Case Studies

Using Predictive Modeling to Avoid the “One Size Fits All” Approach to Name Selection

To compensate for the lack of time in today's busy society, many people employ outside services to help them accomplish the everyday tasks they do not have time to do themselves. One such service provider, a major residential cleaning service franchise, had been engaging in a direct mail campaign to capitalize on this trend using too-broad targeting selects and simple geography to create its mailing lists. Despite the efforts of a few branches that even tried to narrow their lists further to target a smaller population, the direct marketing campaigns were producing few leads, and the branches were not acquiring new customers. The cleaning service approached our team to help its franchisees pinpoint which homes were most likely to hire an in-home cleaning service.

"One Size Fits All" Does Not Work

Predictive modeling is used to create targeted prospect universes that are customized for every campaign. By applying over 750 variables to data on more than 120 million households, a model is generated to score and produce highly-targeted mailing lists that can improve response rates by 7.5-15%.

The cleaning service had been using targeting selects like “homes with incomes of $100K+” to try to determine who was most likely to use a cleaning service and then used geographic selects to create the final mailing lists. To more accurately predict who would be responsive to the franchise’s mail campaign, models were built that targeted homes with incomes greater than $150K, but also took into consideration the value of the home, whether children or pets were present in the home, and length of residence. Our modeling technology found that these elements, which were never before considered using basic selects, were crucial for predicting response rates. The campaign’s results were astonishing.

The Results

In a head-to-head test of modeled names versus conventional targeting, predictive modeling proved to be clearly superior. By examining hundreds of data elements, the modeling system pinpointed the ten most predictive variables, which were then used to generate a model that yielded 49% more responses throughout the direct marketing campaign. In addition to the increased response, the technology generated higher revenue per response, and the cleaning service achieved a 105% increase in ROI.

Having witnessed these outstanding results, the number of franchisees using sophisticated targeting jumped from 1 to 90 branches in just 9 months.

Go Back