Manual Verification is a Hidden Tax on Lender Profitability
John Hardesty is SVP of Revenue with Argyle.
Mortgage lenders spend much of their time reacting to what they cannot control. Rates shift daily. Volume rises and falls with market conditions. Inventory constraints shape borrower behavior. These forces deserve attention, but they can also obscure what’s happening inside a lender’s own operation—inefficiencies that, unlike rate movements or inventory shortages, are actually within their power to fix.
One of the clearest examples is manual verification. It slows cycle times, increases labor and adds friction at nearly every stage of production. The cost to originate a mortgage now runs north of $11,500 per loan, according to Freddie Mac, making even small inefficiencies carry real weight. Spread across the workflow, those costs are easy to miss and even easier to underestimate.
Part of what makes this so persistent is how it gets rationalized. Teams assume the process is functioning well because experienced staff manage it and underwriters know the guidelines. But familiarity breeds complacency. These assumptions create blind spots. Manual steps that could be automated get accepted as inevitable, small delays accumulate as loans move through the pipeline, and the total cost of verification never gets properly measured or attributed because no one is tracking it that way. The first step toward fixing that is knowing where to look.
Three things to measure
Lenders that want to pressure test whether verification is a drag on their business should start with three core metrics: cycle time, pull-through rate, and number of file touches. The problem isn’t familiarity with these metrics, it’s that most lenders aren’t actually measuring them in any systematic way. When asked what their average cycle time is between key milestones, or how that changes with automated versus manual verification, most lenders will tell you they know they should be tracking it, but aren’t. That’s the gap worth closing.
Cycle time exposes where loans stall between milestones—not just application to conditional approval, but any stage in the pipeline where delays accumulate. The more useful question isn’t what cycle time is in the abstract; it’s whether you have a baseline for your own operation and whether you can see how the verification method affects it.
Pull-through rate shows how many applications actually convert into funded loans. According to the Mortgage Bankers Association, the average pull-through rate in the first half of 2025 was 69% for nonbanks and 55% for depositories, and both those figures have been declining since 2022. While multiple factors influence that trend, manual verification plays a role. Most lenders know the industry benchmark. Fewer can tell you their own number off the top of their head.
File touches capture how many times a loan is handled and by how many people. Manual workflows often involve multiple handoffs between loan officers, processors and underwriters. Each touch adds cost and increases the likelihood of rework. The question is whether anyone is counting.
A bonus metric worth watching
Beyond cycle time, pull-through rate, and file touches, there’s a fourth metric worth tracking: repurchase risk. According to recent research from Reggora and STRATMOR Group, the average mortgage loan repurchase rate from April 2023 to October 2024 was 0.49%—or roughly one in every 200 loans—at an average cost of $32,288 per loan.
The good news is that when income and employment are validated through an approved service provider, the GSEs will not enforce representations and warranties on the validated data, providing a meaningful layer of protection against buyback exposure. The bad news is that the teams managing repurchase risk usually operate in a different silo from those focused on operations and may not even be aware of the intrinsic connection between the two.
For lenders selling primarily to Fannie Mae or Freddie Mac, the percentage of loans eligible for rep and warrant relief is an early indicator of buyback risk worth tracking, since R&W relief protects against repurchase requests before they materialize. But when we ask, the answer is almost always the same: they know they should be tracking it, but they aren’t.
What’s at stake over the next 12-24 months
In today’s lower-volume, purchase-focused market, inefficiency already compresses margins. When volume returns — and especially when a refinance cycle kicks in — those same inefficiencies will limit a lender’s ability to scale. Refi borrowers are more sensitive to speed and more willing to switch lenders, which means delays that were merely costly in a purchase market become actively dangerous to pull-through in a refi market.
More than an operational detail, manual verification is a measurable, controllable driver of cost, speed, and risk. The lenders getting ahead recognize that manual verification isn’t costless just because it’s done in-house. They measure what it actually costs them in cycle time, pull-through, quality, and buyback exposure.
(Views expressed in this article do not necessarily reflect policies of the Mortgage Bankers Association, nor do they connote an MBA endorsement of a specific company, product or service. MBA NewsLink welcomes submissions from member firms. Inquiries can be sent to Editor Michael Tucker or Editorial Manager Anneliese Mahoney.)
