It’s not often that we can draw a parallel between the automobile industry and the mortgage and real estate industries. And yet, our space may be on the verge of its own “Model T” moment.
The Model T, you’ll recall, represented the first successful dive into mass production and assembly. Ford’s game-changing assembly-line approach turned the automotive industry (really, every industry) on its head. In just under 20 years, Ford managed to produce over 15 million Model Ts, cutting the production time from 12.5 hours to 93 minutes in under five years. In so doing, Ford obliterated a long-established industry business model plagued by inefficiency, slow production and lack of standardization.
Sound familiar? The mortgage industry has been awaiting its own “Model T” introduction for quite some time. It may finally be here.
In spite of digitalization, the mortgage and title industry is still mired in inefficiencies
Mountains have been written about the inefficiencies inherent to the mortgage and real estate industry, lenders and service providers alike. Ours is a space crying out for more standardization. Every property is different. Every transaction is different. Almost every state, county and municipality has its own, unique set of real estate customs, laws and regulations. Similarly, seemingly every element of a real estate transaction demands numerous siloed service providers working simultaneously with technology and data kept in silos. The result is, ultimately, a nearly unwavering 50 days-to-close average across our industry.
All of this in spite of what has to be considered a digital transformation that has enveloped us in the past five years.
Yes, the world of mortgage and real estate has come a long way in just a few short years. Driven to automate and innovate – willingly or not – by not only a global pandemic but by consumer and client demand as well, as one of the most abrupt and substantial pivots in market conditions we’ve ever seen, mortgage lenders, commercial investors and service providers alike have had little choice but to embrace things like digital closings or automated workflows.
Despite of that rapid growth, we still see way too much key mashing, stare-and-compare and voice mail tag. More often than not, homebuyers still wait weeks and even months to get the keys to their new home.
The good news is that, as happens with most substantial industry-wide transformations, the digitalization of our industry is only about to accelerate.
Naturally, one can’t have a conversation about automation and innovation without invoking AI. While ChatGPT has almost become a caricature of itself or a buzzword, AI solutions are already making a huge difference in the world of real estate, and not just with marketing materials or basic communications.
HOA requirements and documentation: a case in point.
As a case in point, let’s look at the settlement requirement to navigate homeowners associations (HOA) and their vast and disparate array of documents and requirements. It’s not the first thing one thinks about when signing a sales contract on a new home, but when it’s poorly handled, those disparate or undiscovered requirements can derail an otherwise smooth closing process. Worse, in the case of a single family rental investor seeking to purchase a home to rent to others but failing to realize strict leasing prohibitions imposed by an HOA, it can lead to minor disasters.
Identifying an HOA and corresponding property management company then locating and properly processing HOA-related documentation such as bylaws, document fees and the like has long been a primarily manual task for title agents. The biggest issue with identifying HOAs and procuring HOA-related documentation is that there is no dominant, national standard. The result is thousands of HOAs with thousands of varying requirements and mounds of varying documentation.
For the title agent, it meant finding those documents, scraping or collecting them from a website or requesting them manually, then painstakingly scouring them (sometimes dozens to hundreds of pages long) before processing what that particular HOA required to close the deal as well as what would be required of the buyer. At various points of this tedious process, the title staffers likely had to key and rekey some of the relevant data into their own system.
It’s probably not too surprising to know that this process was far from error free in addition to being time consuming. If a required form or fee wasn’t mailed, or wasn’t mailed to the appropriate P.O. Box, days could be added to the process. If a specific form had to be downloaded and completed, or completed in a very unique and specific way, days could be added to the process. Things like leasing/rental restrictions or fencing requirements, especially in SFR-related purchases, had the potential to become deal killers.
AI is already impacting the HOA process
The good news in all of this is that the title industry isn’t “on the verge” of finding an efficient solution. We’re not awaiting “impending” improvements. Although we are in the early days of the transition, service providers are increasingly deploying AI solutions that take days (if not weeks) off the cumbersome process of hurdling HOA obstacles.
Large language models can streamline quality control processes
AI technologies, such as Large Language Models (LLM), are already being used to streamline quality control processes, minimizing errors and inconsistencies. Machine learning algorithms are analyzing vast amounts of data, detecting anomalies and discrepancies that human oversight might miss. From verifying property details to assessing compliance with regulatory standards, AI-powered QC systems offer a level of efficiency and accuracy that traditional methods cannot match.
For example, it only takes AI technology minutes to scan and analyze hundreds of pages of legal documents; find all data points relevant to the title officer and summarize them into a user-friendly report with links to the original text. The same AI solutions are then almost instantly reviewing thousands of data points to determine which forms would be required; where fees would need to be sent and what limitations the home buyer might have in this particular HOA.
In fact, AI is now regularly used to perform its own quality control (QC), repetitively performing certain tasks and then flagging for human review any discrepancies or inconsistencies in the results. These solutions are sorting, prioritizing and deleting data points in an overall process that could cover hundreds of pending transactions and a dizzying amount of information.
The HOA process, once one of the most time-consuming and error prone elements of the entire settlement service phase of a transaction, is not the only one being dramatically transformed by AI. Most know that AI will soon be a standard tool for communications and basic marketing tasks. But it’s also being rolled out throughout the most stubbornly archaic elements of the workflow in all parts of the real estate transaction. In fact, AI will likely prove to be one of the primary catalysts for the mortgage and title industry’s Model T moment, making many of the old ways obsolete.
Vishrut Malhotra is a co-founders of Rexera (formerly InspectHOA.)
Anton Tonev is a co-founder of Rexera.
This column does not necessarily reflect the opinion of HousingWire’s editorial department and its owners.
To contact the author of this story:
Anton Tonev at [email protected]
To contact the editor responsible for this story:
Tracey Velt at [email protected]
