Data and document management has become an extremely important topic for mortgage servicers in recent years. Not only are data and document management solutions important for servicers to achieve operational efficiencies, but they are also practically a requisite for meeting compliance.
To learn more about how data and document management solutions are helping mortgage servicers meet compliance and improve customer service, MortgageOrb recently interviewed Jonathan Kunkle, president of GuardianDocs, a division of LenderLive Network Inc., and Joe Dombrowski, director of product management for Fiserv.
Q: Why has data and document management become such an important topic for mortgage servicers in recent years?
Kunkle: Regulatory pressure has increased to ensure that the consumer is being treated correctly throughout the lifecycle of a residential loan. When onboarding a new loan or transferred loan, there is increased focus on ensuring the correct terms of the contract are loaded into the system of record the first time to make certain the application of each mortgage payment and timeliness of response to changes in terms (i.e., a scheduled adjustment within an adjustable-rate mortgage) are handled appropriately.
Further, the servicer has to ensure that impounds for escrows are correct and insurance and tax payments are properly scheduled. And, on top of that, the investor reporting and payments rely on all of this data being accurate. If it’s a transferred loan, there is a limited window of time when a servicer can make a claim for missing information or docs; once that window is closed, the new servicer has no recourse for the prior servicer’s transgressions. From the consumer’s perspective, he or she may be oblivious (up until the point something possibly goes wrong), but to the servicer, there is great pressure.
Dombrowski: Data has exploded in both volume and variety. The volume of data that a servicer generates and collects has increased over 250% in the last five years. This includes data about nonperforming loans particularly. To be successful in today’s environment, a servicer must manage its data well. This includes data in its servicing system, its communication systems (especially email and online chat), its phone system (data collected via voice mail messages), and both digital and paper document management systems. The touch points for data sharing between servicers and trade partners, borrowers, and others, along with risks involved in handling data, require a servicer to have a comprehensive policy regarding data and document management.
Servicers certainly have picked up best practices from their origination counterparts; however, servicing operations have a unique set of challenges due to the time frame for holding data and documents and the different regulatory requirements for data gathering and communicating. Although the Consumer Financial Protection Bureau’s (CFPB) “Know Before You Owe” requirements really changed data and disclosure for originations, servicers were already working with generally defined “servicing file” requirements from the CFPB and multi-page billing statements.
One nexus point for originations and servicers is the onboarding of loans once closed into the servicing system. There is always the quality control aspect to ensuring the data is transferred accurately and documents are on file to back that data up, including documents that arrive after the closing date. Further, transfers of servicers require similar tools, processes and due diligence to ensure an issue-free experience for borrowers and investors, as well as regulators. A recurring issue for portfolios that have traded hands is the evidence of ownership chain of documents, called assignments. This has been highlighted in foreclosure cases as an area where better tools and process improvements are needed.
Q: What are the features and capabilities that servicers need most from today’s data and document management solutions? Why are these features and capabilities now important?
Kunkle: The answer to this question is, “It depends.” One servicer may have two or three (or more) systems of record (a.k.a. servicing platforms), and each of them may have different onboarding requirements, including the data required and the data elements used for that data. The servicer’s primary focus should be to ensure it has data comparison capabilities. What’s the expected value, and what is the truth? Outside of payment histories, the source of truth is in the loan and supporting docs. There aren’t significant dollars available to a servicer to throw at collecting and perfecting that data, so it also needs a semblance of what’s mission critical, really critical, and what’s nice to have.
Dombrowski: Basic data management solutions should have a way to monitor all changes to data – what it looked like coming in to the servicer (both image and data point); who changed the data, showing a complete audit trail of the before and after values with traceability to each touch; and a way to revert data back to a prior authorized version.
Basic document management systems must have those same tools, as well as a way of tracking when documents are viewed, printed, shared via email or link, etc. Better data and document management solutions will have some level of workflow automation to make the flow of data between systems easily managed to reduce risk. Superior data and document management solutions will integrate the data found in documents to data in the servicing system. This will encompass the ability to read data in a document, compare that data with data found in the servicing system and make changes based on a set of conditions. Data that doesn’t match and doesn’t meet a set of conditions should be promoted to a person to review and authorize or decline the change.
Loan servicing transfers and loan boarding processes generally include some type of document review. Many servicers will select a representative sample of loans to compare documents with data. This manual approach is costly and risky, as it is subject to a human performing a “stare and compare” routine, which can be both mundane and fatiguing. A better approach, which will offset this risk, is to employ tools that will read a document, such as a mortgage note, and then compare the data found in the servicing system and make changes, such as changes to mismatched mailing addresses or other low-impact data, while alerting quality review staff to discrepancies, such as mismatched interest rates or loan amounts.
