Vendor complexity in mortgage fulfillment shows up in one of two ways, and most lenders recognize whichever one they are currently living with.
The first is a BPO problem. Many outsourcing vendors cover only part of the loan lifecycle. A lender brings one vendor in for income verification, discovers that vendor does not touch quality control, and now needs a second BPO relationship to cover the piece the first one misses. Add underwriting support or post-close review, and a lender can end up with three or four outsourcing partners, each responsible for a slice of the file, none accountable for the whole thing.
The second is a technology problem. A lender implements a system to solve one bottleneck, then implements another system for the next one, and ends up owning a stack of platforms that all need to be paid for, monitored, and maintained separately. Some of these systems work well together. Some do not. Either way, fragmentation persists. It has simply moved from BPO relationships to technology licenses.
Both paths lead to the same place: a lending operation managing multiple relationships to do the work of one connected process, and absorbing the cost of the seams between them without ever seeing that cost itemized anywhere.
For risk and compliance leaders, the seams carry a second cost. Every additional vendor is another third-party relationship to vet, monitor, audit, and defend in front of regulators and investors. Oversight obligations scale with the vendor count. So does the audit surface.
The pressure to address this is now measurable. In a July 2026 Mortgage Collaborative member survey, 86 percent of lenders named reducing loan production costs their top operational priority, and 64 percent named vendor and technology consolidation, making it the second-highest priority on the list. Investment bank Houlihan Lokey reached a similar conclusion from the supply side in a May 2026 white paper, finding that rising regulatory and cybersecurity requirements are raising vendor operating costs and pushing lenders toward multiproduct platforms.
The Problem Underneath Both: True Cost is Hard to See
Here is what makes both paths harder than they need to be. Few lenders or servicers have a reliable number for what it costs to perform these functions in-house. Without that baseline, sourcing decisions rest on estimates.
STRATMOR Group has made this point for years: executives need benchmarking data from credible sources to know whether the metrics they track are as strong or as weak as they believe, because peer data is what gives a lender’s own numbers meaning. The stakes behind that observation keep growing. Per-loan production expenses reached $11,898 in the first quarter of 2026, according to the Mortgage Bankers Association’s Quarterly Mortgage Bankers Performance Report. And as MBA Vice President of Industry Analysis Marina Walsh observed of the same data, “disparities between the top and bottom performers remain wide.” Which side of that disparity a lender sits on is difficult to know without a trusted internal cost number.
A lender might be paying more for BPO services than the work would cost done well internally, simply because no internal benchmark exists to compare against. Or a function might look inexpensive on a budget line while the real cost shows up downstream: repurchase demands, indemnification requests, defect cure costs, rework, and findings that surface in post-close QC after the loan has already been sold. Both are the same gap wearing different clothes: a sourcing decision made without knowing the true cost of the alternative.
This is where vendor transparency becomes more than a nice-to-have. A vendor that is honest about cost, including what a function actually costs to do well and where its own pricing sits against that number, gives lenders the missing input. Transparency turns a sourcing estimate into a sourcing decision.
The Fix Starts With The Right Vendor
None of this argues that specialized vendors do bad work, or that consolidation for its own sake solves anything. A lender who replaces five fragmented relationships with two equally disconnected ones has renumbered the problem.
The durable fix is a vendor built to cover both sides of the equation: technology and services, across the full lifecycle of the loan, from the same partner. The reason is easy to miss when evaluating vendors one at a time. Technology alone leaves lifecycle gaps. It automates one piece of the process and leaves the handoffs on either side exactly as fragmented as before. STRATMOR principal Tom Finnegan reached this conclusion as far back as 2021, writing from STRATMOR and MBA benchmarking data that “very little, if any, correlation exists between the money put into automation” and lower fulfillment cost per loan. STRATMOR’s subsequent research helps explain why: of roughly $11,000 spent to produce a loan, only about $500 to $1,000 goes to technology, while roughly 60 percent of the cost to close goes to people. Automating one step and leaving the handoffs untouched moves the bottleneck.
Services alone leave technology gaps for the same reason in reverse. Trained staff get added around the edges of systems that remain disconnected from the next one in line.
The two function as a real solution when one partner owns both, understands the full lifecycle from application through close (or through servicing, for lenders extending the relationship further), and can say honestly where automation reduces cost, where trained professionals outperform a system, and what each option is costing the lender today.
What the right mix looks like
In practice, this means evaluating fulfillment function by function across the loan lifecycle, and asking a different question at each stage.
Where is the work repetitive, document-heavy, and rules-based? Document classification and data extraction, income calculation, audit sample selection, and prefunding checks against GSE requirements all qualify. That is where technology should carry the load, with the added benefit that a platform performs those checks the same way on the ten-thousandth file as on the first, and leaves an audit trail while doing it.
Where does a file require judgment? A nuanced read on a self-employed borrower’s income. A QC exception that needs context a system does not have. A corrective action plan that has to hold up in front of an investor or a regulator. That is where trained professionals add more value than any current AI tool, and where the right vendor keeps people in the loop by design.
A platform performs the same check the same way on the ten-thousandth file as on the first, and leaves an audit trail while doing it. A corrective action plan still needs a professional who can defend it.
A lender working with a fragmented stack rarely gets to ask this question cleanly, because the answer depends on cost and capability data scattered across multiple vendors who have no reason to compare notes. A lender working with one partner who owns technology and talent across the full lifecycle can see the tradeoff and make a real decision about it.
Consolidation means seeing the whole cost
The lenders rethinking their vendor relationships in 2026 are chasing something more fundamental than a smaller contract count. They want a cost per loan they can trust, built on a clear view of what each function costs to do well, whether that function lives in a platform or with a trained analyst.
STRATMOR’s guidance on third-party relationships applies directly: “transparency is non-negotiable.” Best-in-class partnerships give lenders full visibility into work performed on their behalf, down to the audit logs and documentation trails that stand up to regulatory scrutiny. Adding a sixth vendor to coordinate the other five multiplies the oversight burden instead. Visibility comes from a partner with the technology and the domain expertise to cover the full lifecycle of the loan, who is transparent about what things cost, and who can show a lender where the money is actually going.
The full cost of vendor sprawl runs well past the licensing fees and the BPO invoices. It includes the blind spot those arrangements create: the gap between what a lender believes it is spending and what the work actually costs, in cash, in rework, and in risk. Closing that gap is the argument for consolidation. The right vendor comes first. A smaller vendor count follows on its own.
