Key Takeaways
- A single oversized MCA deal wiped out more than $40 million for one funder, and concentration risk remains the most underpriced threat in alternative lending portfolios.
- AI-powered bank verification can detect revenue dependency, customer concentration, and seasonal volatility that static financial documents miss entirely.
- MCA underwriting best practices now require analyzing transaction-level banking data, not just top-line deposit averages, before approving large advances.
- Asynchronous screen recording of live banking sessions gives underwriters visual evidence of account health that PDFs and API snapshots cannot replicate.
- Funders scaling into bigger deals need verification infrastructure that matches the risk, not just the speed, of their deployment.
When One Deal Takes Down an Entire Portfolio
The merchant cash advance industry has a concentration risk problem that keeps getting more expensive. In November 2018, a conglomerate of car dealerships collapsed in California. Within three weeks, 1 Global Capital filed for bankruptcy after losing more than $40 million on that single relationship. The ripple effects lasted years, with investors, syndication partners, and downstream funders absorbing losses they never saw coming.
That story, recently revisited by deBanked's reporting on big deals gone wrong, is not an isolated case. It is a pattern. Funders chasing higher deployment volume approve oversized deals without the verification depth to match the risk. The underwriting that works for a $50,000 advance simply does not hold up when the number climbs to $500,000 or beyond. What separates the funders who survive from the ones who don't is how they verify the underlying business before committing capital. MCA underwriting best practices for big-ticket deals demand a fundamentally different approach, one that AI-powered bank verification is finally making practical.
Why Traditional Underwriting Misses Concentration Risk
Deposit Averages Hide Revenue Fragility
Most MCA underwriters evaluate bank statements by looking at average daily balances, total monthly deposits, and NSF frequency. These metrics work well enough for small advances against diversified revenue streams. They fail catastrophically when the merchant's deposits depend on one or two large customers.
Consider a construction subcontractor depositing $200,000 per month. The statements look healthy. The averages are strong. But 70% of those deposits come from a single general contractor. If that relationship ends, the merchant's revenue drops by more than half overnight, and the funder's advance is suddenly underwater. Static bank statement analysis rarely surfaces this kind of dependency because it treats all deposits as interchangeable.
PDF Statements Cannot Show What Matters Most
Bank statements in PDF format present a flattened view of financial activity. They show transaction amounts and dates, but they strip away the context that matters for concentration risk analysis. An underwriter reviewing a PDF cannot easily distinguish between 50 small customer payments and two large ones that happen to total the same amount. The visual structure of the document obscures the very patterns that indicate fragility.
Even when underwriters manually categorize transactions, the process is slow, inconsistent, and prone to error. Two different analysts reviewing the same statements will often reach different conclusions about revenue concentration. This inconsistency compounds when deal size increases, precisely when accuracy matters most.
API-Based Verification Gives a Snapshot, Not a Story
Open banking APIs and instant account verification tools provide real-time balance data and categorized transaction feeds. These are useful for confirming account ownership and checking current balances. They are far less useful for understanding how a business's revenue composition has shifted over three, six, or twelve months.
API data is structured for aggregation, not investigation. It tells you what happened but not whether the pattern is sustainable. For a $500,000 advance, the underwriter needs to understand revenue trajectory, customer mix stability, and seasonal variation across multiple statement periods. As we explored in our analysis of how real-time balance checks create false confidence in MCA underwriting, a healthy snapshot today does not guarantee a healthy business tomorrow.
How AI-Powered Bank Verification Detects Concentration Risk
Transaction Clustering and Source Identification
Modern AI systems can analyze transaction descriptions, amounts, and timing patterns to cluster deposits by likely source. Rather than treating each deposit as an independent data point, machine learning models group recurring payments from the same payer based on description similarity, amount consistency, and deposit cadence.
This clustering reveals concentration ratios that manual review would take hours to calculate. If a model identifies that 60% of a merchant's deposits originate from a single source, the underwriter receives an immediate flag before the deal progresses. The analysis happens in minutes, not hours, and it is consistent across every deal regardless of which analyst handles the file.
Seasonal Volatility and Revenue Decay Detection
AI models trained on MCA portfolio data can score merchants for seasonal volatility and revenue decay risk. A restaurant in a tourist town might show strong deposits from May through September but drop 40% in the winter months. A landscaping company might have the inverse pattern. These seasonal swings are manageable for small advances but become dangerous for large ones where the repayment timeline spans multiple seasons.
