A real estate investment practice run by a group of realtors, managing transactions of about US$100 million annually.
Their primary service being buying and selling of residential properties from house owners looking for hassle-free quick deals in cash, they were struggling to reduce time spent on lead and campaign management processes, which comprised over 30% of their work hours.
They also wanted to explore if data-driven lead segmentation and/or personalized campaign management could help them improve lead conversions.
Identifying parameters contributing towards lead conversion success or failure – because of very limited digital historical records on lead sources and acquisition factors.
Identifying data sources: As most customers formed part of the public who did not prefer to list for buying or selling homes.
Data Quality issues: A lot of mismatches in lead data from various public and private databases, which were frequently updated.
Content Customization for campaigns: Creating personalized marketing campaign encompassing a large number of target customers across segments was a huge challenge.
Automated Lead Segmentation Module – developed by calculating “lead similarity index” using advance machine learning methodologies on various demographic factors, past interactions, information from third party data sources, etc.
Lead Conversion Dashboard – showcasing conversion metrics across time, comparing between test and control groups, with ability to drill down across revenue metrics, lead segments, etc.
Personalized Marketing Solution – this included
Performed market research to identify relevant data sources and analyzed all the information available to select relevant parameters which could be useful in conversion probabilities calculation for individual leads.
Automated data collection process, with customized modules for each data source depending on source type and structure and an optimum data extraction schedule to deal with updates.
Developed an automated data cleansing and transformation module for getting the data ready for further analysis.
Designed an advanced analytics algorithm to evaluate success probability for each lead and score it.
Developed a mechanism for evaluating campaign effectiveness by identifying a test population having more than a pre-specified threshold value of success probability and a control population chosen at random without calculating success probability.