Manually updating spreadsheets. Dealing with paper jams in the printer. Remember what office life was like in 2005? If you’re feeling nostalgic, you can find many of the same practices still in place in the investor’s reporting offices of loan servicers today.
But why haven’t these offices evolved with changing technology?
It’s partly cultural, but it also reflects how the industry has developed over time. Many of our systems were never designed with data transparency or real-time automation in mind – some servicing platforms even predate the invention of Excel.
When I started in the mortgage industry in the 1990s, data management remained largely manual – we were still using paper ledgers and basic spreadsheets. Excel began gaining traction as companies transitioned from DOS-based to Windows-based systems. At the time, it was revolutionary, giving us far more capability and flexibility than we’d ever had before.
That revolution changed how people worked. Over time, nearly every department learned how to improve their processes, build automation, and quickly solve problems using Excel. These practices became a natural way of working and eventually turned into part of the very culture of investor reporting.
Fast forward 15 to 20 years, and many of the same people who relied on those methods are now industry leaders who’ve passed that knowledge and mindset to the next generation, ensuring that spreadsheets remain embedded in our operational DNA.
The comfort of visibility and the cost of overreliance
There’s also an element of trust in play. People feel more comfortable with what they can see and verify, even if it’s inefficient. A spreadsheet, after all, gives users full visibility into their data.
But that transparency can be deceptive. We’ve all seen how a small formula error affecting just a fraction of a percent can become a costly problem when applied to thousands of rows of data. Some spreadsheets have become so complex that, even though you can technically trace each formula, they exceed our ability to truly understand them.
The challenge is helping organizations recognize that automation doesn’t mean losing control. It means transferring control, shifting from manual processes with limited data integrity and capability to systems that ensure accuracy, visibility, and simplicity. True automation doesn’t obscure the data; it clarifies it, freeing teams to focus on what truly matters.
While Excel still provides some level of scalability, it only scratches the surface of what’s possible with modern, managed systems.
Why outdated workflows persist
So, what keeps these legacy workflows alive? In my experience, the biggest culprits are data fragmentation, limited system interoperability, and dispersed systems of record.
Servicing data lives across multiple platforms – core servicing, accounting, cash management, investor data, and client-specific reporting templates – but these systems don’t communicate cleanly, or sometimes even at all. Even when technology vendors promise automation, their solutions can often sit on top of inconsistent data. As a result, teams still end up manually reconciling the results.
There’s also an understandable resistance to risk. In investor reporting, even a small error can have regulatory or reputational consequences. Many organizations choose certainty over efficiency, at least until they see that modern automation can deliver both.
And then there’s a cost. Excel is affordable, and most servicers don’t recognize the hidden cost of poor quality in investor reporting. Many servicers still view investor reporting as a purely external obligation – a compliance deliverable, not a strategic asset. But that mindset overlooks a major opportunity.
By harnessing the power of all the data collected through investor reporting, servicers can uncover valuable insights to improve upstream and downstream operations, from loss mitigation and foreclosure to cash management and loan boarding. The ROI becomes clear when organizations view investor reporting as an internal catalyst for improvement, not just an external requirement.
Using AI to replace manual processes
Servicers can employ expert systems enhanced by AI that ingest and analyze hundreds of reports from the servicing platform, investor records, and other related data sources in parallel. These systems use comprehensive reconciliation, triangulation, and data validation to detect even the smallest anomalies, effectively flagging issues that often go unnoticed or require significant manual rework when identified by traditional processes.
Unlike tools that automate only surface-level tasks, these new expert systems that are enhanced by AI use advanced, rules-based decisioning to automatically resolve nearly 80% of exceptions and edits. Reporting analysts can therefore shift their focus to more complex or high-value research, knowing that the data remains clean, reliable, and aligned across systems. AI driven insights suggest potential research paths based on the patterns they’ve detected. This guidance not only increases efficiency but also leads to more consistent and accurate conclusions.
Quality results require quality data
High-quality automation starts with high-quality data. Too often, organizations invest in new systems without addressing underlying data issues. And that’s where breakdowns occur.
Automated platforms should strengthen, not replace, human oversight. A well-designed system helps identify and resolve inconsistencies by triangulating information across multiple data sources. While that adds some initial effort, it creates a critical feedback loop: operational teams can pinpoint and correct data issues or improve processes at the source, driving greater efficiency and accuracy over time.
As the underlying data becomes more reliable, automation delivers even better results. Cleaner data produces clearer feedback, and clearer feedback further improves data quality, creating a positive cycle that compounds over time.
This ripple effect goes well beyond investor reporting. Servicers can leverage such feedback to drive operational excellence across the broader organization. As data integrity improves, the entire servicing operation benefits.
How automated systems can earn trust
Ultimately, automation succeeds or fails on one factor: trust.
Once data is consistent and validated, organizations can more confidently introduce automation. While a system can technically operate on bad data, trust in its outputs will erode quickly when the results aren’t reliable.
Trust is earned through transparency, traceability, and performance. Teams need clear visibility into the data used, every processing step taken, and the reasoning behind each result. Over time, consistent efficiency gains, auditability, and alignment with user judgment reinforce confidence – making automation not just accepted but relied upon.
Modernization isn’t about abandoning the tools that got us here; it’s about evolving how we use them. Spreadsheets gave this industry its first taste of digital empowerment. But the same trust and visibility that made Excel revolutionary have also made it difficult to move away from.
As we enter a new era of intelligent automation, the mindset shift is already underway. Automation isn’t about taking control from people. It’s about giving them better tools. Tools that ensure accuracy, improve oversight, and free them to focus on higher-value work.
When we can see automation as an extension of our expertise rather than a threat to it, that’s when real progress begins. By eliminating spreadsheet-based exception management and reducing reliance on end-user computing tools, expert systems become foundational solutions, strengthening current workflows while positioning organizations for more advanced AI adoption in the future.
Jeff Choi is the COO at PMSI.
This column does not necessarily reflect the opinion of HousingWire’s editorial department and its owners. To contact the editor responsible for this piece: zeb@hwmedia.com.
English