Lending's hidden cost is manual insurance verification

Somewhere between the application and the closing table, an underwriter is still calling an insurance agent to confirm a dec page. In 2026.

It's easy to shrug off, because each instance is small. A few emails. A phone call. A PDF to read and re-key. But mortgage production is a thin-margin business right now. According to HousingWire's summary of the latest MBA performance report, independent mortgage banks averaged a pre-tax production profit of $973 per loan in the second quarter of 2026 while spending $10,936 per loan to originate it. Costs have also stayed well above the long-run average of roughly $7,900 per loan since 2008, per the prior quarter's MBA data.

When profit per loan is under $1,000, every manual step carries weight. The trouble with manual homeowners insurance (HOI) verification isn't that any single touch is expensive. It's that nobody tracks the total.

Where the costs hide

Manual verification spreads its cost across people, systems, and timelines, which is why it rarely shows up as its own line item:

• Staff touches. Requesting documents, sending reminders, calling agents, reading declarations pages, checking coverage against requirements, and keying data into the loan origination system. Each step is minor. Together they add up on every file.
• Delayed or failed closings. 64 percent of lenders reported closing issues tied to home insurance either frequently or somewhat frequently over the prior year. Matic's CEO noted that lenders are seeing closings slip or fall through over insurance hurdles.
• Borrower friction. Borrowers are already juggling appraisals, title, and moving logistics. Every extra request for an insurance document is another chance for a stall.
• Errors and rework. Data keyed from a PDF can be wrong, outdated, or altered, and problems tend to surface late, in quality control, investor review, or servicing boarding. Manual data entry is a tax on team time that compounds.
• Post-closing blind spots. Insurance details collected by hand at origination are stale soon after closing, and servicing often has to collect them again. The downstream results are force-placed insurance and escrow disruption.
• Lost capacity. Every hour a processor spends chasing a dec page is an hour not spent on the next loan, which matters most when volume picks up.

A back-of-envelope model

You can estimate the direct labor cost with a simple formula: touches per file, times minutes per touch, times loaded hourly cost, times loan volume. The numbers below are illustrative assumptions, not benchmarks, so swap in your own.

• Touches per file: 4 (request, follow-up, review, data entry)
• Time per touch: 15 minutes, or one hour per file
• Loaded hourly cost: $50
• Volume: 1,958 loans per quarter, the average per company in the MBA's second-quarter 2026 sample

That works out to about $50 per file, or roughly $98,000 per quarter, before counting a single delayed closing, rework item, or borrower complaint. Against $973 of average production profit per loan, that one verification step would consume about 5 percent of profit on every file. Change the assumptions and the answer moves, but the exercise usually reveals a number larger than anyone expected.

Where manual and automated workflows diverge

A typical manual process looks like this:

• Request proof of insurance from the borrower or their agent.
• Wait, then follow up.
• Receive a declarations page or binder, then read it and compare it to requirements.
• Key the data into the loan file.
• Re-check as closing approaches.
• Hand off to servicing, which often starts the process over.

An automated process has fewer moving parts:

• The lender initiates a request from the mortgage workflow.
• The borrower connects their insurance account or uploads a policy document.
• MeasureOne retrieves or processes the policy information and converts it into structured data.
• Verified data is delivered directly into the lender's systems.

The page reports 95 percent coverage and support for all major insurance providers, along with more than 4.2 million documents processed. Because borrowers can either connect an account or upload a document, the process doesn't depend on a single path working for everyone. The second advantage is what happens next: the same data foundation supports verification after closing, not just before it.

What to measure before you automate

You can't build a business case on numbers you don't have. A short baseline goes a long way:

• Touches and minutes per file spent on insurance verification
• Average days from application to insurance-cleared
• Share of files with open insurance conditions near clear-to-close
• Closings delayed or lost because of insurance
• Borrower follow-ups per file
• Post-closing lapses discovered late, and force-placed policy volume
• Servicing calls tied to escrow changes from insurance

Making the business case

The goal of automating HOI verification isn't to remove people from the process. It's to stop paying skilled staff to retype data that a system can retrieve. Manual insurance verification is one of the last analog steps in otherwise digital workflows, but its verification fits into the wider origination stack. For lenders already automating income and employment checks, insurance belongs in the same workflow.

The most persuasive ROI case uses your own numbers. Bring your touches per file, monthly volume, and delay rates, and compare them with an automated workflow that includes intelligent document processing for policy uploads and insurance monitoring after closing. 

Verification is only half of the insurance data story, though. Lenders aren't the only ones working with incomplete information. In our next post, we'll look at what insurers are missing without verified borrower data, and why that gap matters to both sides of the transaction.