For most families, buying a home is the single most significant investment made during their life. The residential market is very competitive, with several high-quality players with pan-Indian operations. Many of these prominent players have in-house design teams that conduct extensive market research to understand customer requirements. The design teams incorporate customer needs into design specs when building new projects. However, catering to every customer's need is an inefficient process.
Therefore, Real Estate companies need to understand customers' requirements, tastes & preferences and incorporate the most desirous specifications into their product (home for the customer). Typically, Real Estate companies collect customer information in customer feedback from various customer bases. The companies can garner the customer feedback information generated by large data pools, incorporate most customer requirements, and make well-informed data-based that satisfies the customer.
In this study, researchers explored a leading Real Estate developer in India to build a database that incorporated their customers' needs and preferences and translated this into design specifications for their homes. The entire focus was to integrate customer-centric ideas into their product development. The study intended to understand the existing data, create a robust data collection process, and build an automated analysis method.
Data used for the study spanned 25 Real Estate projects of the developer across eight cities in India. The researchers used Product data and Customer Care data for data analysis. Product data was collected before and during the project. It included details like customer demographics, socio-economic data, design preferences, etc. However, the Customer Care data was collected post possession of projects and included details like purchase experience, service levels, overall customer satisfaction, willingness to recommend to others, etc. The researchers faced a big challenge because the data was in separate databases in an unstructured format.
Researchers used relevant keyword count (related to design) as a primary tool to measure design preferences. Then they split the data into Design and Non-Design data based on keywords. The study considered only the data related to further Exploratory Data Analysis plan.
In the study, the researchers created the 'NPS Category' (Net Promoter Score) with different ranges to get insights into customer opinion mining. They used four data points to gather customer reviews. Researchers separated the data as Promoters, Passives, and Detractors depending on whether customers promote and recommend the projects to others, are neutral for various parameters, or give negative feedback.
The researchers used N-gram to explore the client's data and maintained the Design keywords in a static file (Dictionary.xlsx).
The study was able to identify the top desired attributes of real estate projects for customers and the top challenges faced by them. The researchers shared the customer's preferences and challenges with the firm’s design team for improvement in future projects for a continuous process.
Viswanathan M., Sarvabhotla A.S.Y, Attuluri B., Ganesan N., Agoorkaisetty P.K., 'Data-Based Decision Making in Real Estate,' 2021