Shopify Capital originated $4.2 billion in merchant cash advances and loans in 2025 — up 40% from $3 billion in 2024. PayPal has passed $30 billion in cumulative small business loan originations over twelve years. Square Loans facilitated $1.3 billion in a single quarter. Toast Capital crossed $1 billion in cumulative restaurant lending by mid-2024. These are not banks. They are the same platforms that process the SMB’s payments, host their storefront, and track their daily sales. They know the business’s revenue in real time — not from financial statements filed quarterly, but from the transaction data flowing through their system every hour. They underwrite from what the business actually does, not what it reports. The float killer is embedded lending that bridges the gap documented in UC-161, UC-162, and UC-163 — using data the traditional credit system cannot access.
Analysis via 🪺 6D Foraging Methodology™
The traditional bank underwriting process for a small business loan requires financial statements, tax returns, a credit check, collateral documentation, and weeks of processing. The bank evaluates the business based on historical reports that may be months old, through a risk model designed for stability. The Fed SBCS found that only 41% of applicants received all the financing they sought, and 24% received none. The system is slow, document-heavy, and structurally biased against the businesses that need capital most — young firms, seasonal firms, and firms with variable revenue. Platform lenders invert this model entirely. Shopify sees every transaction that flows through a merchant’s store. Square sees every card swipe at the register. PayPal sees every payment received. Toast sees every restaurant ticket. The underwriting decision is not based on what the business reported six months ago. It is based on what the business did yesterday.[1][2]
Shopify Capital’s trajectory demonstrates the scale of this advantage. Launched in 2016 with $9 million in Q1 merchant cash advances, Shopify Capital reached $4.2 billion in originations for full-year 2025 — a roughly 19,500% increase over a decade. The growth is accelerating: $3 billion in 2024, $4.2 billion in 2025, with $1.78 billion in outstanding receivables on the balance sheet as of mid-2025. The programme has expanded from the U.S. to eight countries, adding Germany, the Netherlands, Ireland, and Spain in 2025. Shopify’s write-offs remain manageable: $103 million in loans and $28 million in merchant cash advances written off in the most recent fiscal year against $4.2 billion originated. The current rate (percentage of loans not delinquent) was 91.9% at year-end 2025 — slightly down from 93.7% in 2024, reflecting the scale of growth but not signalling systemic credit deterioration.[3][4][5]
We are capitalising on untapped opportunities in SMB.
The competitive landscape confirms this is not one company’s experiment. PayPal announced in March 2025 that it had surpassed $30 billion in cumulative global loan originations for small businesses across twelve years of operation. Square Loans (Block) facilitated 129,000 loans tied to $1.3 billion in originations in Q3 2024 alone, up 17% year-over-year, with successful secondary-market loan sales to investors demonstrating capital-markets confidence in the model. Toast Capital crossed $1 billion in cumulative restaurant lending by May 2024, offering loans of $5,000 to $300,000 through a WebBank partnership specifically designed around restaurant seasonality and profit margins. Non-bank lenders now account for 42% of all small business financing, up from 25% in 2018, and SMBs are 2.6 times more likely to approach a non-bank lender first than they were seven years ago. The structural shift from bank-centred to platform-centred lending is not emerging. It has arrived.[6][7][8][9]
The structural innovation is not that platforms lend money. It is that the lending is embedded in the platform the merchant already uses. The Shopify merchant does not apply for a loan through a separate process. They receive an offer inside their Shopify dashboard, generated by Shopify’s analysis of their actual sales data. The repayment is automatic: a fixed percentage of daily sales is deducted before the merchant sees the revenue. If sales are high, repayment accelerates. If sales are slow, repayment slows. The float killer adapts to the very seasonality and variability that traditional credit models penalise.[3]
This embedded model solves three specific problems that UC-161, UC-162, and UC-163 identified. First, it eliminates the application friction that the Fed SBCS documents — the weeks of paperwork, the credit checks, the document gathering. The merchant receives an offer; they click accept. Second, it adapts repayment to revenue timing — the very mismatch that creates the 90-day float. Fixed monthly loan payments on variable revenue (the traditional model) compress the float during slow months. Percentage-of-sales repayment (the platform model) expands and contracts with the business. Third, it addresses the credit underwriting bias against seasonal businesses that UC-162 identified: the platform sees the seasonal pattern as normal because it has years of the merchant’s transaction history. It does not confuse a February dip for a structural decline.[10]
The model is not without tension. UC-138 (The Algorithm Tax) documented how platform dependency extracts value from SMBs. The float killer creates a new dependency: the merchant who relies on Shopify Capital for working capital is more deeply locked into the Shopify ecosystem. The lending relationship reinforces the platform relationship. The merchant who might have switched to a competitor stays because their capital access depends on their Shopify history. This is the structural trade-off at the heart of embedded lending: the platform solves the float problem but deepens the dependency that created the algorithm tax. The benefit is real. The lock-in is also real. Both are structural.
