Calculating Customer Lifetime Value on Azure For Real Estate Property Management

Alex Scroggins

Khushboo Mehta

Data economics is an emerging discipline concerned with quantifying the economic benefits of data. 

An example of how a company can put data economics into practice is by using its data to calculate lifetime value (CLV) of its customers, then using CLV (and other attributes, such as credit score) to segment its customers into groups that can be used for next-best action marketing.

YDC recently applied these concepts in the real estate industry for a property management company leveraging Microsoft Azure.

The customers of property management companies are property owners. Property management companies work on behalf of property owners to attract and retain tenants, collect rent, maintain and repair rental units, evict bad tenants, and other activities. Each of these activities has corresponding data tracked in a property management system. 

Using the property management system as inputs, we calculated a lifetime value of the various apartment communities managed by the property management company.

The data flow of the solution is presented below.

Figure 1: Architecture

A description of the data pipeline is discussed in the following sections:

Data Lake

The data lake contains files from the property management system including rental income per apartment unit, tenant income, tenant delinquency, and so on. These files were important inputs for the lifetime value model built using Azure Data Factory.

Azure Data Lake Storage Gen2 (ADLS) was used for the data lake. An example file from the data lake is shown below. 

Figure 2: ADLS Data Lake Example File: Rental Income Per Unit


A rules-based model, using files in the data lake as inputs, computed the lifetime value of each apartment community and property owner.

Azure Data Factory was used to execute the rules-based model as a data pipeline. A screenshot from the data pipeline is below. We plan to extend the model to leverage AI/ML using Azure Machine Learning to predict tenant churn for each property given historical tenant information.

Figure 3: Azure Data Factory Data Pipeline

Data Warehouse

Results from the model were written to the data warehouse in a dimensional model. Dimensions of the model included property owner, apartment community, and unit (unit contains such details as number of bedrooms, number of bathrooms, etc.). These dimensions represent a natural hierarchy: property owners own one-to-many apartment communities; apartment communities contain many units. 

Azure Synapse Analytics was used as the data warehouse. A screenshot of owner lifetime value is shown below.

Figure 4: Azure Synapse Analytics


Power BI was used to create executive dashboards over the data. Data was summarized at the community and property owner level.

An example dashboard is shown below. This dashboard summarizes CLV for the property management company (ABC Real Estate Partners) and compares it to a key competitor (XYZ Investment Group Holdings).

Figure 5: Example Power BI Dashboard


As a reminder, the customers for the property management company are property owners. Property owner lifetime value was sent into the CRM to help sales and marketing identify the most valuable property owners. The business was interested in using lifetime value to assign segments to its customers. For example, a “High Value” segment could be assigned to property owners having a lifetime value equal to or above a threshold.

Microsoft Dynamics 365 was the CRM used by the property management company. An example screenshot of how Lifetime Value was integrated for customers is shown below.

Figure 6: Lifetime Value Integrated to Microsoft Dynamics 365

Other news

We use cookies to ensure we give you the best experience on our website. If you continue to use this site, we will assume you consent to our privacy policy.

This website uses cookies. By continuing to use this site, you accept our use of cookies as described in our Privacy Policy