Marqeta Business Model: How Modern Card Issuing Powers Square, DoorDash, and Instacart

In 2010, Jason Gardner had a frustrating realization. He was trying to build a prepaid card product and discovered that the card issuing industry was stuck in the 1980s. Legacy processors used batch systems. Integration took months. Customization was nearly impossible.

Gardner decided to build what he wished existed: a modern, API-first card issuing platform. Marqeta would let any company create and manage card programs with the same ease that Stripe brought to payment acceptance.

The early years were difficult. Card issuing is heavily regulated. You need bank partnerships, network relationships, and compliance infrastructure. Gardner spent years building the foundation before Marqeta could issue its first cards.

The breakthrough came with the gig economy. DoorDash needed to pay Dashers instantly. Instacart needed cards for shoppers to buy groceries. Traditional card issuers couldn't handle these use cases. Marqeta's Just-in-Time (JIT) funding was perfect - fund the card at the exact moment of the transaction.

Then came Square. Cash App needed a card program, and Marqeta won the deal. Square became Marqeta's largest customer, eventually representing about half of revenue. The relationship was transformative but also created concentration risk.

Marqeta went public in 2021 at a $15 billion valuation. The stock soared initially, then crashed with other fintech stocks. But the business kept growing. By 2025, Marqeta processes over $200 billion annually, powering cards for the biggest names in fintech and gig economy.

The company that started because Jason Gardner couldn't find a modern card issuer now defines what modern card issuing looks like — though Gardner himself stepped down as CEO in January 2023, succeeded by Simon Khalaf. Khalaf stepped down in February 2025, and Mike Milotich — Marqeta's CFO at the time — took over as interim CEO before being named permanent CEO in September 2025. Patti Kangwankij was named CFO in February 2026, completing the leadership turnover.

Traditional card issuing was broken:

Legacy Technology

Card processors built their systems in the 1970s-80s: - Batch processing (not real-time) - Mainframe-based - Limited APIs - Months to integrate

Inflexibility

Traditional issuers offered: - Standard card products - Limited customization - Fixed rules - No programmability

Slow Time-to-Market

Launching a card program took: - 6-12 months minimum - Extensive paperwork - Multiple integrations - Compliance hurdles

No Modern Use Cases

Traditional issuers couldn't support: - Instant virtual cards - Just-in-time funding - Real-time controls - Gig economy payments

The Gig Economy Problem

DoorDash needed to: - Pay Dashers instantly - Fund cards only for specific orders - Control spending by merchant - Scale to millions of workers

No traditional issuer could do this.

Gardner's Insight

What if card issuing was rebuilt with modern technology? APIs instead of batch files. Real-time instead of next-day. Programmable instead of fixed.

Marqeta rebuilt card issuing for the modern era:

1. Modern APIs

RESTful APIs for everything: - Create cards programmatically - Set controls in real-time - Get instant notifications - Integrate in days, not months

2. Instant Card Creation

Create cards instantly: - Virtual cards in milliseconds - Push to Apple/Google Pay - Physical cards shipped fast - No waiting

3. Just-in-Time (JIT) Funding

Revolutionary capability: - Fund cards at transaction moment - No pre-funding required - Approve/decline in real-time - Perfect for gig economy

Example: DoorDash funds a Dasher's card only when they're at the restaurant buying food for an order.

4. Programmable Controls

Set rules on every card: - Merchant category limits - Spending caps - Time restrictions - Geographic controls

5. Real-Time Everything

No more batch processing: - Instant authorization - Real-time notifications - Live reporting - Immediate updates

6. Global Scale

Issue cards worldwide: - 40+ countries - Multiple currencies - Local compliance - Global platform

Marqeta is a B2B card-issuing platform: it earns a slice of every transaction its customers' cards process, not consumer fees. In FY2025 it generated $625M of net revenue (+23%) on $383B of total processing volume (TPV, +31%).

A crucial nuance: net revenue overstates the business because most of it is pass-through card-network cost. Investors watch gross profit ($437M in FY2025, +24%) as the true margin signal. Marqeta turned adjusted-EBITDA positive ($110M) in FY2025, narrowed its GAAP net loss to just $14M for the year — and then posted its first-ever GAAP profitable quarter in Q1 2026 ($7.8M net income on $165.8M net revenue, +19%).

