By Michael Detwiler

Is the industry at a turning point in choosing mortgage origination technology? One technology executive says yes.

There has been some banter in the industry (articles, roundtables, water-cooler discussions) relative to the health of certain mortgage origination platforms—traditional loan origination systems (LOSes) in particular. Some discussions among thought-leaders in the industry have gone so far as to suggest that the LOS, as we know it today, is dead. Is the LOS dead? After careful consideration, I have reached the conclusion that the mortgage software we have come to think of as the LOS is a dinosaur, and it's about to become extinct. Something similar has already happened in the manufacturing industry. Software designed to manage the individual components of an operation has been replaced by enterprisewide tools that make it possible to manage an entire concern much more profitably, but without sacrificing flexibility. Enterprise resource planning (ERP) and supply-chain management have revolutionized the business of manufacturing.

Now certain technology providers, such as my firm, Greenwood Village, Colorado–based Mortgage Cadence, are looking at financial services and concluding that at the end of the day we are just manufacturing mortgages. Exactly like the manufacturing of any other product, the loans we write require certain raw materials (data) and suppliers (settlement service providers) that make up the supply chain required to build a mortgage.

If we are just a different kind of manufacturing operation, can the mortgage lending industry follow a trail blazed by other industries leading toward an enterprisewide set of automation tools that will streamline our processes, squeeze costs out of our operations and increase our profitability?

Soon, lenders will discover that new breeds of mortgage automation, true enterprise lending solutions (ELSes), are already available, and in my judgment it will render the traditional LOS obsolete. The capabilities of these new systems will go far beyond what today's users have come to expect. In fact, the way the tools are architected will allow companies to leverage them in ways that users of today's LOSes are only dreaming about. It might surprise these lenders to know that other forward-thinking mortgage originators such as Alpharetta, Georgia–based NetBank Inc., Irvine, California–based Homefield Financial Inc. and Santa Fe, New Mexico–based Thornburg Mortgage Inc. are already using this technology and enjoying its benefits.

I believe the death of the traditional LOS is an unavoidable consequence of the new technologies that have emerged over the past few years. Not every vendor that talks about ELSes is capable of delivering on the promise, so I'll also give you a glimpse of this new breed of tool that mortgage originators will be using in the future.

A tool of the past
"Faster, better, cheaper."

These are the orders that have been coming down from executive offices all over the mortgage industry for at least the last decade. Up until the very recent past, it was fairly easy to deliver on these mandates. Lenders could gain efficiencies and lower costs by automating any function in the mortgage process. So over the past few years technologies have been developed, implemented and moved into production, providing point solutions for different parts of the mortgage process.

It took too long to have a person call the credit bureau and request a file, so we automated the ordering and delivery of credit reports. It was difficult to transfer the data from the report into the LOS without causing errors, so we built automation for the LOS to bring the liabilities in automatically. And so it went, all along the mortgage supply chain.

About midway through the last century, the American automotive industry found itself in a similar situation. Having developed powerful new machines, manufacturers were building cars faster, but they needed a better way to retool their lines in order to produce different products. In response, the industry invested in technology that in effect turned the entire assembly line into an automated tool.

Today mortgage lenders have some form of automation for most of the work in the mortgage loan origination process. All we have to do is press a button and just about any information, product or service we require can be ordered, received and pulled into our process. Actually, we should say all we can do today is press that button.

In the pursuit of faster systems that cost less to operate, we have automated ourselves into a box. Instead of our modern systems helping us do business better, they now tell us how we have to do business and what processes we have to follow. By automating the industry's old ways of doing business, we have made it difficult, if not impossible, to re-engineer our "production lines" for greater efficiency. So today's automation comes with a price—rigidity.

The reason the traditional LOS has been unable to keep up with the times is that it attempts to automate traditional lending processes, locking lenders into workflows that are no longer profitable and in some cases product-specific. The traditional process for originating a loan is very manual and task-based.

