worth examining because it came closest. Founded in 2019, Yotta offered a savings account where interest was distributed as lottery-style prizes. The product generated real consumer excitement and grew to meaningful deposits. But it operated entirely on traditional banking rails, with deposits held at Synapse and Evolve Bank.

When Synapse collapsed, Yotta's depositors were caught in the fallout. The company eventually pivoted to crypto-adjacent products before shutting down its prize-linked savings offering. The product-market fit was real. The infrastructure was not.

Why onchain, and why now

The crypto industry has spent a decade building infrastructure. Most of the consumer applications that infrastructure was used for have been, honestly, extractive. Speculative tokens, leverage trading, negative-EV gambling. The rails are powerful. What was built on top of them has been largely disappointing.

But the rails themselves have matured to a point where they can support real financial products. Stablecoin supply has crossed $200 billion. Onchain lending markets like Aave and Euler hold billions in deposits with years of operational history. Verifiable randomness functions from Chainlink and others provide auditable, tamper-resistant random number generation. Smart contract security tooling has improved dramatically.

This is the infrastructure that Ample needs and, critically, it did not exist even three years ago at the required level of maturity.

The stablecoin moment

The broader macro environment matters too. There is a visible regime shift happening in crypto. The speculative chain-and-token cycle that dominated 2020 through 2023 is giving way to something more durable. Three themes are converging: stablecoins as the dominant use case, tokenization of traditional financial assets, and fintech products built on crypto rails rather than legacy banking infrastructure.

Ample sits at the intersection of all three. It is a stablecoin-centric offering, with tokenized assets on the roadmap, delivered through a fintech experience, oriented around the most fundamental financial behavior there is: saving.

How Ample works

The mechanics are straightforward, which is the point.

Users deposit supported stablecoins into the protocol. Those assets are supplied to onchain lending markets where yield accrues continuously. Rather than distributing that yield proportionally to each depositor, as a traditional savings account would, accrued yield is pooled into a prize budget. At defined intervals, that budget is distributed to randomly selected participants using verifiable randomness.

Principal remains segregated from pooled yield at all times. Your deposit does not transfer to an intermediary institution. You maintain ownership of your assets. The smart contracts that govern distribution logic are transparent and auditable. When the protocol charges fees, they are explicit and programmatic, not embedded across layers of intermediaries.

This architecture preserves the expected value of conventional savings while altering the distribution of returns across time and participants. You are not losing anything relative to a standard lending position. You are experiencing the same aggregate return in a psychologically different structure.

We are not competing on rate. We are competing on experience, while preserving positive expected value. The differentiation is in how yield is distributed and perceived.

Behavioral finance as product design

I think about this through the lens of product design rather than financial engineering. The most successful consumer fintech products in history succeeded because they amplified an existing behavioral tendency.

Credit cards separated consumption from payment, reducing the salience of spending. American Express built an empire on that insight. Brokerage platforms introduced gamification to increase engagement. Robinhood brought millions of new participants into equity markets, for better and worse, by making trading feel like a game. Peer-to-peer payment apps created social feeds around money movement. Venmo turned a transaction into a social signal.

In each case, the underlying financial product was not novel. Credit existed before Amex. Brokerages existed before Robinhood. Payments existed before Venmo. What was novel was the behavioral insight applied to the experience layer.

Prize-linked savings follows the same pattern. Savings accounts exist. Yield exists. What Ample changes is the experience of earning that yield, exploiting probability weighting to make saving feel more engaging without introducing additional risk or negative expected value.

The critical difference from many fintech predecessors: this is not extractive. Robinhood monetized through payment for order flow, effectively selling user trades to market makers. The business model created a tension between user outcomes and company revenue. Prize-linked savings has no such tension. The provider benefits from retained deposits and participation in yield. The user retains principal and participates in variance funded by underlying yield. Interests are structurally aligned.

On competitive positioning

Traditional savings products compete on rate. Rate competition compresses margins and commoditizes providers. In rate-driven markets, trust and balance sheet strength tend to favor incumbents. A startup offering 4.2% instead of 4.0% is not defensible.

Prize-linked savings competes on something different entirely. The expected return remains anchored to underlying yield. The differentiation is in the shape of the return, how it is distributed and perceived. Engagement becomes endogenous to the product structure rather than dependent on sustained marketing spend.

This matters for retention. One of the persistent challenges in consumer fintech is that users acquired through rate promotions leave when rates normalize. Churn is structurally high because the value proposition is a commodity. Prize-linked savings creates stickiness through the experience itself, through the anticipation of draws, the excitement of winning, the psychological satisfaction of participating.

Where Ample goes from here

From this base, expansion follows three axes:

Asset breadth

Today it is stablecoins. Over time, the same structure extends to any yield-bearing digital asset: ETH, SOL, BTC (via liquid staking or lending), and eventually tokenized real-world assets like gold, treasuries, and money market instruments. Each new asset class brings its own yield profile and its own depositor base.

Yield modularity

Rather than routing deposits to a single lending market, the protocol can programmatically allocate across multiple yield sources, optimizing for risk-adjusted return. This creates a more resilient and competitive prize budget over time.

Distribution as infrastructure

This is the axis I am most excited about. Ample does not need to be the consumer-facing product forever. The protocol can serve as savings infrastructure that wallets and fintech platforms integrate. Any app with a balance could offer prize-linked savings powered by Ample under the hood. This is how you get from thousands of direct users to millions of indirect ones.

If you take the UK Premium Bond market as a directional proxy, the addressable market for this type of product is enormous. Over 130 billion pounds in a single country with 67 million people. Now imagine that product, natively global, accessible to anyone with a stablecoin balance, composable into any wallet or fintech app.

What crypto should be building

I want to close with something broader because I think it matters.

Over the past decade, the dominant consumer use cases for crypto have been speculative. Memecoins, leverage trading, negative-expected-value gambling dressed up as "DeFi." The industry has extracted enormous value from retail participants through products that exploit behavioral bias rather than serve it. This has contributed, fairly, to the negative perception of crypto by the mainstream.

Ample is built on the same rails. It uses the same stablecoins, the same lending markets, the same programmable infrastructure. But it points that infrastructure in a fundamentally different direction. Instead of monetizing impulsive behavior, it promotes saving. Instead of extracting from users, it aligns with them. Instead of negative expected value, it preserves positive expected value.

Savings products compound with trust. The longer someone holds, the more value accrues, for the user and for the protocol. That alignment is what makes this durable.

I believe this is what crypto should be building. Products that use the genuine technical advantages of onchain infrastructure, transparency, composability, global accessibility, programmable yield, to create financial products that are better for the people using them. Not marginally better. Structurally better.

Ample is live today. We are just getting started.