In the spring of 2015, Jason Gardner, the founder of Marqeta, a payments processor, left a grim board meeting and went for a walk with his lead investor, Arnon Dinur of 83North. Facebook had pulled the plug on a joint initiative, and Marqeta had fallen far short of its revenue target.
Walking the streets of Emeryville, California, a small town between Berkeley and Oakland, Gardner told Dinur that Marqeta wouldn’t last long with the cash it had. He needed to buy time. The company moved to a weekly budget, and Gardner volunteered to cut his own salary by 40%. “Ninety out of one hundred entrepreneurs would have asked for more money,” Dinur says.
“I wanted to show my commitment to not only the board but the company, that I’m willing to do anything to get to the next step through determination,” Gardner says. “It’s almost like inflicting pain on yourself. . . . It gets you to understand what’s at stake here.” Two other executives also volunteered to slash their salaries by 40%, and the startup didn’t lay anyone off.
Four years later investors were meeting again at Marqeta’s office, now a 16-story building in Oakland with the company’s name on it. This time the meeting’s tone was different. Revenue had doubled every year since 2016, reaching close to $150 million in 2018, a source tells us. Marqeta was finalizing plans to raise at least $250 million at a valuation of about $2 billion, nearly quadrupling its value from two years ago.
Despite a rough start, Marqeta has had the right idea since its founding in 2010. It pursued a niche in the payments-processing business that had seen little innovation in over a decade: card issuing and processing, which involves deciding whether a debit card transaction should be approved. It iterated through three business models, staying frugal along the way and ultimately landing on an open-software platform that outside engineers could easily plug into. It identified Square and Instacart as future winners, recruiting them as customers and latching on for the ride.
Gardner, 49, is hardly the type you’d expect to become a successful Silicon Valley entrepreneur. Growing up in a middle-class family in New Jersey, the son of a stockbroker father and a legal mediator mother, he had a motley assortment of jobs in high school, from working at a thrift shop to selling tie-dyed shirts on New Jersey trains on his way to some of the 80 Grateful Dead concerts he attended. Although he wasn’t a coder, he liked hanging around Radio Shack and taking apart radios and TVs. In college at Arizona State, he worked as an assistant to Senator John McCain in his early 20s but decided politics wasn’t for him.
Later he worked in sales at research companies like Gartner, eventually founding a payments startup in 2004 that allowed people to pay rent electronically. Gardner was so short on cash over the next few years that he sometimes put mortgage payments on his credit card. In 2007 he sold the company to MoneyGram for $28 million and stayed on as an executive for two years.
Then he started to think about new uses for debit cards. His first idea was a prepaid loyalty card sold at grocery stores where you could pay $50 for $55 worth of items from retailers like Jamba Juice. Gardner called it Marqeta (after a woman he and a friend had traveled with in Prague), brushing off the fact that a marketing firm was already using the name Marketo. He raised $6 million from investors like the Israeli VC firm 83North.
It took him almost two years to release the product, because he’d built an entirely new payment processor along the way. He opted not to partner with one of the big companies that had been doing issuing and processing for decades, like First Data or FIS, because he thought going solo would give him speed and flexibility. But the loyalty card flopped. Gardner learned that it took too much capital to scale a consumer retail product and that he wasn’t good at consumer-facing design. “I like the complexity of building infrastructure,” he says.
Gardner’s second business product was commissioned by Facebook: a gift card that you could send to friends and was redeemable at places like Target and Olive Garden. Facebook launched it in January 2013 but was disappointed in the sales and shut down the card about a year later.
Around this time, companies like the communications software maker Twilio were starting to let clients access their technology and customize it via application programming interfaces, or APIs. Gardner chose that approach for Marqeta’s third product and announced it in late 2014.