The traditional LOS tailors to this; therefore, it incorporates manual processes into the product design instead of automating these activities. Some mortgage professionals may not mind this, as it provides them with great job security; however, anytime you have to say the word "manual" or "work-around" in relation to your technology solution, you're surely not getting the full return on investment you were initially seeking.

It's not that yesterday's technology providers were creating bad software, no more than the Ford Model T was a bad car. But times change and, unfortunately, as the industry has added new products to its menu and new channels to its origination efforts, lenders have found a need for a more flexible, adaptable solution. Like the automotive industry, the mortgage business must now move up to the next level, to systems that can automate processes without human intervention.

One way that companies are trying to create a better solution out of these legacy tools is to build interfaces between the disparate systems in order to force them to share data. Sharing information seamlessly among modules is a worthwhile goal. Unfortunately the older architectures these systems were built upon don't respond well to that task, making it difficult and expensive. Seldom does the effort actually return the required results. For example, many legacy applications are closed-ended systems.

Additionally, they have their business rules and logic hard-coded into the application. It requires a lot of work to extract data to make them accessible to third-party systems. And when those data can't be extracted, it requires those rules to be duplicated in an external system, thus making versioning and maintenance a huge burden. Putting a new face on an old system is like applying a fresh coat of paint on a car that won't run: It costs money and the end result looks much better, but it still won't get you where you want to go.

Today, most of the LOSes in service are conglomerations of "best-of-breed" or "best-available" modules cobbled together to force the lender and its customer down a pathway that leads to the closing table. While getting to that destination is the end result all parties desire, everyone is forced to settle for the process that a particular platform requires. A quick glance at the mortgage industry's customer-satisfaction ratings will convince you that the system is broken. The fact that systems integration/consolidation is one of the leading information technology (IT) initiatives, according to a 2005 consumer survey of recent mortgage borrowers conducted by CFI Group USA LLC, Ann Arbor, Michigan, suggests that lenders are not satisfied with this approach.

Powering the lending enterprise
One of the most powerful concepts to come out of the Internet's explosive beginning was the idea of software as a service. Service-oriented architectures (SOAs) take advantage of this by enabling core systems to reach out and interact with other software applications, called services.

SOA makes it possible to integrate with any new technology that is developed. If a new credit module became available, for instance, an LOS built on an SOA would be able to reach out and interact with it with minimal work required by the lender's information technology (IT) department. Meanwhile, legacy LOSes would require new interfaces to take advantage of the new tool. Building and maintaining these interfaces would be time-consuming and expensive, forcing some lenders to concede they could no longer afford the new best-of-breed component.

It's my view that in order to meet the needs of today's lenders, origination platforms must include (at a minimum) Web services, integrated interfaces and bundled services, and must be built on an SOA.

ELSes are designed to manage data instead of tasks. This is a fundamental change in thinking that will signify the end of the traditional LOS. Stop and think about this concept for a moment. These new systems can define and automate processes based on and triggered by changes in loan data. They are powered by business-rules management engines that determine what must occur in the process of closing the loan. These SOA-based core systems can then act in accordance with these rules.

The most successful of these new systems will embed mortgage-specific rules engines with predefined events and actions designed to ace the competition. These engines will be an inherent part of the software and understand all data specific to the mortgage transaction. With this kind of power built in, lenders can quickly develop processes for originating any kind of loan. With a different set of business rules, the same core system can perform other work, such as originating a different type of loan, servicing the portfolio or selling whole loans into the secondary market.

What's possible right now?
In October 2005, Thornburg Mortgage announced it went live on a new lending platform, Mortgage Cadence. Mortgage Cadence is an ELS that contains the combined functionality of LOS, automated underwriting system (AUS), business rules management, workflow and imaging. Thornburg Mortgage's original intentions for seeking a new system were to enhance its overall business scalability and business processes through increased automation.