Q: What are the advantages/disadvantages of outsourcing data and document management to a third-party provider? If a servicer decides to outsource this function, will it need to keep some data and document management expertise in-house?
Kunkle: If the servicer chooses a vendor or platform wisely, there are limited disadvantages and a significant upside. Keep in mind that all automated document recognition (ADR) and optical character recognition (OCR) platforms run on a standard open source code that has been in existence for quite some time. A good vendor can run multiple ADR/OCR components within a single application, allowing the servicer to benefit from the age-old equation of using the right tool for the right job.
Conversely, if the servicer chooses to insource the build and installation of ADR/OCR, it is then wedded to the technology source, with the prospect of an additional investment if it doesn’t work out or if requirements change. In an outsource model, that vendor is constantly evaluating the market, looking for a better tool. Then, not only is the implementation of that tool shared across all clients, but all clients also benefit from the advantages of using the best one for the job.
Moreover, if the servicer chooses the right vendor, there is a very limited risk in outsourcing 100% of the business to a sole provider. In fact, there may be more risk and cost in trying to manage more than one. However, most large servicers and banks want to control enterprise content management (ECM). So, even if the vendor is an exclusive provider, the image capture and recognition of the documents is an upstream process. Each image is copied to the ECM, regardless. This, in turn, creates a robust disaster resilience strategy as an extra copy that is still available in the vendor’s system in the case of an emergency.
Dombrowski: In a highly networked world, we outsource many functions not only in businesses, but also in our personal lives. Many readers of this publication outsource their tax data to an online service each spring as they plow through questions gathering data to ensure a speedier and more beneficial tax return statement. Most servicers outsource some aspect of their businesses today to gain efficiency or lower cost.
There are three fundamental elements in mind when considering outsourcing data and document management: philosophy, privacy and practicality. First, a servicer must be willing to consider that some portion of its data will be outside its walls. Its business has a philosophy about data, perhaps unspoken, but there is a working set of values and beliefs at play in the company. If a servicer is using a hosted servicing solution, it has already embraced the philosophical approach of allowing key data outside of its walls. It may be using a hosted document provider. There, too, it has made a decision to allow key data outside its walls.
Privacy concerns and how that data is handled while being used and at rest is an element to consider. Risk officers or staff involved in assessing risk will have a hand in the decision to outsource. Risk mitigation must be weighed to determine if there is an acceptable level of risk with managing the data either externally or internally.
Finally, a servicer must determine if it can effectively operate and grow its business with the data and document management tools it has today. The practicality of working with internal and external systems embraces the ability to expand, costs involved, disaster recovery, education, monitoring and oversight or management.
The best approach balances considerations about privacy and security, always-on access, and the ability to rapidly change. Regardless of outsourcing or keeping in-house, there should always be someone on staff internally who understands where the data is collected, stored and used. This individual or group should have data paths mapped out and a basic level of expertise to monitor any external data and document management.
Q: What do you think are the biggest challenges facing data and document management vendors serving the mortgage servicing industry today? Are there features and capabilities that servicers are asking for that simply can’t be met as of yet? How can things be further improved?
Kunkle: The biggest challenge is cost. The technology that exists today extracts 100% of every piece of data from the source documents, compares it with the expected values and creates workflow applications to route exceptions to the individuals best geared to solve for that discrepancy. On a servicing transfer, the investor may have to only pay the subservicer $20 in total to onboard each loan. By the time the servicer deploys its resources to manage the data transfer and onboarding of that data, its resources just managing this can easily cost the subservicer more than $20 per loan. Adding in the ability to then compare the provided data with the source of truth becomes a cost center, not an advantage. However, the savvy servicer can look at the return on investment on the costs of onboarding a loan incorrectly, the compliance and reputational risk of doing it wrong, and the costs of cleaning up a mess that could have been prevented.
Dombrowski: As a solutions provider, we focus on staying ahead of bad actors to protect data and documents while supporting efficient means to manage and access data and documents. Having that data available to a servicer as needed and being able to still protect it from risks is a primary focus. Many servicers regularly evaluate their data and document management needs, and many providers are able to offer them solutions that fit their requirements. There are not many new features and capabilities that cannot be addressed.
However, a nascent idea in loan servicing is whether data can be employed to improve operations. With advances in analytics and harvesting data from documents, along with better information about data (metadata), there could be ways to eliminate and reduce risk, improve processes, and inform better decisions. Servicers are beginning to think along these lines, and industry providers are stepping up to the challenge. As servicers embrace data science, the industry will begin to see improvements in the quality of service.