1. Introduction
HomeRiver Group (HRG) is one of the nation’s leading residential property management companies. They’re known for their exceptional level of service – both to tenants and homeowners, and strive to find ways that technology can help them continue to execute on their mission. Like many property management companies, though, they’re always looking for ways to better predict the behavior of their clients, particularly amongst their homeowners.
2. Problem
HRG wanted us to explore whether it was possible to predict which homeowners were at the greatest risk of unenrolling in HRG’s property management services. When a homeowner churns, it has a significant financial impact on HRG’s business, so predicting churn has significant implications on their business.
3. Action Steps
HRG hired Auril AI to address this issue. Our project had 3 main goals:
- Assess the state of data sources HRG could mine for features predictive of churn
- Identify other workflows that could be the source of future features to enhance churn prediction systems
- Study the predictive power of currently available data features in predicting owner churn for use in data-science-driven predictions.
By starting with conversations with key business leaders, we came to understand the workflow that fueled the various processes to service homeowners. This allowed us to make an informed deep-dive into currently available datapoints and make an exploratory analysis of the potential for predicting owner churn, allowing us to inform HRG’s already considerable efforts to mitigate homeowner churn.
4. Results
Our study surfaced some fascinating findings that informed HRG’s churn prevention strategies, and gave context to their already-planned work to deploy new software to track employee relationships.
“The Auril AI team was instrumental in helping us accelerate our data science roadmap, and we’re excited to partner with them on our upcoming AI initiatives” – Jamie Wilson, Chief Technology Officer, HomeRiver Group
By studying currently available data, focusing on elements such as placed work orders on homeowner properties, email frequency & content between HRGs team and homeowners, and homeowner demography, we found a set of interesting relationships that suggest new strategies for engaging with homeowners.
One specific finding had to do with the relationship between placed work orders and churn. Going in, our default hypothesis was that homeowners who are spending a lot of money making repairs and improvements to their properties were more likely to churn, perhaps due to their being unhappy with that spend. In fact, the opposite ended up being true, and the presence of recent work orders was a signal of engagement, and the owners without recent work orders were 12 times more likely to churn.
As a result of this exploration, we were able to design a homeowner churn risk index for HRG to put into production. This could be used for a variety of purposes, including stratifying churn risk amongst specific homeowner sub-populations and designing new processes to improve how HRG serves their valuable customers.
5. Call To Action
Are you facing a similar problem in your organization? Do you wish to make progress and hit achieve your most pressing goals like HomeRiver Group did?
Auril AI offers a free consultation to help you gain clarity around your main challenges and how to solve them.
Through our Data Science and AI Strategy expertise, we can help turn your biggest challenges into results, just like how we helped HRG.