Humanizing AI to improve the home buying journey

By BrokerOwner.net November 13, 2019 Housing News

There’s no question AI and machine learning are having a transformational impact on the home buying and refinancing journey. And while the long-term reverberations of AI’s expanding role are fascinating and for some worrying, companies are realizing that marrying people with machine intelligence is the path to capitalize on AI’s potential.

Artificial intelligence has generated lots of unrealistic expectations. There’s no shortage of vendors, or internal project plans liberally sprinkled with references to computer vision, supervised learning and other forms of the technology, with little connection to its real capabilities. Merely calling a CRM “AI-powered,” for example, doesn’t make it any more effective, but it might help with fundraising.

When assessing AI,
it’s useful to view it through the lens of business capabilities rather than
technologies. Broadly speaking, AI can support three critical business needs:

Automating processes
Gaining insight through data
analysis
Engaging with customers and
employees

Therefore, it is
important to isolate the problem you are trying to solve before starting your
AI initiatives.

Use of AI

The most common use of AI today is business process automation, often called Cognitive Automation. It is more advanced than previous automation efforts because it can mimic human input and information consumption across multiple systems concurrently. Robots are now “reading” legal and contractual documents to extract provisions using natural language processing.

According to a 2017 article in Bloomberg News, JPMorgan Chase introduced a system for reviewing commercial loan contracts; work that used to take loan officers 360,000 hours can now be done in a few seconds.¹ If it is a process you can outsource, it’s probably one you can leverage AI to massively improve speed and accuracy.

The
next use is gaining insight through data analysis, called Cognitive Insight. This
entails the use of algorithms to look for patterns and gain insights on massive
amounts of data. It’s the super-sized analytics of the past. AI can help you
predict what a customer is likely to buy or identify credit fraud in real-time.
We can now analyze hundreds of fields across millions of records to accurately
predict that a consumer is about to sell their house 90 days before traditional
triggers.

Real estate agents can better target their marketing efforts and lenders can actively engage customers in their portfolio retention efforts. The ability of AI to analyze photos and recognize objects allows for an enhanced consumer search experience. The most significant advancement over previous analytics is the models can get better — that is, today’s AI can leverage new data to improve its ability to make predictions or put things into categories.

Finally, Cognitive Engagement projects “talk” to employees and customers using natural language processing chatbots and intelligent agents. Think of systems that offer 24/7 customer service chatbots addressing a broad and growing array of issues all in the customer’s natural language. Companies have typically used these systems internally to engage employees because they are hesitant about turning customer interactions over to machines. However, these systems are getting better every day. Chatbots are used to engage “lost” or “dead” leads with resuscitation rates high enough to justify the project.

Home Captain, a Charleston-based, Conversion Optimization System (COS), leverages chatbots to help their lending partners recover leads after dozens of failed attempts to contact the consumer. They average 6.4% contact rates on aged leads 30-90 days old, ultimately warm transferring the consumer back to the lender. Realtime leads have a 29.9% contact rate and a 10.1% warm transfer rate. These chatbots don’t sleep and can optimize contact patterns far more effectively than their call center counterparts.

AI should
supplement service offerings

With all the advancements in data science, machine learning and artificial intelligence, there remain limitations. Machine-learning systems must be taught how to perform the work they’re designed to do. As these systems increasingly reach conclusions through processes that are opaque (the black-box problem), they require human specialists in the field to explain their behavior to non-expert users. This is especially important in highly regulated industries, such as lending. Regulators want to know how that credit decision was made, etc.

More
important, purchasing a home is often the biggest transaction in a consumer’s
life. They, especially first-time homebuyers, want someone to shepherd them
through the process. They want someone to explain to them what is happening in
the context of their specific situation. Sometimes they just want to be
reassured.

As we
consider cognitive technology projects, companies need to reimagine processes
and workflows to ensure that humans and machines augment each other’s strengths
and compensate for weaknesses. A good example is a company that uses chatbot
technologies to ascertain interest and incubate leads, but its final task is to
set an appointment where a human takes over. Intelligent scaling meets human
capital.

There
a growing term in data science replacing the “artificial” in AI; it’s
“Collaborative Intelligence.” Pairing the leadership, social skills, and
creativity of humans with the speed, scalability, and analytic capabilities of
machine learning. A great business requires both kinds of capabilities.