AllReferralCodes

Banks barely changed for decades, then a handful of small teams spent the 2010s picking them apart, one specific complaint at a time. None of them set out to build "a bank with an app." Each one found a single narrow thing traditional banks did badly, and built the entire company around fixing just that.

Simple was founded in 2009 by Josh Reich, making it the first real U.S. neobank and the one that proved a bank account could live inside a clean app instead of a branch. It had no banking license of its own — it ran on top of a partner bank — which was the template almost every neobank after it would copy.

Simple never scaled the way its successors did. BBVA acquired it in 2014 for $117 million, and the app was quietly shut down in May 2021 after BBVA sold its U.S. business to PNC. Simple's real legacy isn't its own numbers — it's that Chime, N26, Monzo, and Revolut all effectively started from the blueprint it drew first: strip out the branch, keep the license arrangement in the background, put the whole relationship in an app.

N26: mobile-first for people who'd never chosen a bank

N26 launched in Berlin in 2013, aimed squarely at a generation that had no attachment to a branch and no interest in acquiring one. Where older banks' apps were bolted onto a decades-old back end, N26 built the back end and the app at the same time, so the product felt native to a phone rather than ported to one. It became one of the clearest signals that European regulation and appetite for mobile-first banking had both matured enough to support a bank built with no physical footprint at all.

Nubank: undercutting a five-bank cartel

Nubank's founding problem was the most concrete of any neobank on this list. When David Vélez started the company in Brazil in 2013, five banks controlled roughly 80% of the country's banking market, and charged some of the highest fees and interest rates anywhere in the world — annual fees on basic credit cards, monthly fees on ordinary services, charges for things like fraud protection and text alerts that most markets treat as standard.

Nubank's opening move wasn't even a full bank account — it was a credit card with no annual fee, built and issued without Nubank needing a banking license of its own at first. The pitch was simple because the market had made it simple: charge people less than the incumbents did, and be easier to deal with while doing it. It worked at a scale few neobanks anywhere have matched — Nubank now serves more than 100 million customers across Brazil, Mexico, and Colombia, and went public on the NYSE in December 2021 as one of the most valuable digital banks in the world.

Chime: banking for people the old system charged the most

Chime, founded in 2012 by Chris Britt and Ryan King and live from 2014, picked a different target inside a market — the U.S. — that already had plenty of banks. Its target was overdraft fees, historically one of the largest sources of fee revenue for American retail banks and one that fell hardest on people living closest to their balance.

Chime's answer was structural rather than promotional: instead of charging account fees and overdraft fees, it made its money the way a debit card issuer usually does — from the interchange fee merchants pay on each transaction — and used that to offer fee-free banking and features like early access to direct deposits. It didn't need every customer to be profitable in isolation the way a fee-funded bank does, so it could serve the exact customers overdraft-fee banking had been built to profit from, and not charge them for it.

Revolut: built by traders who were tired of losing money on their own trips

Revolut was founded in 2015 by Nikolay Storonsky and Vlad Yatsenko. Storonsky had spent years as an equity derivatives trader at Lehman Brothers and then Credit Suisse, which meant he understood exactly how exchange rates were quoted at institutional desks — and exactly how much worse the rate was that ordinary customers got when they used a bank card abroad or sent money to another currency.

Revolut's first product was narrow on purpose: a card that gave people the real interbank exchange rate instead of the marked-up one, aimed at frequent travelers who felt that markup on every trip. It's since expanded into a much broader financial app, but the founding niche was exact — one specific, well-understood inefficiency that a trader would have spotted immediately and a typical bank customer would never have had the visibility to notice.

Wise: the same complaint, solved peer-to-peer instead

Wise (launched as TransferWise) grew out of almost the identical frustration, arrived at from a personal moment rather than a trading desk. Co-founders Kristo Käärmann and Taavet Hinrikus, two Estonians living in London who met in 2007, built the company after Käärmann transferred a £10,000 Christmas bonus to his Estonian bank account and realized the exchange rate his bank quietly applied had cost him roughly £500 versus the real mid-market rate.

Their first fix wasn't even a company — it was a personal workaround. Hinrikus would move euros into Käärmann's Estonian account while Käärmann moved pounds into Hinrikus's British one, so both ended up with the currency they needed without either paying a bank to convert anything. That peer-to-peer trick became TransferWise in 2011, built around a simple promise: the real mid-market rate, with the fee shown upfront instead of hidden inside the exchange rate itself.

Monzo: building the bank in public

Monzo was founded in 2015 by Tom Blomfield, who had been part of the founding team behind what would become Starling Bank before leaving to start his own. Monzo launched first as a prepaid coral-pink card in open beta, with the roadmap and the company's decisions discussed openly with the community actually using it, before it held a full banking license of its own. The card's now-famous color wasn't an accident of branding — it was there to be recognizably different from every bank card already in every wallet, at the exact moment someone pulled it out.

