Look Who’s Talking – Mason Levy, CEO of swivl

Posted by Erica Shatzer on Jun 1, 2023 12:00:00 AM

On Nov. 30, 2022, OpenAI launched a prototype of ChatGPT, a large language model (LLM) that generates human-like responses to questions and statements known as “utterances.” Within two months, ChatGPT had reached 100 million users. Then, on March 21, 2023, Google opened up a user waitlist for its own AI chatbot that’s aptly dubbed Bard. 

Although the ethicality of these innovative conversational artificial intelligence (AI) programs has been called into question, and many educational institutions are hastily establishing codes of conduct for student users, conversational AI—similar to other technologies—is only expected to progress. In fact, a newer, “multimodal” version of ChatGPT, called GPT-4, was released on March 14, 2023, enabling users to enter different kinds of input. Enhanced versions will undoubtedly continue to emerge from end-user feedback and increased usage, further minimizing the instances of false information known as “hallucinations.” 

While generative AI will likely be regulated in the future, perhaps limiting usage applications, today it can be used by self-storage operators to improve their customer service, streamline their operations, and strengthen their brands. Here, Mason Levy, conversational AI expert, with over 10 years in the field, and CEO of swivl, his third conversational AI-focused company, discusses those possibilities and more. 

Entering Self-Storage 

Levy, who’s lived in 10 states over the last nine years, has been using self-storage for numerous years to store his belongings between, during, and after moves. Sometimes renting units at multiple locations across the country at once, including a period when he had storage in Chicago, Berkeley, and Denver, Levy always utilized technology to secure and maintain his unit rentals. Other swivl team members have had similar experiences, with another co-founder living an RV life for about 18 months. 

“I don’t know exactly how many storage units I’ve rented in the past, but I’ve never once walked into a management office to rent a unit,” says Levy. “I don’t think that I’ve even ever talked to somebody on the phone to rent a unit.” 

It was during his “contactless” interactions with self-storage that he identified a need for AI and automation within the industry and sought out to make the rental experience more seamless. “The product itself, we are huge consumers of, in general, and so we saw that opportunity to come in and build something that we, as consumers, would like to engage with,” he says about the formation of swivl. “We saw issues with building and deploying conversational AI in general and set out to solve those foundational issues.” 

Looking to assist enterprises with their inefficiencies, swivl formed a relationship with Extra Space Storage—a liaison that enabled the company to expand its reach and product offerings. Currently, 3,500 self-storage locations are using swivl’s AI-enabled products in English, and additional languages may be available in the future. Swivl also hosts a tech-focused podcast called Self Storage Lab. 

Utilizing AI 

These days, most people, regardless of their age, are familiar with technology and use it in their daily lives. Younger generations, however, are more inclined to seek out tech-based alternatives to bypass human interactions than older generations, perhaps because it has been a more significant part of their lives. For instance, Gen Z, those born between 1997 and 2012, is the first generation in which all of its members were born after the internet went public in 1993.

But having so much easily accessible information at our disposal has essentially programmed society to expect immediate information, rapid responses, instant access, and on-the-go offerings. What’s more, as Levy points out, the nomadic and minimalist lifestyles that many “zoomers” (those aged 18 to 26) and millennials prefer to live create demand for self-storage when they need somewhere to store their excess belongings. 

These realities are especially important when current and future interactions with self-storage facilities are taken into consideration. Since 60.5 percent of customers rent from the first facility they contact, according to the SSA’s 2020 Self Storage Demand Study, self-storage operators need to meet customers where they are—on their smartphones.

On this topic, the SSA’s 2020 Self Storage Demand Study revealed that 37.8 percent of renters first learned about their self-storage facility online via search engines, store websites, aggregators, social media channels, or reviews. When making first contact with a facility, 8.7 percent of renters used email and 8.5 percent used the facility’s website. As for making rental payments, 8.8 percent of renters preferred to pay with various smartphone apps or online options. 

The industry’s adoption of newer technologies will only augment these numbers. And that’s where conversational AI-enabled products come into play. Levy states that conversational AI-enabled products can be used as a solution for assisting customers. Similar to a call center, chatbots can prevent self-storage facilities from missing leads when managers are busy during office hours. They also provide a round-the-clock alternative for customers, like Levy, who’d rather type an inquiry into a chat box than make a phone call. He gives an example of someone who needs emergency storage at 2 a.m. “Conversational AI can be there to help.” 

“So, regardless of where that customer is finding you, I think operators really should think of this as an omnichannel approach,” he says. “It’s really allowing for customers to engage with your business, or brand, wherever they want.” 

Levy goes on to say that conversational AI-enabled products like swivl can make product recommendations (such as what unit size to rent), provide access codes, answer questions, provide site-specific information, and more. Unlike a standard FAQ page on a facility’s website, which may not have the answers a customer is seeking, conversational AI-enabled products can give more personalized responses to customers’ questions. 

“Maybe I don’t identify with any of those questions,” he says. “Even if my answer is kind of somewhere in the mix of that, I want to ask my question … it really allows for customers to feel heard, cared for, and then ultimately been sold or serviced.” 

Conversational AI-enabled products are able to provide those kinds of personalized responses from “training.” Just like onboarding a new employee, Levy and his team at swivl have “trained” the company’s natural language model with self-storage-specific data. “… This is where that human element comes in,” he says. “We’re constantly reviewing and annotating and training models and testing models to see what that accuracy is.”

In addition to a chatbot application, conversational AI can be utilized to reply to social media posts and reviews in the same fashion. With more customers looking at reviews and how companies respond to them, replying with messages that are consistent with the company’s brand is becoming more important than ever. The same is true for social media posts; customers who take the time to leave feedback or make comments are appreciative when the company retorts.

As for more general purposes, conversational AI can assist operators with generating text for emails, websites, letters, advertisements, telephone scripts, and much more.

“There are other benefits to implementing conversational AI,” Levy adds. “From an operator’s perspective, you’re reducing labor costs by adding conversational AI into the mix.” 

At the facility level, swivl has about an 80 percent automation rate. “Our average unit reservation rate from chat is around 25 to 30 percent,” says Levy. “When we show inventory, 25 to 30 percent of the time we have somebody say that they want to reserve that unit. We’ve seen a 17 percent decrease in service-related calls, and the opposite to that is a 90 percent decrease in agent escalations, which is exciting. So when somebody’s engaging with our AI systems, we’ve seen a really large drop in the necessity of getting an agent involved.” 

Getting On Board 

As conversational AI continues to improve and become more human-like, generating more emotional and deeper responses, its utilization will increase and broaden. But regardless of when and how you plan to use it in your operations, Levy stresses the importance of “having buy-in from the entire team” before implementation. 

“It is operations, sales, contact center; it is the entire marketing, everyone working underneath, that you have to have leadership buy-in to say this is how it’s going to work,” he says. “And so having that buy-in from the very top is key.” 

Once you have buy-in for the product and its paralleling objectives, clearly communicate those goals to everyone in the organization. Levy says it’s the same process you should be taking for the implementation of any new technology to make it “as successful as possible.” 

While Levy acknowledges that the self-storage industry had been slow to adopt new technologies in the past, COVID quickly brought to light the need for contactless options and proved that many customers prefer human-less channels of communication. If that disruption didn’t spur you to take action, there’s no better time than the present to get on board with AI and other automations that can improve customer service, reduce expenses, increase sales, and strengthen your brand.

Erica Shatzer is the editor of Messenger, Self-Storage Now!, Self-Storage Canada, and the annual Self-Storage Almanac.