Leveraging Data, Analytics, and Artificial Intelligence in the Guidewire Ecosystem.
In recent years, insurance companies have dedicated more resources to take advantage of technology trends like data management, analytics, and artificial intelligence to improve efficiency and customer experience. Guidewire, a leading provider of software solutions for the insurance industry, has developed many tools to unlock innovation potential for data, analytics, and artificial intelligence projects.
In this blog post, we will explore some products created by Guidewire to help insurance companies be on top of these technology trends and we’ll talk about generative AI, giving some examples of how insurance companies can use it.
- Integration Data Manager
Integration Data Manager, recently included in Guidewire Hakuba version, is a document storage service solution that stores JSON documents outside of Guidewire InsuraceSuite applications. With this new service, the customer can enrich InsuranceSuite data with data from third-party systems without storing the information in their core data model. Having the data from external systems integrated with the InsuranceSuite enables a streamlined user experience for accessing the data without the need to log into multiple systems. Also, it provides the tools to make better data-driven decisions.
- Guidewire Predict
Predict leverages AI throughout the insurance lifecycle. It is a platform to build predictive models and can be used to integrate real-time results in PolicyCenter and ClaimCenter. With this feature, each company can deploy its own models and obtain real-time predictive insights. Different business cases take advantage of this product, for example: Automatically assign claims to specific skilled groups, improve quoting process, perform risk assessments to quotes, commercial risk assessments and pricing. The real-time recommendations can be customized by each insurance company as needed; for example, it may be used to trigger business rules, activities, underwriting issues, or any custom notification.
Hazardhub provides accurate information for property and casualty in seconds. The companies can use the data provided by HazardHub data to pre-fill data into core systems, improve their underwriting processes considering this additional information (property characteristics and risk scores), and make better decisions about when to do inspections based on the property information. HazardHub provides scores for multiple perils, including earth, wind, fire, water, and manmade disasters. These scores may be used to get a more accurate underwriting analysis.
Canvas helps companies to geo-visualize claims and policy locations. In case of a catastrophe, this product can help insurance companies to proactively take actions like reaching out to policyholders and providing useful information considering their current situation about the catastrophe. The companies can use a data-driven approach to prioritize hardest-hit areas and identify potential fraudulent claims.
Compare gives you feedback about how the company is performing. This product uses key claim metrics and compares them with peers and industry benchmarks. Insurance companies can use the result of this comparison to improve claim processing efficiency by identifying which metrics to improve. The data is provided by Guidewire Data Cooperative, which is a customer consortium. It is necessary to join this group to use this product. The data provided is secured, anonymized, and continuously updated.
Cyence improves data-driven underwriting decisions by assessing cyber risk and projecting loss ratios and insights considering today’s cyber-attacks. This product uses the latest data available from the companies. In the new Guidewire version (Hakuba), Model 6 was released; this version improves the accuracy of previous versions. This product includes dashboards at the company and portfolio level that can be used to understand the company risk quickly. Additionally, comparing how the risk has changed between months is possible.
- Guidewire Explore:
Explore is a product to get many actionable insights from the data available to be analyzed. Companies can personalize search data using formulas that can be saved and reused in the future. You can set up alerts for key performance indicators. Explore includes artificial intelligence capabilities to enhance data analysis by suggesting filters, attributes, and measures to liveboards. Explore also includes SpotIQ, which is an artificial intelligence capability to create automated insights from your data, point out important trends, and can be used to identify outliers, trends, and correlations.
- Data Cooperative Predictive Scores (Subrogation Likelihood Score)
Guidewire provides scores that indicate the likelihood of recoverable subrogation on a collision claim. This feature provides the solution detail (description, estimate, and score) and the predictive factors contributing to calculating the score. The score is created with claim information from US companies. The customer can import the “Data Cooperative Subrogation Model” solution from the “Analytics Manager” section in ClaimCenter, and customize the input variables, applicability, and appearance in the UI based on their needs.
Generative Artificial Intelligence:
In simple terms, Generative Artificial Intelligence is a technology that can create things just like humans do, but it’s powered by computers instead of human brains. Generative AI learns from many examples and then uses that knowledge to come up with new and original creations on its own. From a more technical perspective, Generative AI uses neural networks, and it is trained following an “unsupervised learning” approach from data without explicit guidance or labelled examples. Some models used for generative AI are Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs). Generative AI can provide significant benefits to companies. Here are several ways in which generative AI can assist insurance companies:
- Get accurate answers from unstructured data:
Companies can use AI generative models to retrieve accurate information from unstructured data using natural language processing (NLP). This information could be available to users via an interface or by enhancing a chatbot solution. Having the required information on time can improve internal search experiences and customer interactions. Contact centres is a use case for this feature; agents can provide faster responses and improve customer experience using this functionality. As a reference, a product that uses NLP to get answers from unstructured data is AWS Kendra.
- Increase business insights from unstructured data:
Large amounts of unstructured data can be transformed into structured data using generative AI. This feature can reduce the time to obtain critical information from the market and provide readable documents, reports, or KPIs that can be used to take specific actions by companies. Companies can deploy architectures to read unstructured data and include AI generative solutions to consume it. AWS provides an architecture that can be used as a reference for this functionality.
- Product Development:
Generative AI can help the design process of products based on specific inputs and constraints. Developing this feature will speed up the initial ideation phase and may be used to optimize existing products based on client feedback. Each company may design their own models based on their needs.
- Improve Developer Productivity:
The developers can improve their productivity by receiving code recommendations based on natural language comments. This action will reduce the interaction with documentation and help developers with unfamiliar languages to focus on the expected functionality. As a reference, AWS Codewhisperer is a tool that can be used to achieve this objective.
Guidewire products leverage data, analytics, and artificial intelligence to transform the insurance industry. They provide different products ready for the customer and frameworks to build specific solutions adapted to each customer.
Companies should analyse the rapidly evolving new technologies to determine possible market impacts. It is optional to have big innovation teams. Still, at minimum, it is necessary to have key roles that are constantly investigating and learning new technology trends and can bring those topics to be discussed in strategic meetings for each company.
Companies that embrace these innovations are better positioned to adapt to changing customer expectations and market dynamics while mitigating risks effectively.