Revenue decay detection goes further, identifying merchants whose month-over-month deposits are declining even when individual months still look acceptable in isolation. A business depositing $150,000 this month, $140,000 last month, and $130,000 the month before is on a trajectory that should change the underwriting calculus for any deal above a certain threshold. In 2026, the funders who are catching these patterns early are the ones building durable portfolios.
Visual Verification Through Live Banking Session Recordings
The most sophisticated layer of concentration risk detection comes from watching how a merchant's banking portal actually looks during a live session. Asynchronous screen recording, the approach Exact Balance uses, lets applicants record their banking portal at their convenience while AI guides them through the required steps. The underwriter then reviews the recording on demand.
This visual evidence reveals details that no PDF or API feed can capture. The underwriter sees the actual transaction list in the bank's interface, can observe account naming conventions that suggest related-party transactions, and can spot patterns like large round-number deposits that indicate intercompany transfers rather than genuine customer revenue. For big-ticket deals, this layer of verification transforms the underwriter's confidence from hopeful to informed.
Matching Verification Infrastructure to Deal Size
The 1 Global Capital collapse was not caused by a lack of data. It was caused by a mismatch between the size of the deal and the depth of the verification applied to it. That mismatch persists across the industry today, particularly among funders scaling rapidly into larger transactions.
Milestone Capital Partners recently closed an $11.5 million corporate note to support its specialty finance platform's growth. Credibly has built tokenized loan pool structures. Fund Street issued a $45.5 million investment-grade note. These are funders deploying institutional-scale capital through MCA and adjacent products. The verification infrastructure supporting those deployments needs to be proportionally rigorous.
For funders at this scale, the cost of a single bad deal is not just the lost principal. It is the impact on facility covenants, investor confidence, and the ability to raise the next round of capital. As our analysis of concentration risk in big MCA deals details, the verification gap widens as deal size increases because the incentive to close quickly grows alongside the risk.
Exact Balance addresses this gap by making thorough verification fast rather than making fast verification seem thorough. The asynchronous workflow eliminates scheduling friction, so merchants can complete their banking session recording in minutes. The AI-guided process ensures every required screen is captured. And the underwriter can review the recording at their own pace, pausing, rewinding, and scrutinizing the details that matter for concentration risk assessment. Speed and depth stop being opposing forces.
Frequently Asked Questions
What is concentration risk in MCA lending?
Concentration risk in MCA lending refers to the danger of a merchant's revenue depending heavily on a small number of customers, a single industry, or a specific geographic market. When a large portion of the merchant's deposits come from one source, the loss of that source can make the advance unrecoverable. This risk is especially acute in big-ticket deals where the advance amount represents a significant portion of the funder's portfolio exposure.
How do MCA lenders verify bank statements for large deals?
For large deals, leading MCA lenders go beyond reviewing PDF bank statements. They use AI-powered transaction analysis to cluster deposits by source and identify revenue concentration. Many also require visual verification of the merchant's live banking portal through screen recordings. Platforms like Exact Balance enable this by having applicants record their banking session asynchronously, providing the underwriter with video evidence of actual account activity rather than a static document that could be altered.
Can AI detect fake or manipulated deposits in MCA applications?
Yes. AI models can identify patterns consistent with fabricated deposits, including round-number amounts deposited at regular intervals, deposits that appear and are immediately withdrawn, and transaction descriptions that do not match typical business activity for the merchant's industry. When combined with live banking session recordings, AI verification makes it significantly harder for applicants to present a false picture of their financial health.
What additional verification should funders require for big-ticket MCA deals?
Big-ticket MCA deals should include extended statement review periods covering at least six months, AI-powered concentration analysis identifying revenue dependency on specific customers, live banking session verification through screen recording, and cross-referencing of deposit patterns against the merchant's stated business model. The verification depth should scale with the advance amount, not remain constant regardless of deal size.
Conclusion
Concentration risk has already destroyed entire MCA portfolios, and the industry's memory is shorter than its ambition. As funders push into larger deals backed by institutional capital, the verification infrastructure supporting those deals must evolve beyond PDF reviews and API balance checks. AI-powered bank verification, from transaction clustering to live session recordings, gives underwriters the tools to catch revenue fragility before it becomes a portfolio-level crisis.
Exact Balance was built for exactly this workflow. Asynchronous screen recordings eliminate scheduling delays while AI guidance ensures every critical screen is captured. Underwriters review on their own time, with full audit trails for compliance. Visit exactbalance.ca to see how async bank verification fits into your underwriting process for deals of any size.