The cascade has a dual origin in D6 (Operational/Platform) and D3 (Revenue/Capital). D6 captures the embedded infrastructure: the lending is built into the platform’s operating system, not bolted on. Shopify Capital lives inside the Shopify dashboard. Square Loans lives inside the Square ecosystem. The operational integration IS the innovation. D3 captures the capital injection: $4.2 billion from Shopify alone in 2025, filling the float gap at a scale that no community bank programme can match.
D5 (Quality/Underwriting, 35) captures the decision quality improvement: underwriting from real-time transaction data produces more accurate credit decisions than underwriting from quarterly financial statements. The platform knows if the merchant had a good week. The bank knows if the merchant had a good quarter — three months later. D1 (Customer, 30) captures business continuity: the merchant who bridges the float maintains service quality, keeps inventory stocked, and does not turn away customers. D2 (Employee, 28) captures the payroll dimension: float-killing capital keeps employees paid through cash gaps. D4 (Regulatory, 15) captures the current regulatory framework: most platform lending uses merchant cash advance structures (purchase of future receivables) rather than traditional loan structures, operating in a regulatory space that is less constrained but also less protective.
UC-161 established the structural mismatch: median 27 buffer days, 56% borrowing for operating expenses, 24% receiving no financing at all. UC-164 maps the structural response. The float killer does not eliminate the 90-day float. It bridges it — with capital that is faster to access, adapted to revenue timing, and underwritten from actual business performance rather than reported financial statements. The $4.2 billion Shopify originated in 2025 is capital that flowed to merchants whose 27-day buffer was insufficient. The float still exists. The bridge is new. → Read UC-161
UC-138 documented platform dependency as a structural extraction mechanism. UC-164 reveals the other side: the same platform dependency that extracts fees also delivers capital access that traditional banks cannot match. The merchant pays the algorithm tax AND receives the float killer from the same platform. The dependency deepens because the lending relationship reinforces the platform relationship. This is the core tension: embedded lending solves the float problem while deepening the ecosystem lock-in that created the algorithm tax. The benefit and the cost are inseparable. → Read UC-138
UC-150 documented how platforms mediate trust in the trades through reviews and matching. UC-164 reveals the financial parallel: platforms mediate trust in lending through transaction data. The platform knows the merchant’s revenue because it processes the payments. This real-time data creates a trust relationship that banks — who see only quarterly reports — cannot replicate. UC-150 mapped trust mediation for service; UC-164 maps trust mediation for capital. Both use the platform’s data advantage to solve a problem that traditional institutions cannot. → Read UC-150
-- The Float Killer: 6D Amplifying Cascade
FORAGE float_killer
WHERE platform_lending_annual >= 3000000000
AND non_bank_lending_share_pct >= 0.40
AND embedded_underwriting = real_time_transaction_data
AND repayment_model = percentage_of_sales
AND current_rate >= 0.90
AND platform_count >= 4
ACROSS D6, D3, D5, D1, D2, D4
DEPTH 3
SURFACE float_killer
DRIFT float_killer
METHODOLOGY 88 -- Shopify SEC filings (10-K, 10-Q, 8-K: origination volumes, receivables, write-offs, current rates). Block/Square SEC filings (10-Q: loan origination volumes, secondary market sales). PayPal SEC filings (10-Q: merchant receivables, working capital/business loan data; CEO/CFO earnings call commentary). Toast SEC filings (8-K: capital programme data). deBanked origination tracking (industry publication, multi-quarter tracking since 2014). PYMNTS.com platform lending analysis. FunderIntel Q3 2025 earnings breakdown. The Logic (Shopify Capital deep analysis). Crestmont Capital (42% non-bank market share). Fed SBCS 2024 (baseline credit access data).
