Insurance company triples home insurance cross-sell by automating predictive

A leading insurance and financial services provider wanted to identify and execute cross-sell opportunities for various insurance types. Pecan worked with the marketing and sales teams focused on auto and home insurance.



higher campaign


improved model performance

2 weeks

to effective vs. 6 months from in-house model

The challenge

Negative customer experience with unfocused upsell campaigns

The sales team  was conducting generic upsell campaigns aimed to offer home insurance bundles to current auto policyholders. The traditional outreach approach resulted in low upsell conversion rates, unproductive sales reps, and negatively impacted customer experience.


Tailored, targeted home insurance upsell to maximize conversion

Pecan’s platform ingested all raw historical sales and customer data and built a predictive model to help identify the highest likely customers to purchase home insurance. 

The primary end result was 3x the campaign conversions. But like many instances of upgrading standard BI to predictive analytics, benefits spanned the entire customer journey. Sellers saved time by not having to target accounts unlikely to convert, while customer experience benefitted from campaigns only going to those who were likely to convert.


Tripled conversion rate from 5 to 15%

In only a few hours, Pecan delivered a fully functional model that was able to meet and even outperform a competing model developed by the internal data science team over 6 months. Pecan’s model improved the conversion rate by 3x (5% to 15%). Moreover, Pecan was able to enhance performance even further (up to 10x) by combining their model with the internally developed one maximizing impact and demonstrating the powerful combination of traditional data science along with the Pecan platform.

Want to get started with predictive analytics?