The pattern underneath all of them

None of these companies pitched themselves as "a bank, but better." Each one picked a single specific failure of the existing system — no mobile-native experience, predatory fees in a concentrated market, overdraft charges aimed at people who could least afford them, hidden markups on currency conversion, a banking relationship with zero transparency — and built the whole company as the fix for that one thing. The full-service "everything app" each of them looks like today came after, once the initial wedge had already won trust.

That's also why so many of them, banks and fintechs alike, lean on referral programs to grow: a banking product gets more useful to everyone the more people are already using it, and the fastest way to reach that density is still the same one PayPal used in 1999 — get an existing customer to bring someone they trust.

Every referral program you see today — $20 off here, a free month there — is running a bet that was first placed at scale by PayPal in 1999. The mechanics have barely changed since. What's changed is how much companies have learned about which version of the bet actually pays off.

PayPal: the original blitz

PayPal's problem in 1999 wasn't building a payments product, it was getting anyone to use it. Peer-to-peer online payments only work if the other person is also on the platform, so the company needed density fast, in a market with no existing habit to piggyback on.

Their answer was blunt: pay people to sign up, and pay them again to bring a friend. New users got $10, sometimes $20, just for opening an account, and both the referrer and the new signup got another cash bonus when the referral worked. There was no cap, no clever mechanic — just cash, deposited directly.

It worked almost too well. PayPal's user base went from around 1 million to 5 million between March and the summer of 2000, growing at 7–10% a day at points. By 2002 it had crossed 100 million accounts. Peter Thiel, one of PayPal's founders, later described the arithmetic candidly: exponential user growth funded by a bonus that cost real money, with no revenue growing to match it. The company was reportedly paying out tens of millions of dollars in signup and referral bonuses in its first year alone.

That's the part most retellings of PayPal's growth story skip: the program was not sustainable at that payout level, and PayPal knew it. The reward was walked down over time, from $20 to $10 to $5, and eventually phased into other incentives entirely as the network reached critical mass on its own. The referral program's job was never to run forever — it was to buy density before competitors could establish it first, and it did that job so well that it also nearly became a case study in burning cash for growth.

Dropbox: paying in a currency that costs (almost) nothing

Dropbox launched its referral program in September 2008 with a very different insight: PayPal's problem was that cash bonuses cost cash. Dropbox's product was storage, and marginal storage was cheap. So instead of paying people money, Dropbox paid them in more of the product — 500MB of extra space for both the referrer and the person they invited, up to a cap of 16GB.

The effect on growth was dramatic and well documented. Dropbox's signups went from around 100,000 to 4 million in about 15 months, a roughly 3,900% increase. At its peak, referrals accounted for as much as 35% of all daily signups, and Dropbox has said the program permanently lifted signups by around 60% even after the initial spike faded.

What made it durable, where PayPal's program wasn't, is that the reward and the product were the same thing. Every dollar PayPal spent on a referral bonus was a dollar that could otherwise have gone to revenue. Every gigabyte Dropbox gave away cost fractions of a cent and, more importantly, made the product itself more useful to the very people most likely to keep using it and referring others.

Robinhood: selling access, not cash

By the time Robinhood was preparing to launch in 2013, giving away money or product wasn't the only lever left. Robinhood didn't have a live product to hand out yet — it hadn't launched. So it sold something else entirely: your place in line.

The app opened a waitlist well before it was available, and the only way to move up that list faster was to refer other people. There was no cash and nothing tangible changed hands — the reward was purely positional, a smaller number next to your name. Scarcity and status turned out to be enough: the waitlist crossed a million signups before Robinhood officially launched in March 2015, driven in part by an organic post on Hacker News that took on a life of its own, and the company got there without meaningful paid advertising spend.

Robinhood's version is the clearest proof that a referral program doesn't need a large budget line at all — it needs a reward that feels valuable to the specific audience being recruited. For early fintech adopters watching a waitlist counter, "sooner" was worth more than a coupon would have been.

What actually separates a referral program that works from one that doesn't

Line the three up and a pattern falls out that has held up in every referral program since:

The reward should cost the company less than it's worth to the user. Dropbox's storage and Robinhood's queue position both cost the company close to nothing per unit while feeling meaningfully valuable to the recipient. PayPal's cash bonus was the opposite, and that mismatch is exactly why it had to be wound down.

Both sides need a real reason to act, not just the referrer. All three programs paid or benefited the new signup as well as the person doing the inviting. A one-sided reward asks someone to spend social capital on your behalf for nothing in return.

The reward should reinforce the product, not sit next to it. Storage for a storage product and queue priority for a scarce product both deepen the reason someone already wanted to sign up.

Know when to turn it off. PayPal's bonus was a deliberate, temporary subsidy to solve a cold-start problem, and the company treated it that way, ratcheting it down as the network effect took over.

That last lesson is the one still playing out today, just with different products. Neobanks, crypto exchanges, and investing apps run referral programs for the same cold-start reason PayPal did, and most of them are just as deliberate about capping the reward and the payout window. The mechanics PayPal, Dropbox, and Robinhood pioneered are still the three basic shapes a referral program can take: pay in cash, pay in product, or pay in access.