Problem Statement
Our client, a leading US-based PE firm specializing in the real estate sector, wanted us to pull data from multiple sources and integrate it to build dashboards and derive meaningful business-driven insights for their portfolio investments spread across ten states in the United States of America. Traditionally, the PE firm relied upon labor-intensive research methods by maintaining as many Excel files as the number of sources and portfolio sites. The JMI Analytics Team was roped in to simplify the entire workflow process, delivering value and offering solutions in three broad areas:
- Consolidation of financial and operational data from multiple sources for investment sites spanning multiple states.
- Understanding the financial health of portfolio sites by building interactive dashboards.
- To grasp the key real-estate operational metrics at a micro-level to disseminate derived insights to the intended regional property managers.
- Build a detailed tenant profile for each tenant across property sites.
JMI Implementation
- The JMI Team reworked the entire data inflow process for data processing by building a data lake for quick and easy access to the requisite data.
- Perform extensive data pre-processing steps to convert unstructured and semi-structured data coming in from various sources into a structured, human-readable format for data processing.
- Derive intelligent KPIs based on current and historical financial data and build interactive visual dashboards.
- Built dashboards based on real estate operational efficiency to disseminate derived insights to the intended regional property managers.
- Integration of multiple utility service providers bills such as water, electricity, broadband, garbage, and sewer present across various formats such as PDF and handwritten notes on a unified scale.
- Individual property-wise, profile-rich dashboards were built based on clustering techniques. These dashboards also helped to understand tenant behavioral patterns at a granular level. Predictive analytics based on machine learning were employed to classify tenants based on their propensity to make late payments and default on loans.
Results
- The process efficiency of the consolidation of data from multiple sources onto a unified data lake led to considerable resource savings and a reduction in turnaround times.
- Intelligent KPIs and insights derived from financial metrics helped the client optimize the financial health of portfolio investments and also showcase the findings to potential investors. This led to a 20% increase in interest from prospective investors.
- Dashboards developed to gauge real-estate operational efficiency helped the client achieve more granular control over the functioning of property managers, enabling data-driven decision-making.
- Rich-profiling of tenants spread across properties by means of clustering techniques helped the client optimize rent collection strategies, thereby minimizing delays in payments.