The big risk is concentration: Block (Cash App) once drove 68% of revenue and is still ~44-45%, and a December 2025 pricing step-down is expected to trim 2026 growth. DoorDash, Affirm, Klarna and European expansion are diversifying the mix by design.

Marqeta's growth followed the gig economy:

Phase 1: Building Foundation (2010-2016)

Built the platform. Got bank partnerships. Obtained licenses. Slow, foundational work.

Key milestones: 2010 founded, 2014 first cards, 2016 gig economy traction.

Phase 2: Gig Economy Explosion (2017-2020)

DoorDash, Instacart, Square all signed. JIT funding was the killer feature. Volume exploded.

Key milestones: 2017 Square deal, 2019 unicorn status, 2020 $60B TPV.

Phase 3: IPO and the Block Problem (2021-2023)

Marqeta went public at a $15B valuation, then the fintech multiple collapsed and the stock followed. The deeper issue was concentration: Block (Square and Cash App) accounted for the lion's share of volume, so Marqeta's fate was tied to a single customer. The work of these years was unglamorous - signing new programs, renegotiating the Block relationship, and proving the platform could grow without leaning on one account.

Key milestones: 2021 IPO, 2023 diversification and Block renewal.

Phase 4: Record Volume, Real Margins (2024-2026)

The diversification thesis played out in the numbers. FY2025 total processing volume reached $383 billion, up 31%, with net revenue of $625 million (up 23%) and gross profit of $437 million. Crucially, Marqeta turned adjusted-EBITDA positive at $110 million while narrowing its GAAP net loss to just $14 million — and then delivered its first-ever GAAP profitable quarter in Q1 2026. European expansion became a real growth engine — accelerated by the acquisition of TransactPay (announced Feb 2025, completed Aug 2025), a UK/EEA-licensed E-Money Institution and Mastercard principal member, which gave Marqeta full card-program-management capability in Europe instead of a partner-dependent model — and powering cards for the likes of Affirm and Klarna - not just Block - showed the platform had finally broadened its base.

Marqeta's moat has multiple layers:

1. Technology Platform

Modern, API-first platform: - Years to build - Continuous improvement - JIT funding unique - Hard to replicate

2. Customer Relationships

Deep integrations with major customers: - Square/Block - DoorDash - Instacart - High switching costs

3. Network Relationships

Visa and Mastercard partnerships: - Preferred partner status - Direct connections - Years to build - Competitive advantage

4. Regulatory Infrastructure

Licenses and compliance in 40+ countries, combined with years of audit history, make Marqeta a "compliance-as-a-service" layer for fintechs. In Europe specifically, Marqeta acquired TransactPay (announced Feb 2025, completed Aug 2025) — a UK/EEA-licensed E-Money Institution and Mastercard principal member live in 25 countries and 16 currencies — giving it full card-program-management capability in the region instead of relying on a partner-dependent model.

5. Square/Block Technical Lock-in

The deep technical integration with Square (Cash App/Square Card) creates a massive volume advantage that allows Marqeta to negotiate superior rates with card networks.

6. Embedded Finance Ecosystem

Marqeta's toolkit for non-fintechs (like DoorDash/Instacart) creates a structural advantage by embedding card issuing into the operations of the gig economy, a market legacy processors can't serve effectively.

Challenges to the Moat:

Stripe Issuing bundles the same capability into a developer suite millions already use. Galileo (SoFi, ~158M accounts) competes for modern programs. And the moat's biggest crack is its anchor: Block (~44-45% of revenue) negotiated a December 2025 pricing step-down that will trim 2026 growth — proof that a concentrated customer holds pricing power over the platform.

The Moat Question:

Marqeta's moat is real — switching costs and certifications are genuine — but it must convert technology leadership into a broader, less Block-dependent customer base before rivals close the gap.

Marqeta's journey from an Oakland startup to a global infrastructure giant offers powerful lessons: