ZestyAI launches AI roof age solution

ZestyAI has launched Roof Age, an AI-based solution that helps insurers determine the ages of roofs.

Kumar Dhuvur, co-founder and head of product at ZestyAI.

The performance of asphalt shingles in older roofs generally starts declining after five years, and that decline accelerates after 10 years, according to Kumar Dhuvur, co-founder and chief product officer of ZestyAI. “Carriers need confidence in the true age of the roof for appropriate underwriting, rating, and placement of schedule modifications,” he stated in a written response to questions.

Roof Age uses a combination of verified building permits and verified aerial imagery from satellites and aircraft, not drones, to assess the age of roofs, according to Dhuvur. 

This combination gives carriers more data with more accuracy than they would get using just permit data or just aerial images separately, he added. “Permit-based products suffer from low coverage, leaving a significant data gap,” Dhuvur stated. “Aerial imagery-only solutions offer limited historical data, typically only covering the past six years, making them unsuitable for older roofs.”

Roof Age has 90%accuracy and covers almost all of the 48 contiguous U.S. states, according to a company press release. The solution gets completion dates from re-roofing permits and pairs that with AI analysis of photos from the past 20 years. Roof Age’s AI is trained on examples of confirmed re-roofing projects to be able to tell the difference between roof replacements and other repairs or additions. 

Roof-related claims total about $19 billion each year. Carriers depend on policyholder or agent-reported data, and inaccuracies contribute to the cost of these claims, according to ZestyAI. The company recently surveyed 517 homeowners from across the U.S. and found that 63% did not know the age of their roof if they were not in the home the last time it was replaced.

Roof Age fits into ZestyAI’s product suite, which includes Digital Roof, a solution that provides digital twins of structures, and peril-specific risk models for hail, wind and storms.