How OnDeck's Small Business AI and Growth Data Reshape MCA Underwriting Best Practices

Small Business AI Adoption Is Changing What MCA Underwriters See

OnDeck's newly released report, titled "Small Businesses Lean Into Growth and AI," paints a picture of an applicant pool that looks nothing like it did even two years ago. Small business owners are not just using AI for marketing or customer service. They are using it to manage cash flow projections, generate financial reports, and optimize the very banking data that MCA underwriters rely on to make funding decisions. For funders following MCA underwriting best practices, this shift demands a serious rethinking of how bank verification actually works.

The report, published via PR Newswire, highlights that a growing share of small businesses are in active growth mode, taking on more capital, expanding operations, and leaning on technology to move faster. For MCA lenders, this creates a double-edged reality. More merchants need funding quickly, which drives deal volume. But the same AI tools that help merchants grow also give them the capability to present cleaner, more curated financial profiles, sometimes masking the very risk signals underwriters need to catch.

This article breaks down what OnDeck's findings mean for MCA verification workflows, why traditional document-based approaches are losing ground, and how async video-based bank verification closes the gap that AI-generated documents have opened.

Why AI-Savvy Merchants Break Traditional Verification Workflows

AI-Generated Financial Documents Look Increasingly Real

The uncomfortable truth for MCA underwriters in 2026 is that AI-generated bank statements, profit and loss reports, and cash flow summaries are now nearly indistinguishable from authentic documents at a glance. Tools that were once limited to basic formatting can now replicate institution-specific layouts, transaction patterns, and even balance trajectories that mimic real business activity. A merchant who understands what a funder looks for, consistent daily deposits, low NSF frequency, healthy average balances, can use generative AI to produce statements that check every box.

This does not mean every applicant is committing fraud. Most are not. But the existence of these tools means that AI document verification for lending must go beyond surface-level checks to remain effective. Pixel-level document analysis helps, but it is fundamentally a reactive approach. Fraudsters iterate faster than detection models update.

Growth-Mode Merchants Present a Unique Risk Profile

OnDeck's data shows that small businesses are not just surviving; they are actively pursuing expansion. Growth-mode merchants tend to carry higher expenses, more variable revenue, and shorter operating histories in new business lines. For MCA underwriters, this creates a tension. A merchant with rapidly increasing deposits might look like an ideal candidate on paper, but those deposits could be funded by another advance, a short-term contract, or even a personal loan being run through the business account.

Traditional bank statement review struggles to distinguish between organic revenue growth and manufactured deposit activity. The numbers can look identical. What separates a genuinely growing business from a stacking risk is the context behind the numbers, context that only becomes visible when an underwriter can watch a merchant navigate their live banking portal in real time.

Async Video Verification Closes the Gap

This is precisely why async screen recording of live bank portals has become a critical layer in modern MCA underwriting. When an applicant logs into their actual banking environment and navigates through account summaries, transaction histories, and balance details while being recorded, the resulting video provides verification signals that no static document can replicate.

Consider what a screen recording reveals that a PDF does not: the URL in the browser bar, the session-specific elements of the banking portal, the natural scroll behavior through transaction lists, and the consistency between different account views. An applicant who has fabricated a bank statement cannot easily replicate the live banking environment of a major Canadian or American financial institution. The friction of faking a live session is orders of magnitude higher than editing a PDF.

Exact Balance was built specifically for this use case. Applicants receive a secure link, record their banking portal at their convenience using browser-based screen capture, and submit the recording for underwriter review. No scheduling, no software installs, no live call coordination. The AI-guided recording coach walks applicants through each step and verifies completion in real time, ensuring that the footage captures exactly what the underwriter needs to see.

Adapting MCA Underwriting for AI-Era Merchants

Rethink What Counts as "Verified"

The OnDeck report should push every MCA funder to ask a basic question: what does "verified" actually mean in our workflow? If the answer is "we received a bank statement and it looked reasonable," that standard is no longer sufficient. Verification in 2026 must include provenance, meaning evidence that the financial data came from the actual banking institution, viewed in a live environment, by the actual account holder.

Async video verification provides this provenance. A timestamped recording of a live banking session, stored securely with a full audit trail, is a fundamentally different kind of evidence than a downloaded PDF. It is evidence of interaction, not just data.

Velocity Matters More Than Ever

Growth-oriented merchants do not wait around. OnDeck's findings confirm what every MCA broker already knows: the fastest funder wins the deal. But speed without verification discipline leads to losses. The genius of asynchronous verification is that it removes the scheduling bottleneck without removing the verification itself. An applicant can record their banking session at 11 PM on a Sunday. The underwriter can review it at 7 AM Monday morning. No calendars were harmed in the process.