Previously, Thornburg Mortgage's internal loan review process was mostly manual, which limited its ability to increase efficiencies and productivity. Working with Mortgage Cadence's technology consulting division (3t Systems) and using its best practices paradigm called MC POM (Mortgage Cadence Process Optimization Methodology) to assess its business lending systems, 136 aggregated activities out of approximately 170 were identified as opportunities for process improvement and cost reductions. Additionally, 110 of these recognized activities could be either partially or fully automated through Mortgage Cadence BRMS, the embedded business rules management system (BRMS). This system also manages areas such as workflow, data validation, events/alerts and security.

By using the system's service-oriented architecture, Thornburg Mortgage expanded its paperless lending environment—utilizing embedded business rules, custom application interfaces and a custom integrated imaging system. In addition, correspondent lenders and mortgage brokers now have direct access to the company's lending platform via Web services. By invoking open architecture within Mortgage Cadence Software Developer Kit, these Web services provided Thornburg Mortgage the ability to build a customized Web site with a unique brand, look and feel specific to Thornburg. Under the hood, the services automatically leveraged preconfigured business rules, automation and common data in the system.

Anne Beckett, senior vice president of loan operations for Thornburg Mortgage, cites Mortgage Cadence's technology capabilities and roadmap as a key differentiator in the company's selection process. "Mortgage Cadence's automated workflow is what initially sold us on the product. The system gives us the ability to track and automate a loan through the entire loan cycle. This ability will save our correspondents countless hours, increase our overall customer satisfaction and retention, and ultimately lead to more business in the future. It is a great fit for our business strategy," Beckett says.

Thornburg Mortgage projects departmental cost savings resulting from the process improvements in both its front and back office. These cost savings stem primarily from the reduction or elimination of manual tasks and operational expenses related to new loan setup, faxing, office supplies and courier services—a direct result of the integrated imaging capabilities. There are also additional benefits relating to easy document retrieval and storage. The balance of cost benefits comes from performance improvements in Thornburg Mortgage's pricing desk, information technology, processing, underwriting, postclosing and servicing areas. Furthermore, risk was significantly reduced or eliminated relating to loan files, data transfer, declined loans, intra-departmental communication and off-system calculations.

Altogether, Thornburg Mortgage anticipates it will experience a cost benefit of more than $3 million over the next three years and a cost benefit of almost $6 million over the next five years. The company estimates that for its correspondent lending activity alone, it will experience a 23 percent cost reduction. There will also be an estimated reduction in cycle time of approximately 18 percent.

This real-world example helps exemplify the benefits that are possible with enterprise lending solutions. ELSes give lenders the ability to create or simply configure any system they require to power their enterprise. It's like having a virtual assembly line that can extend to any supplier you need to get just the right part (data) for any type of loan you want to manufacture.

This is why many of the legacy LOSes in use today are no longer competitive, and why many LOS projects fail—they cannot meet the business need. No longer can an LOS that simply provides a means of checks and balances, along with elementary automation, be considered an advanced technology solution for lenders. In fact, in today's complex lending environment, it's not a solution at all. As new software and capabilities are developed anywhere along the mortgage supply chain, these LOSes will fall further behind, as will the lenders that use them, in my view.

Advantages to the new paradigm
You can call just about any vendors in our space and ask them for their paradigm-changing technology, and they'll be happy to talk to you. Few, however, actually have developed anything that will change the way the mortgage industry does business. SOA, in and of itself, won't do that, either. But it opens the door and enables it to occur.

Today's automated systems have been set up to automate outdated manual processes regardless of the way they are structured, making it faster and easier to do the things mortgage lenders and servicers have always done. The old processes that lenders used to move a loan from application to closing were meant to assist humans in processing these files more effectively, but not necessarily more efficiently.

Automated systems don't need to care about the forms used for a particular part of the transaction. They don't need to mimic human processes in order to effectively move information from a consumer credit repository to the liabilities section of the 1003. In fact, an automated system doesn't even need to know what a 1003 is in order to gather all of the information required to effectively underwrite, approve and fund a loan.

