“One thing we’re trying to do here … is not serve one stakeholder unconditionally, but make sure we can go to any of those stakeholders — internally and on the platform — and provide a service useful to the current customer,” Horneff explained.
New York-based Noldor launched in late 2021 and has gradually worked its way up to more than 12 employees and counting.
As a data aggregation company, Noldor’s flagship platform interacts with MGAs, delegated authorities, Lloyd’s masters and more. The company does this, Horneff said, to access “structured, unstructured or pseudo-structured” risk exposure claims data, which is then ingested, normalized and transformed into something “more validated and strong”. This allows for easier data consumption between stakeholders in the delegated authority system, including carriers, reinsurance agents, Lloyd’s syndicates or vendors who may need data to provide their services to the MGA- of and coverage holders.
The technology is designed to integrate with any entity that has delegated signing authority, regardless of their existing technology stack. It allows Noldor’s platform to use artificial intelligence and machine learning to aggregate data, uncover hidden drivers of loss reports, and automate back-office functions such as reporting.
In July, Noldor announced that it had raised a $10 million seed funding round led by the DESCOvery group at DE Shaw, a global technology investment and development firm based in New York City, and other strategic investors participated. The founders of Noldor launched the company at the DESCOvery venture studio.
The elephant and reusable data
Horneff turns to a parable to explain the company’s technological approach.
You are probably familiar with the parable of the blind Indians who catch the elephant. One catches a bunch. One catches the tail and they all break different things. The problem I’ve seen firsthand … is that the data access requirements for MGAs and coverage carriers are exactly like that image,” Horneff said. “Carriers care a lot about modeling their cats. Reinsurance agents care much more about generating reinsurance submissions, and each has their own specific need for how that data is being distributed.”
The job of the Noldor, he said, is to “sit on the strife” and create reusable data later that can fill many of those cases. Its integration with an MGA is for border reporting [a report prepared by an insurer for a reinsurer listing assets covered or actual claims paid]but can also be used to help generate reinsurance brokerage submissions.
“It requires us to get more data and make sure that we are, every day, reconciling and validating the data,” Horneff said. “[We’re] making sure we’re reporting things that might break and trying to do our best to allow the pipeline of data acquisition into data analysis to continue and run smoothly.”
In other words, the Noldor help simplify data exchanges with MGAs.
“These MGAs are sending six different things to different people,” he said. “What we’ve allowed them to do is send it to one person and deliver it to six other people … it’s a single point of contact so we can act as a clearinghouse of data for access to MGA data.”
Main ingredients
Data mining technologies, optical character recognition (OCR), and web crawling (a computer program that automatically searches websites for certain keywords) help power the Noldor platform.
“We just go super high,” Horneff said. “We can leverage AI and machine learning to train ourselves on how we’re mining data and start automating some of the human steps required to validate that data.”
In addition, Horneff explained, Nordor can help reduce costs through internal tools that allow it to grab data without having to rely on an engineer to encode the data.
“We’re building an internal technology stack that allows this to be done with a business analyst,” Horneff said. “I can lower the cost of doing that [data] drafting while gaining the advantages of the expertise of someone who may have spent 20 years in the industry and knows how to dictate, but may not know how to code how the data should be translated.”