customer stories
Customer Stories
University Federal Credit Union surfaces discrepancies in the first hour and cuts underwriting scenarios 80%
Mortgage | Classify, Analyze, Inspect and Detect
University Federal Credit Union (UFCU) is a 90-year-old credit union based in Austin, Texas, serving members across Texas and beyond. UFCU turned to Ocrolus, a vertical AI workflow and analytics platform for lenders, to standardize income calculations and document indexing across a mortgage operation it had run manually for 25 years.
Today Ocrolus runs on every loan file before it reaches a processor, so loan officers get a recommended income number and see discrepancies within the first hour of receiving credit documents, and the underwriting desk has roughly 80% fewer income scenarios to work through.
Challenge
After 25 years of manual income worksheets, University Federal Credit Union needed a way to give loan officers confident income numbers without routing roughly 1,179 scenarios a year through the underwriting desk. Problems with a file tended to surface late too, once underwriting was already deep in it.
Solution
Ocrolus was made a required step on every loan file the moment credit documents arrive, so loan officers get an income number they can stand behind and see discrepancies before the file ever reaches underwriting, with classification, indexing and document integrity checks running on the same pass.
Result
• 80% fewer scenarios hitting the underwriting desk in the first week
• Undisclosed garnishment flagged in the first hour instead of five days later
• Over an hour saved per transaction across income scenarios, indexing and loan approval
• 30 to 45 minutes saved per loan on indexing alone, with the biggest gains on self-employed files
• Required on every loan file for consistency and fair lending
Challenge: Manual income worksheets pulled underwriters to the front of the file
For UFCU, every file moved by hand. Sales teams and processors gathered pay stubs and W-2s, filled in an income calculator worksheet, saved it as a PDF and sent it to underwriting, where the underwriter rebuilt the worksheet any time something did not reconcile.
“It was very easy to make mistakes because it was such a manual process,” said Erin Simmons, Senior Manager of Operations at University Federal Credit Union.
The bigger cost was where the work landed. Loan officers did not have confidence in the income numbers they were producing, so they sent scenarios to underwriting for a second opinion to be sure a pre-qualification would hold. And because nobody looked at the file closely until underwriting did, a problem like an undisclosed debt could sit undetected for days.
“We had roughly 1,179 scenarios cross our underwriters’ desk in 2025,” Simmons said. “We had roughly 294 hours of underwriting time that was being used on scenarios.”
Indexing was a second opportunity. Representations and warranties were a third. UFCU sells heavily to Fannie Mae, Freddie Mac and the Federal Home Loan Bank, but a workflow built around pay stub and W-2 checklists meant the team was using agency reps and warranties on far fewer loans than they wanted. Simmons and her underwriting manager saw the headroom and made it a requirement for any technology partner.
Solution: One standard, required on every loan file
UFCU spent months defining what it actually needed to solve before it started shopping: income calculations its loan officers could stand behind, a fix for manual indexing, and reps and warranties the team was not yet taking advantage of. Just as important was the ability to adopt in stages, without being pressured into automating underwriting before the team was ready.
Income came first. Ocrolus produces a recommended income calculation on files with the required documentation, and UFCU told loan officers they could issue a pre-qualification on that number without waiting for an underwriter, who can still accept or change it.
“When we use Ocrolus’ recommended income, our team can just move forward,” Simmons said. “Ocrolus is backing us up, so we’re going to back up that number.”
Classification runs on the same pass. Every document is sorted and named the same way on every file, which replaced a manual indexing process built up over years in Encompass.
Ocrolus Inspect then compares borrower documents against the application and credit data, and surfaces anything that could stop a deal before the file reaches processing or underwriting. A loan officer ran a borrower’s pay stubs and W-2s through Ocrolus and Inspect flagged a $1,200 per month garnishment not disclosed on the application.
“Inspect picked it up within the first hour of that loan officer having those credit documents from the member,” Simmons said. “It caught something that would have been caught roughly five days later had we not used Inspect and reviewed that tool up front.”
Address mismatches, name discrepancies and Social Security number errors surface the same way, which keeps the back-and-forth away from the member and the realtor. Automated conditions are the next phase for UFCU, so underwriters get guideline-cited conditions written for them.
Ocrolus Detect runs as a document integrity check, flagging documents that show signs of alteration. For a credit union without the layered review of a much larger shop, that mattered more than UFCU expected.
“We’re loving what Ocrolus is picking up on Detect, and it’s giving us a peace of mind that we have support behind us in the decisions we’re making,” Simmons said.
Adoption followed the same staged approach. The team got comfortable and confident with Ocrolus and then made it mandatory to use on every file.
“All loans are run through Ocrolus before they’re ever moved for processing,” Simmons said. “At the end of the day, it’s going to save us time, and it’s going to provide consistency when it comes to our income calculations.”
Result: Fewer scenarios, faster catches and time back in every department
The scenario backlog cleared almost immediately.
“In our first week, we saw the number of scenarios that were hitting our underwriting scenario desk drop by 80%,” Simmons said. “This was all because the loan officers had the tools right there at their fingertips. Huge win.”
Across the file, the savings compound.
“We’re saving over an hour per transaction by using Ocrolus,” Simmons said. “That’s through the income scenarios, through the indexing of the file and then on through the approval of the loan.”
The indexing gains turned out to be the ones UFCU had not planned for.
“The processors are saving anywhere from 30 to 45 minutes per loan on the indexing alone,” Simmons said.
Because every document is classified and named the same way, the benefit does not stop at processing.
“You’ve saved underwriting time, you’ve saved sales time, you’ve saved your compliance team time, closing, who knows exactly where documents are, and post-closing,” Simmons said. “We’ve saved far more time than we ever expected with Ocrolus.”
Fannie Mae reps and warranties moved too. The underwriting manager built training for sales and operations on Fannie Mae Day 1 Certainty, and the behavior changed.
“We’re already seeing huge wins with our reps and warrants,” Simmons said. “Our sales team is asking for less documentation because they’re realizing they don’t need all the documents we had been asking for for so long.”
Key outcomes:
- 80% fewer income scenarios reaching the underwriting desk in the first week of launch
- Undisclosed obligations, Social Security number errors and signs of document alteration surfaced within the first hour instead of five days later
- Roughly 294 hours of annual underwriting scenario time addressed at the source
- Over an hour saved per transaction across income scenarios, indexing and loan approval
- 30 to 45 minutes saved per loan on indexing alone, with the largest gains on self-employed borrowers
- Expanded use of Fannie Mae reps and warranties, with fewer documents requested from members
- One consistent process on every loan file, including pre-qualifications
Why University Federal Credit Union recommends Ocrolus
What UFCU wanted from a technology partner was not a feature list. It was a partner willing to work at the credit union’s pace.
“We were really looking for a partner that was going to customize the experience for us, that would slow down when we needed to slow down and speed up when we needed to speed up,” Simmons said.
That is how implementation ran, and it is why the team is comfortable taking on automated conditions and post-closing next.
“Ocrolus truly does feel like a partnership,” Simmons said. “They value what their partners need, and have a team that is ready, willing and able to put out a product that works for today and tomorrow’s environment.”
“In our first week, we saw the number of scenarios that were hitting our underwriting scenario desk drop by 80%. Ocrolus values what their partners need, and has a team that is ready, willing and able to put out a product that works for today and tomorrow’s environment.”
Erin Simmons, Senior Manager of Operations at University Federal Credit Union