PERFORMANCE 40 -- The evidence base is anchored in SEC filings — the hardest possible data source: audited financial statements, quarterly and annual reports filed with the Securities and Exchange Commission. Shopify, Block, PayPal, and Toast are all publicly traded companies with mandatory disclosure requirements. Origination volumes, receivable balances, write-off rates, and current rates are all audited figures. The 42% non-bank lending share comes from Federal Reserve and CRA data aggregation. Confidence (0.80) reflects the SEC-filing anchor combined with the recognition that platform lending is still growing rapidly and the long-term credit performance of these portfolios through a full economic cycle has not been tested.
FETCH float_killer
THRESHOLD 1000
ON EXECUTE CHIRP amplifying "Shopify Capital: $4.2B originated 2025 (+40% YoY from $3B in 2024). $1.78B outstanding receivables. Expanded to 8 countries. 91.9% current rate. Write-offs: $103M loans + $28M MCAs against $4.2B originated. PayPal: $30B+ cumulative over 12 years. Value-added services revenue +16% driven by merchant credit. Square/Block: 129K loans, $1.3B originations in Q3 2024 (+17% YoY). Secondary market loan sales demonstrating capital-markets confidence. Toast Capital: $1B+ cumulative by May 2024. $5K-$300K restaurant-specific loans via WebBank. Non-bank lenders: 42% of all SMB financing (up from 25% in 2018). SMBs 2.6x more likely to approach non-bank first. D6+D3 dual origin: embedded lending infrastructure + capital injection at scale. Percentage-of-sales repayment adapts to the very revenue variability that traditional credit penalises. Platform data advantage: real-time transaction history vs quarterly financial statements. Trade-off: float killer deepens platform dependency (UC-138)."
SURFACE analysis AS jsonRuntime: @stratiqx/cal-runtime · Spec: cal.cormorantforaging.dev · DOI: 10.5281/zenodo.18905193
UC-138 documented how platforms extract value through fees, visibility rules, and dependency. UC-164 documents how the same platforms deliver value through embedded capital. The merchant pays Shopify’s subscription, pays Shopify’s transaction fees, competes for visibility within Shopify’s algorithm — and then receives working capital from Shopify Capital when the float threatens survival. The extraction and the rescue come from the same source. This is not a contradiction. It is the structural logic of platform ecosystems: the deeper the dependency, the more the platform knows, and the more it knows, the better it can lend. The data that enables the algorithm tax also enables the float killer.
Banks can lend money. Online lenders can lend money. What platform lenders do differently is adapt repayment to real-time revenue. A traditional loan requires $5,000 per month regardless of whether the business had $50,000 or $20,000 in sales. A Shopify Capital advance takes a fixed percentage of each sale — if sales double, repayment doubles; if sales halve, repayment halves. This structural alignment with the float is the innovation. It does not eliminate the cost of capital. It eliminates the timing mismatch between repayment obligations and revenue reality. For the seasonal business documented in UC-162, this is transformative: the winter slowdown reduces both revenue and repayment simultaneously.
In 2018, 25% of small business financing came from non-bank sources. By 2025, that figure reached 42%. SMBs are 2.6 times more likely to approach a non-bank lender first than seven years ago. Online lender applications have increased for five consecutive years in the Fed SBCS. This is not a temporary fintech trend. It is a structural migration of SMB lending from institutions that see quarterly reports to platforms that see daily transactions. The migration is driven by speed (instant offers vs weeks of paperwork), approval rates (platform data produces yes/no faster), and relevance (platform lenders understand the business because they operate it). Traditional banks are losing SMB lending market share not because their rates are higher, but because their model is slower.
Shopify Capital’s 91.9% current rate and manageable write-offs are from a period of economic growth and rising e-commerce adoption. The model has not been tested through a recession where merchant sales decline 30–40% across the portfolio simultaneously. The percentage-of-sales repayment model is theoretically resilient to downturns (repayment scales down with revenue), but the credit losses from merchants who close entirely have not been observed at scale. UC-165 (The Credit Tightrope) will explore whether the model survives a stress scenario — and whether the regulatory framework evolves to match the scale of platform lending before the next downturn tests it.
The 6D Foraging Methodology™ reads what others call “fintech lending” and finds the amplifying cascade underneath. One conversation. We’ll tell you if the six-dimensional view adds something new.