Maximize your efficiency with our Likelihood to Sell Score, an AI-powered predictive metric that helps you identify clients likely to list and sell their homes.
We've fine-tuned our algorithm to provide more accurate predictions by incorporating a broader range of data points, including market trends and individual client behaviors.
Continue reading to learn how we built this feature, how it works, and how you can incorporate it into your outreach initiatives.
The Homebot Likelihood to Sell Score is an innovative machine learning model designed to accurately predict your homeowners’ likelihood to list and sell their home (or not) within the next 9 months.
Every client in your account will receive a score, except for buyers and clients with an incomplete profile (missing home address). The higher the score, the more confident we are that they are likely to sell their home within nine months. Likely to Sell scores are under the CLIENTS tab, as shown in the image below:
The scores are updated daily, so even if a client isn't looking at their report, their score will change based on the data we pull. For example, our interest rates are obtained every Thursday from Freddie Mac, which will affect the score.
If you add a client and don't see a score right away, please wait until the following day before concluding that there's an issue with your account.
Please note: Your client needs to be in their home for at least one year to get a score.
How we built the Likelihood to Sell model:
Our data at Homebot is extensive, with over 150,000,000 rows of information available for analysis. The score is generated by a machine learning model trained on all of our activity data from 2020 to 2024. We utilize CMA requests, direct messages, interest rates, and general user interactions within the reports to determine the most important signals from Homebot that we can use to predict a sale.
However, we do not need a client to engage with Homebot to determine if they are likely-to-sell within the next 9 months. In addition to the information above, we use demographic information, their location, time spent in the house, their current interest rate, equity, and home details. Engagement helps us get even more accurate, but we don't need it to predict sellers.
How to leverage the Likelihood to Sell score:
What actions would you take if you knew one of your clients was likely to sell their house within the next nine months? Could you create more targeted campaigns or reach out early to provide personalized home-selling support?
Through Homebot, you could:
Send a personalized video message to a client with a high score using BombBomb or any other screen recording application (we really like Zight at Homebot!).
Send all of your homeowners a Video Email outlining the 5 tips for getting their home ready to sell.