This is where the paradigm will be changed. Software doesn't need to be limited to documents and processes. Software only cares about data. Learning to think about the information independently of everything else is what will really allow lenders to move into the new paradigm. When data become the most important thing, automated systems will be free to seek it out in the most efficient manner, speeding up the process and squeezing costs out of the transaction.

When the SOA-based core processing system is managed through an effective business rules management engine, it will quickly become apparent that by changing the rules and the services involved, the same flexible technology platform that processes mortgage loan applications will be capable of handling the mortgage loan servicing function. Like today's manufacturers, mortgage lenders will be able to swiftly retool their system to handle other jobs.

Core systems will no longer be locked into specific lines of business. The problems with data trapped in silos will go away as these new enterprisewide platforms gain access to all of the data outside the company—and not just inside the company. SOA-based systems are capable of reaching out to any source available on the network by connecting to their published Web service. Partners and suppliers will become more accessible, and sharing information securely with them will be easier.

In terms of the LOS, you can expect an ELS to offer all the functionality of today's LOS at the same time it offers anything and everything the business managers have decided to build into the firm's business rules.

If data know no boundaries, neither will the company's employees. They will have access to the system from anywhere, which will free them up to operate effectively beyond the company's physical walls. Any Internet connection will give them access to what they would expect in their own cubicle.

Reaching out to another system on the network for data is not very different from reaching out to another node in the network to share processing. In fact, SOA is ideal for load-balancing. Systems based on this architecture are highly scalable. It is very cost-effective to add additional processing capabilities to these platforms, allowing the core systems to grow with the enterprise.

A time for change
Some would argue that the current crop of loan origination systems will be secure for many years, as the companies that bought them attempt to recoup their investments. Unfortunately, there are a number of trends that will speed these older systems into their graves, in my view.

Perhaps first among these trends is the speed at which new software services are coming to the market. Compliance, documentation and paperless loan processing are all hot areas where vendors are racing against each other to bring slick new tools to lenders and servicers. Every time a lender has to pay someone to develop a custom interface between its LOS or servicing system and a new tool, it will fall farther behind its competition. Manual integration cannot compete with the speed and accuracy of SOA and Web services. As with manufacturers before them, competitive pressures will force many lenders to move into these powerful new systems sooner rather than later.

Compliance is vital to any business organization. Lenders depend on outside compliance experts to keep them on track in a shifting legislative and regulatory landscape. A mistake here can be fatal to a mortgage banking enterprise. Lenders will demand the best compliance tools available. As vendors vie to meet this need, lenders with outdated systems will not be able to keep up.

The success of MISMO is testament to the fact that the industry is moving swiftly toward the paperless mortgage. To accomplish this goal will require a number of electronic components, including eClosing, eSignature, eRecording and eVaulting. As these new tools come online, ELSes will allow lenders to easily integrate them into their processes. Those lenders that depend upon outdated technology will not be so lucky.

Finally, after years of searching, vendors are answering lenders' demands for end-to-end mortgage origination systems. Once lenders realize that these systems are really just a collection of legacy automation tools cobbled together into a proprietary process that locks them into an outdated method of doing business, their disenchantment will fuel their move to newer, better technology.

ELSes will stand in stark contrast to the limitations of the LOS. The effect will be to draw a line in the sand. Traditional LOSes will not be able to cross that line and fulfill the future needs of mortgage lenders.

It is hoped you will have made the right choice with your technology provider when this comes to fruition. Otherwise, you might be in the market for another loan origination system much sooner than you think.

Michael Detwiler is chief executive officer of Mor tgage Cadence, Greenwood
Village, Colorado (www.mortgagecadence.com), a firm that provides ELSes and
other technology platforms for a range of industries. He can be reached at
mdetwiler@mortgagecadence.com.