We live in a world filled with messaging. 2.1 billion people use some messaging app. In fact, most of us use only about 7 apps on our phones and a number of those are almost always a messaging tool – SMS text, Facebook Messenger, Telegram, WhatsApp. Messenger alone sees about 1 billion messages per day and 700 million monthly active users. 1 in every 5 minutes spent on a mobile phone in the U.S. is spent on either Facebook or Instagram.

Climb realizes our agents live in their messengers because that’s where our clients are. There’s a lot of good real estate in these tools. People spend 50 minutes per day between Facebook, Messenger or Instagram.

  • 49% of smartphone owners ages 18-24 use a messaging app (source)
  • 37% of smartphone owners ages 30-49 use a messaging app (source)
  • 24% of smartphone owners ages 50+ use a messaging app (source)

In the world of real estate, personalized and timely responses are a major key to a professional’s success. By leveraging the massive messaging infrastructure we live in today, Climb Labs is working in partnership with Structurely to develop real estate-specificchatbots to give our agents and clients access to an emerging communication channel.

holmes demos

A chatbot (“bot” for short) is an application that lives in a messaging tool, like Facebook messenger, Slack, or What’sApp. What’s attractive about that, is there isn’t a setup or learning curve to a separate app, which means no additional download is required. Just pull up your favorite messaging tool and start chatting with your favorite bot – such as the CNN Bot, Wall Street Journal Bot, or the Poncho Bot for weather.

“Chat apps will come to be thought of as the new browsers; bots will be the new websites. This is the beginning of a new Internet.” – Ted Livingston, Founder of Kik

Bots are built with artificial intelligence (AI). AI requires huge amounts of data to be effective and this data must be highly tuned towards the use case in which it seeks to understand. In real estate, data can look like almost anything as it is applied differently to each unique home buyer or seller.

Bot Understanding

The bot understands the price and property features of an open house search

In order for us to create a bot that brings value to our clients, we must define a limited scope for our AI bot to understand. To do this, we are reading market data and limiting the Natural Language Understanding (NLU) of the bot to comprehend market questions such as:

  • “How many homes are on the market with 3 beds?”
  • “Can you show me some homes with 3 beds/2 baths for sale in Russian Hill?”
  • “What’s their average price?”
  • “How long are they generally on the market for?”
The bot requests data from a predictive days on market model

The bot requests data from a predictive days on market model

By limiting the AI bot’s scope to understanding market questions we can begin to generate Neural Networks that understand buyer and seller’s behaviors in the market. This lays the foundation for building up the bot’s intelligence for more advanced questioning. By training the bot with real conversations powered by real data, artificial intelligence becomes more powerful when interpreted by humans. In the personable world of real estate, humans will continue to dominate but machines have their place too.



Share this:
Facebooktwittergoogle_pluspinterestlinkedinmailFacebooktwittergoogle_pluspinterestlinkedinmail

Related Projects