This is a meaningful operational advantage. As we explored in our analysis of how OnDeck's 93% growth optimism data reshapes MCA underwriting, the funders who can maintain rigorous verification while matching the speed expectations of growth-minded merchants will capture a disproportionate share of quality deal flow.

AI Is Now on Both Sides of the Table

Perhaps the most important implication of OnDeck's report is that AI is no longer just a tool for lenders. It is a tool for applicants. Merchants are using AI to prepare their applications, optimize their financial presentations, and in some cases, fabricate supporting documents. This means the arms race between verification and deception has entered a new phase.

Funders who rely exclusively on document-based verification are bringing a knife to a gunfight. The merchants who present the cleanest-looking statements are not necessarily the best risks. They may simply be the most skilled at using AI to curate their financial image. Video verification of a live banking session cuts through this noise because it tests something AI document generation cannot easily replicate: real-time interaction with a bank's authenticated web application.

The Federal Reserve's small business lending data continues to show that alternative lenders serve a growing share of the market, which means the pressure on MCA funders to get verification right is only increasing. Losses from poor verification do not just hurt individual deal economics. They erode the institutional credibility that funders need to secure warehouse lines and attract capital partners.

What a Modern MCA Verification Workflow Looks Like

Consider a practical scenario. A broker submits a deal for a restaurant owner in Toronto who has been operating for three years and is expanding to a second location. The merchant's bank statements show strong deposit activity, rising balances, and minimal NSF transactions. Everything looks clean.

Under a traditional workflow, the underwriter would review the statements, perhaps run them through an OCR tool, and approve the deal. Under a modern workflow incorporating async video verification, the underwriter would also send the applicant a recording request through Exact Balance. The applicant would log into their RBC or TD banking portal, navigate through the requested date ranges and account views, and submit the recording.

During review, the underwriter might notice that the live portal shows a pending transfer from a personal account that inflates the business balance, something invisible on the monthly statement. Or they might confirm that the deposit pattern visible in the statement exactly matches what appears in the live portal, increasing confidence in the deal. Either way, the underwriter has made a better decision, and the entire process added perhaps 15 minutes of applicant time and 10 minutes of review time. No scheduling call. No time zone coordination.

As we analyzed in our piece on how SMB lending fraud is concentrating in MCA, the funders who build layered verification into their standard workflow are the ones best positioned to scale without proportionally scaling their fraud losses.

Frequently Asked Questions

How does AI adoption by small businesses affect MCA bank verification?

AI tools now allow small business owners to generate professional-looking financial documents, including bank statements that closely mimic authentic institution formats. This makes document-only verification less reliable for MCA underwriters. Funders should supplement statement review with live banking session verification, such as async screen recordings of the applicant's actual bank portal, to confirm that submitted documents match real account data.

What are MCA underwriting best practices for growth-oriented merchants?

Growth-mode merchants often show rapidly increasing deposits and higher expenses, which can mask stacking risk or unsustainable cash flow patterns. Best practices include verifying deposit provenance through live banking portal review, checking for transfers from personal accounts or other funding sources, and using asynchronous video verification to confirm that statement data matches the live banking environment. Speed remains important, but not at the expense of verification rigor.

Why is async video verification harder to fake than bank statement PDFs?

A bank statement PDF can be edited with widely available tools or generated entirely with AI. A live banking portal, by contrast, is an authenticated web application hosted by the financial institution. Replicating the full interactive experience of a bank portal, including session-specific elements, dynamic content loading, and institution-specific URL structures, is far more difficult than editing a static document. Async video recording captures all of these signals in a timestamped, tamper-evident format.

How does Exact Balance fit into an MCA verification workflow?

Exact Balance replaces live verification calls with asynchronous screen recordings. Funders create a verification request, the applicant receives a secure link, records their live banking session using browser-based screen capture with AI-guided coaching, and submits the recording. The underwriter reviews the footage on their own schedule, checks transaction authenticity, and marks the verification as complete. The entire process is tracked with a full audit trail for compliance documentation.

Conclusion

OnDeck's report confirms what forward-thinking MCA funders have been sensing: the applicant pool is getting more sophisticated, more growth-oriented, and more capable of presenting curated financial profiles. MCA underwriting best practices must evolve to match. Static document review alone no longer provides adequate fraud protection or risk assessment. Async video verification of live banking sessions adds a verification layer that is both harder to defeat and faster to execute than traditional phone-based bank checks.

Exact Balance provides this layer as a turnkey solution built specifically for Canadian and North American MCA lenders. If your team is still scheduling live verification calls or relying solely on document review, visit exactbalance.ca to see how asynchronous bank verification fits into your underwriting workflow.