Our Method

Embedded finance is a system.
The architecture layer determines how it works.

The program model, bank terms, and fee structure are going to get defined either way. The only question is whether architecture defines them — or vendor defaults do.

Architecture → Outcomes

The architecture decisions determine the business outcomes. Every one of them.

Vendors don't tell you this because it's not in their interest to. The program model choice determines your margin ceiling. The bank partner choice determines your float and interchange economics. The compliance design determines your launch risk and investor credibility. These decisions are not technical — they are business model decisions.

Architecture Decision Business Outcomes Affected Cost of Getting It Wrong
Program model Margin ceiling, compliance exposure, speed to market, economics ownership 12–18 months re-architecture, $300K–$800K/yr economics gap
Bank partner Float capture, interchange terms, approval odds, regulatory durability, product ceiling Wrong bank: restricted products, compliance friction, re-diligence cost
Fee structure Revenue per transaction, gross margin profile, customer willingness to pay No fee structure = no fee revenue. Retroactive pricing changes customer backlash.
Infrastructure stack Cost to serve, operating leverage, product flexibility, migration cost when outgrown Tight coupling: 4–6 months extra migration time per component
Compliance design Launch risk, bank acceptance, regulatory durability, investor credibility Missing BSA/AML: bank diligence stalls 3–6 months. Post-launch remediation: $200K+
Go-live model Time to revenue, operational failure risk, team capacity requirements Underprepared ops: exception queues, reconciliation breaks, customer experience failures

Programs that skip the architecture layer don't just fail operationally — they leave the interchange economics, float revenue, and fee structure to be defined by whoever they talked to first. That's a revenue problem, not just a structural one.

The six components below are not a service menu. They are a sequence. The order matters as much as the components. Decisions made in bank partner selection constrain what's possible in compliance design. Decisions made in compliance design constrain infrastructure selection.

0→$1B
Built a digital bank from zero to $1B in under 12 months. $4B in 5 years. 20+ payments and lending products launched.

The architecture method on this page isn't a framework we developed in a consulting room. It's what we learned building real programs under real constraints — where the decisions on this page had real consequences. We've been on the other side of every one of them.

The Critical Decisions

The decisions that determine whether a program succeeds

DECISION 01
Program model
BaaS, direct MTL, or hybrid. The foundational decision that shapes compliance exposure, margin structure, and flexibility ceiling.
DECISION 02
Economics & margin structure
Interchange capture, fee design, revenue sharing, and float economics. Designed before infrastructure selection or inherited from vendor defaults.
DECISION 03
Regulatory & sponsor bank strategy
Which bank archetype fits the program. What commercial terms to negotiate. What compliance posture the program requires.
DECISION 04
Product & distribution model
How the program reaches customers. What the product design requires from the architecture. Where product and compliance intersect.
DECISION 05
Partner structure
Which infrastructure partners serve the architecture. How relationships are structured. What the operating model across parties looks like.
DECISION 06
Compliance & controls framework
KYC/KYB design, transaction monitoring, OFAC requirements, bank compliance expectations. Defined before infrastructure is selected.
DECISION 07
Monetization Architecture
How the program generates revenue — interchange capture, fee design, float economics, speed tier pricing — must be designed as part of the architecture, not discovered after infrastructure is in place. See how we design payments revenue →
Why Programs Fail

The five architecture failure patterns

These are the structural problems we find in virtually every program that calls us post-launch. They are all preventable — at the architecture stage.

01
Vendor selected before program model defined
The BaaS demo happens before the economics are modeled. The program gets built around vendor capabilities and defaults rather than program requirements.
02
Economics inherited from vendor defaults
Fee structures, interchange sharing, and revenue splits accepted from vendor term sheets rather than designed from program requirements. Hard to renegotiate after launch.
03
Compliance requirements surface after partner selection
Bank compliance requirements discovered after contracts are signed. Infrastructure choices that worked for the product don't work for the compliance framework.
04
Operating model ambiguity across teams and partners
No clear operating model across internal teams, sponsor bank, and infrastructure partners. Accountability gaps surface at launch and create costly delays.
05
Architecture not designed for growth trajectory
BaaS economics that work at low volume become untenable at scale. The program wasn't designed for the volume trajectory — restructuring is expensive mid-flight.
System Architecture

Embedded finance is a system — not a vendor decision.

Distribution and infrastructure already exist. What's usually missing is the architecture layer that connects them.

Software Platforms
Embedded Finance Distribution
ERP Platform Vertical SaaS Marketplaces Workflow Platforms
ExpandUp
Embedded Finance Program Architecture & Orchestration
Core Program Design Decisions
Right Model
monetization
Right Partners
bank + infrastructure
Right Packaging
product + UX
Right Governance
program lifecycle
Payments Infrastructure
Payment Rails
ACH RTP Cards Stablecoin
Regulated Banking Layer
Sponsor Bank Compliance Bank Core / Ledger

Distribution and infrastructure already exist. What's usually missing is the architecture that connects them.

Payments Monetization

Architecture and monetization are the same problem.

The interchange capture structure, float economics, fee design, and program model alternatives that determine whether your payment flow generates revenue are all decided at the architecture stage. Most programs treat monetization as a separate problem. It isn't.

See the full payments monetization framework →
Want us to assess where your program is in this sequence? We can identify which architecture decisions have been made, which are missing, and what the gaps are costing you.
Start the conversation →

Start with the architecture.

Every embedded finance program starts with an architecture decision — whether or not it's made explicitly. We help you make it intentionally, before the infrastructure constrains what's possible.