Ping Data: Fast-tracking SOVs into your underwriting ecosystem

Stuart Mercer, Chief Executive of Ping Data Intelligence, introduces his AI-driven insurtech and its revolutionary approach to data accuracy, now integrated and available to insurance carriers through AdvantageGo’s Underwriting workbench.

What is Ping’s mission statement?

Ping’s mission is to revolutionise commercial property data through the innovative use of machine learning to process unstructured data and provide exposure intelligence for property risks across Property, Terrorism, Realty and Builders’ Risk business lines.

As a leading provider of global exposure technology to the property (re)insurance industry, we are harnessing the power of artificial intelligence to create data ingestion, cleansing and geocoding solutions that will help to inform more accurate risk modelling and underwriting processes.

Our goal is to provide a global portfolio of easy-to-use data services and products, which integrate with existing underwriting systems to deliver actionable data within minutes, in a pre-determined format which will enable clients to carry out advanced geospatial analysis and support all aspects of the underwriting process.

Through our partnership with AdvantageGo, our aim is to enable users to cleanse and reformat vast amounts of unstructured property data in a matter of minutes, a capability which in its wider market context we believe will help to transform the way that property data is processed in the London insurance market.

What are SOVs and why are they important?

Statements of value (SOVs) are key documents submitted by insureds to intermediaries, detailing the values of buildings, contents and business interruption exposures for each property to be insured. SOVs are subsequently used by underwriters to assess the insurable value of client assets, calculate premiums, and inform exposure management.

When I joined the (re)insurance market, one thing that was clear to me was the poor quality of the data which the industry relied upon. This was having a huge negative impact on the ability of carriers to understand the risks in their portfolios.

Still today, tens of thousands of SOVs are submitted daily to insurance businesses, typically in unstructured formats that often require the use of external data-formatting services to manually reformat data. This a time-consuming process adding billions of dollars of unnecessary cost to the insurance value chain. That is simply not sustainable, and with the capabilities now available, not acceptable.

Step by step, what does Ping’s data ingestion, data cleansing and geocoding system do?

The ingestion, cleansing and geocoding process from start to finish is designed to be as simple and straightforward as possible for the client.

Users of Ping don’t need to integrate or install software, they simply submit the SOV file – typically in Excel format – via an email or make an API call to our platform and receive the scrubbed or cleansed SOV in return, usually within three minutes of sending the original file.

Ping’s sophisticated data ingestion system uses machine learning to cleanse the data, before supplying it to the user in a structured format which automatically aligns with any specific underwriting rules they might apply. The process ensures a high degree of property data accuracy which is ready to be used instantly for catastrophe modelling and underwriting purposes.

As part of this scrubbing, our geocoding capabilities are designed to provide the highest levels of location precision. By combining Ping’s proprietary information with a range of external data sources and leveraging multiple geocoders, we’re able to deliver that level of location accuracy from the individual property level right up to the portfolio level.  

What does the AI enable it to do, which otherwise wouldn’t be possible?

One of the critical aspects of the way that Ping is structured is that we have built on deep and broad (re)insurance experience and expertise and underpinned that market knowledge with the machine-learning capabilities that inform our proprietary global exposure technology. It’s that unique combination that has helped us to bridge the gap between the insurance and technology professions.

Through the advanced application of our machine-learning capabilities, we are able to provide clients with rapid data enhancement while removing the need for expensive, time-consuming manual processes. The platform can augment, highlight, and enrich incomplete data in an instant, acquiring and integrating data from multiple internal and external sources, and converting SOVs and ACORD apps into a structured Snowflake database while preserving full data provenance and metadata.

Our AI-powered technology enables property data augmentation and visualisation, enhancing exposure intelligence for property risks across multiple property business lines. Further, as new data becomes available that can be integrated instantly into the ecosystem, while the inherent flexibility of the Cloud infrastructure means that that ecosystem can evolve to meet the growing data needs of the user at any point.

Like Ping, AdvantageGo is focused on helping underwriters to underwrite better; how much of a difference does Ping’s approach make to them?

Insurers currently spend a lot of time, resource and money on scrubbing data – that should not be the case. It’s about facilitating companies to develop the most efficient underwriting ecosystems possible, which starts with the way data is processed.

The phrase we like to use is ‘Powered by Ping’. With our platform, users can reduce the cost of data cleansing by 50% or more and receive the results in a fraction of the time. In fact, we’re able to do in minutes via an API a function that is taking many teams days to complete, enabling those insurance specialists to devote their time and energy exclusively to the central task of analysing and applying that data.

Users also have the confidence of knowing that the decisions they make are based upon the highest levels of property-data accuracy. That fact that the geocoding information for a property or entire portfolio of properties is accurate and the data as complete as possible provides an incredibly robust base for all underwriting and modelling activities.

The process also removes the need for numerous manual touchpoints and dramatically reduces the requirement to rekey property information on an annual basis. By using Ping, users automatically gain access to an extensive Snowflake data lake, with the data provenance and metadata preserved.

With their data properly organised and stored, clients can access numerous capabilities: automated professional property submissions; auto-population of in-house underwriting frameworks; and advanced geospatial analysis for improved diversification. That’s the real power of Ping.