How LendingTree's SMB Lending Recovery Signal Reshapes AI Underwriting for Merchant Cash Advance

LendingTree Just Confirmed What MCA Funders Suspected: The Rebound Is Here

During LendingTree's Q3 2026 earnings call, CEO Scott Peyree made a statement that should have every MCA funder and ISO broker paying attention: "Q2 was a trough in the performance of small business lending, and we did see strong sequential growth." He added that the company is "still on the path to getting back to our all-time records of Q4 of last year and Q1 of this year." For anyone operating in alternative lending, this is not a vague optimism signal. It is a leading indicator from one of the largest small business lending marketplaces in the country.

What makes this signal particularly relevant for the merchant cash advance space is timing. When LendingTree sees recovery, the downstream effect on MCA application volume typically follows within weeks. Brokers who source leads through marketplace channels start pushing more submissions. Merchants who paused their funding searches re-enter the pipeline. The question is not whether volume is coming back. The question is whether your underwriting infrastructure can handle it. AI underwriting for merchant cash advance is no longer a competitive advantage; it is the baseline requirement for funders who want to participate in the recovery without drowning in manual review backlogs.

Why a Lending Recovery Breaks Manual Underwriting First

The Mechanics of a Volume Spike

Lending recoveries do not ramp gradually. They spike. Merchants who delayed funding during a soft quarter submit applications in clusters, often within the same two-week window. ISO brokers who held back on marketing spend start dialing again. The result is a compression of demand that hits underwriting teams all at once.

Manual underwriting teams are sized for steady-state volume. A typical MCA funder with three to five underwriters can process a predictable number of deals per day when each file requires downloading bank statements, keying data into spreadsheets, cross-referencing deposits against reported revenue, and flagging inconsistencies by eye. When volume jumps 30 to 50 percent over a few weeks, the math breaks. Deals queue up. Response times stretch from hours to days. Merchants who need capital quickly move to the next funder on the broker's list.

This is the throughput ceiling that platform lenders have already solved through embedded data access. Independent funders do not have Shopify's transaction feed or Square's point-of-sale data. They have bank statements. The speed at which they can collect, parse, and analyze those statements determines whether they fund deals or watch them walk.

The Quality Mix Shifts During Recovery

One underappreciated dynamic of a lending recovery is that the quality mix of incoming applications changes. The first merchants to return are typically the strongest: businesses that paused discretionary capital raises during uncertainty but maintained healthy cash flow throughout. These are exactly the deals every funder wants. They also move the fastest because they know they are fundable and expect a competitive process.

Winning these deals requires sub-hour turnaround on initial qualification. That means collecting bank statements, extracting monthly revenue and average daily balance data, checking for NSFs, and identifying existing MCA positions, all before a human underwriter ever opens the file. AI-powered document extraction handles the parsing. Automated bank statement analysis surfaces the numbers that matter. The underwriter's role shifts from data entry to decision-making, which is where human judgment actually adds value.

LendingTree's recovery signal also suggests that weaker merchants will follow the strong ones into the pipeline within a quarter or two. When that happens, the ability to quickly distinguish between a $90,000-per-month business with clean deposits and a $90,000-per-month business with circular transfers and manufactured balances becomes critical. AI fraud detection trained on lending-specific patterns catches anomalies that a tired underwriter reviewing their fortieth file of the day will miss.

Async Document Collection as a Speed Multiplier

Speed in MCA underwriting is not just about how fast you analyze documents. It is about how fast you receive them. The traditional workflow, where a broker emails a merchant, the merchant downloads statements from their bank portal, the merchant emails them back, and the broker forwards them to the funder, introduces hours or days of latency at every step.

Async document collection eliminates most of that latency. Let's Submit generates a secure upload link that goes directly to the merchant. The merchant drops their last four bank statements, a government ID, a void cheque, and a signed application into one portal, often from their phone. No email chains. No file format confusion. No waiting for business hours. Documents land in the funder's pipeline ready for AI extraction the moment they are uploaded.

During a volume recovery, this collection speed is the difference between funding a deal on day one and getting a submission that three other funders have already quoted. LendingTree's data suggests the race is about to start again.

What AI Underwriting for MCA Actually Looks Like in Practice

The phrase "AI underwriting" gets thrown around loosely in fintech marketing. For MCA funders evaluating their options in 2026, it helps to be specific about what the technology actually does and where human oversight remains essential.

At the document ingestion layer, AI handles classification (identifying whether a file is a bank statement, a tax return, or a void cheque), extraction (pulling transaction-level data, monthly totals, account holder names, and institution details), and normalization (converting data from hundreds of different bank formats into a consistent schema). This is not generative AI writing creative text. It is computer vision and structured data extraction, trained on millions of real financial documents.