With Marqeta’s API, companies that wanted to issue debit cards could authorize transactions themselves and set the criteria for accepting them. “We move the system of record or ledger to our customer,” Gardner says. And companies no longer needed to separately solicit relationships with a card network (like Visa), a bank, a transaction processor and a plastic card manufacturer. Marqeta had built those partnerships and wrapped them up in one package.
Once customers were set up, Marqeta would make money the same way Visa and Mastercard do, by taking a cut of every transaction. How much? Marqeta is mum, but we’re told the average fee is roughly 1% before rebates to clients.
One of its first clients was DoorDash, the San Francisco food delivery company whose thousands of “Dashers” retrieve takeout meals from restaurants on behalf of customers. With Marqeta, DoorDash has issued debit cards that don’t work unless a Dasher is at the correct restaurant, and it won’t authorize transactions for values that exceed the customer’s order amount. Over the past two years, Marqeta’s technology has helped delivery companies cut fraud in half, to 5% or less.
In 2016, Marqeta’s trajectory tilted upward. Instacart let Marqeta power the debit cards its freelance delivery people used to buy groceries. When Square decided to issue a virtual debit card paired with its fast-growing money-transfer app, Square Cash, Marqeta helped it build the product within six weeks, rather than the months a traditional issuer-processor would have taken. Square also used Marqeta to create the plastic debit card it released the following year.
Kabbage, a small-business lender, signed on, issuing a Marqeta-powered debit card that let customers spend some of their loaned funds at retail merchants. It took Kabbage “months versus quarters” to release the cards, says Kathryn Petralia, Kabbage’s president. “All traditional providers remain cumbersome.”
Two insights drove Marqeta’s success. First, instead of focusing on banks as customers—as Fiserv, Tsys and FIS have done profitably for decades—it looked sideways, directly targeting tech-enabled service providers in the new economy. Second, by taking an API approach, Marqeta sped up the setup process and catered to companies that want to control how digital payments are authorized.
In 2017 Alipay, the Chinese payment app that has more than 900 million global users, signed on to enable Chinese nationals to use the app at U.S. retailers while traveling. Brex, the credit card startup, became a customer. Marqeta raised $25 million that year from investors like Visa, Granite Ventures and 83North, while bringing in $70 million of revenue, we estimate. It doubled its staff to 160 employees.
“Payments is deceptively complex,” says Omri Dahan, Marqeta’s chief revenue officer. “There are a range of varied interests in the ecosystem who need to be balanced just to produce one card swipe that’s successful. . . . You have to get used to getting punched in the face every day.”
It helps to have healthy clients. Square went from a $4 billion market value in 2016 to $32 billion today, becoming Marqeta’s largest customer and processing more than $5 billion in volume through Marqeta last year, estimates Brett Winton at Ark Investment Management, a large Square investor. In three years, DoorDash’s valuation has gone up tenfold to $7 billion and Instacart’s fourfold to $8 billion.
Dahan says Marqeta hunts for clients in large markets that are being disrupted. And it thinks like a venture capitalist. “What do we think of the company? Is it well funded? Well led? How are their engineers?”
Marqeta expects to double revenue again this year. Dahan says its clients are evenly spread across a handful of industries, which include lending, delivery, e-commerce, travel and one he thinks is the most promising: digital banking. In addition to Square, which has 15 million monthly users for its Cash app and is looking more and more like a bank, Marqeta is working with (as yet unnamed) digital-first “challenger banks” in Europe.
What could go wrong? FIS has APIs with features that are similar to what Marqeta offers and could make them better if its bank customers want more. Stripe, the payments giant that’s worth $22.5 billion, released a card-issuing service last year at what is rumored to be half Marqeta’s price.
“You can put 50 engineers on something and maybe move five times faster,” Dahan says. “But being in-market, scaling programs, learning from those programs? That is not something that can be fast-forwarded.”
For now, Marqeta has something else in its favor: its clients’ trust, which is rather important in a money-handling business. But there’s nothing permanent about even this advantage, so Jason Gardner is going to be looking over his shoulder. He’s had some practice with that.