How AI is reshaping roofing outreach

By Jesse Sanchez. 

Find out how roof-condition imagery and data collection can refine outreach efforts.  

Roofing contractors have no shortage of marketing tools. They can invest in digital ads, purchase mailing lists and saturate neighborhoods with postcards. With more channels available than ever, contractors now have greater opportunity to connect with the right homeowners at the right time and deliver more targeted, effective outreach. On this episode of Roofing Road Trips®, host Karen Edwards spoke with Reworked.ai co-founders Shyam Prabhakar and Fred Castonguay about a strategy designed to address that inefficiency at its source. Instead of trying to capture homeowners who are already searching online, the company’s model focuses on identifying properties where the roof is likely nearing failure, then helping contractors prioritize those households. 

“We’re not identifying intent, we’re identifying need,” Fred said. That distinction shapes the entire approach. Most homeowners do not actively shop for a new roof until a leak, storm damage or inspection forces the issue. As a result, traditional marketing that targets visible buying intent may overlook a much larger group of aging roofs. According to Fred, that gap represents a significant opportunity: homes with roofs “at or near failure” where the owner may not yet recognize the risk. 

To reduce bias and guesswork, the two said they grounded their system in large-scale data rather than contractor assumptions. Shyam described the challenge of building that foundation, saying one of the hardest parts early on was “cracking the how,” specifically “getting to know all those attributes and data points about an individual” and comparing that information against verified roof replacement records. Fred explained that the team analyzed “20 million records of homeowners,” including five million who had replaced a roof and 15 million neighbors, to establish patterns tied to documented outcomes. By training the model on real replacement activity, the system is designed to rely on measurable data instead of assumptions about ideal customers. 

Imagery is the second pillar. Fred said the company uses computer vision to evaluate roof conditions and partnered with Eagleview for higher-quality aerial images. “Eagleview has the best images in the industry,” he said. The system flags visible deterioration such as missing tabs, discoloration and other indicators roofers associate with end-of-life conditions. 

Shyam said contractors receive results through what he described as a “completely self-serve” platform, where users can log in, select their target area and generate prioritized lists that are delivered in spreadsheet form. He explained that the structure is intentional, allowing contractors to plug the data directly into the outreach systems they already use rather than forcing them to adopt a new marketing process. 

Listen to the podcast or Watch the interview to learn more about AI-driven roof targeting!