At the analysis layer, automated systems calculate the metrics that drive MCA decisioning: average monthly revenue, average daily balance, number of NSFs over 90 days, evidence of existing positions (daily or weekly debits consistent with MCA remittance patterns), and revenue trend direction. A clean application summary, like the one Let's Submit auto-generates showing legal name, average monthly revenue, average daily balance, NSF count, and time in business, replaces twenty minutes of manual spreadsheet work.

At the decisioning layer, the human underwriter reviews the AI-generated summary, checks edge cases, and makes the funding call. This is the correct division of labor. The AI handles volume and consistency. The human handles judgment and exceptions. Trying to automate the final decision entirely introduces risks that most independent funders are not equipped to manage, from regulatory exposure to model drift. But automating everything upstream of that decision is not optional anymore.

Square's lending arm, for context, has consistently beaten its 4% loss target for over a decade by relying on embedded transaction data for underwriting. Independent funders will never have that data depth. What they can have is the fastest possible path from raw bank statement to structured, verified application data. That is the gap AI underwriting fills.

How to Position Your Underwriting for the Volume Surge

LendingTree is not the only signal pointing toward recovery. Federal Reserve survey data on small business credit demand has shown sequential improvement across multiple quarters, and OnDeck's growth confidence report found that 93% of small businesses expect growth, the highest reading in recent memory. The convergence of these data points suggests that MCA funders have a narrow window to upgrade their infrastructure before the wave hits.

Practically, this means three things. First, move document collection off email and onto a structured, async platform. Every hour saved in collection is an hour gained in time-to-offer. Second, implement automated bank statement parsing that feeds directly into your underwriting workflow. If your team is still manually keying deposit totals from PDFs, you are paying senior underwriter salaries for data entry work. Third, build fraud detection into the ingestion pipeline rather than bolting it on after the fact. Fabricated cash flow patterns are easier to catch when AI reviews every statement at intake than when a human reviewer spot-checks files after a deal is already in committee.

The funders who will win the recovery are not necessarily the ones with the most capital. They are the ones who can process the most qualified deals per day without adding headcount. That is what AI underwriting for merchant cash advance delivers.

Frequently Asked Questions

What is AI underwriting for merchant cash advance?

AI underwriting for merchant cash advance refers to using machine learning and computer vision tools to automate the extraction, analysis, and verification of financial documents, primarily bank statements, that drive MCA funding decisions. Rather than having human underwriters manually key data from PDFs, AI systems classify documents, pull transaction-level detail, calculate metrics like average monthly revenue and NSF counts, and flag fraud indicators. The human underwriter then reviews the structured output and makes the final funding decision. This approach increases throughput, reduces errors, and shortens time-to-offer.

Why does LendingTree's lending recovery matter for MCA funders?

LendingTree operates one of the largest small business lending marketplaces in the US. When their CEO confirms that lending volume has troughed and is recovering, it signals that application flow across the broader alternative lending ecosystem, including MCA, is about to increase. Funders who are not prepared to handle higher volume efficiently will lose deals to competitors with faster turnaround times.

How does async document collection speed up MCA funding?

Async document collection removes the back-and-forth email chains that typically delay MCA applications. Instead of waiting for a merchant to email bank statements to a broker, who then forwards them to a funder, the merchant receives a direct upload link. They can upload statements, IDs, and signed applications from their phone at any time. The documents land immediately in the funder's pipeline, often with AI extraction already applied, cutting hours or days from the pre-underwriting process.

Can independent MCA funders compete with platform lenders like Shopify and Square on underwriting speed?

Independent funders will not replicate the embedded data advantages of platform lenders, who underwrite from real-time transaction data. However, by combining async document collection with AI-powered bank statement analysis, independent funders can dramatically close the speed gap. Platforms like Let's Submit automate the collection and extraction steps so that an underwriter receives a clean, structured application within minutes of a merchant uploading their documents, rather than hours or days later.

Conclusion

LendingTree's Q3 recovery signal is a clear warning to MCA funders: volume is coming back, and the funders who can process it fastest will capture the best deals. Manual underwriting workflows that worked during a soft quarter will buckle under a volume spike. AI underwriting for merchant cash advance, from automated document collection through intelligent bank statement extraction, is the infrastructure that separates funders who scale from funders who stall.

Let's Submit gives independent funders and ISO brokers the tools to compete: async upload links that collect documents in minutes, AI extraction that turns raw bank statements into clean application data, and a pipeline that ensures your underwriters only touch deals that are ready for a decision. Visit letssubmit.ca to see how async verification fits into your workflow before the next wave of applications arrives.