90% Analysis, 10% Data Wrangling: How Vave Flipped the CAT Analytics Equation

Vave flipped the traditional CAT response model: from 90% of time spent on data preparation and 10% on analysis, to nearly zero time on data prep and over 90% on making high-value decisions based on insight. During the active 2024 hurricane season, this translated to loss estimates that were both fast to market and accurate, requiring no material reserve adjustments.

Vave is a digital-first property MGA that processes underwriting decisions through an API-driven platform. Every risk the company evaluates flows through automated systems that leverage building characteristics, geospatial attributes, proprietary scoring models, and external hazard data. 

Analytics sits at the center of this model. As Tim Spencer, Head of Analytics, explains: “The purpose of analytics is to derive value from the data Vave holds.” 

Vave operates lean by design and prioritizes automation and technology over headcount. This philosophy shaped their approach to catastrophe risk management from the very beginning and led them to EigenRisk.

Unlike many case studies, this isn’t a turnaround story. Vave didn’t adopt EigenPrism to fix something broken. They adopted it to avoid building broken processes in the first place. 

Spencer’s experience prior to Vave shaped this decision: “My pre-Vave experience around doing event response was always that you spent 90% of your time grappling with the exposure data to work out exactly what your in-force portfolio was at any given time, and then 10% of your time doing the actual valuable analysis.” 

Rather than repeat this pattern, Vave integrated EigenPrism early in its journey. The result: they were able to skip the manual workflows entirely and build automated CAT intelligence into their operating model from day one.

Automated Moratoriums via API 

Vave’s original use case was automating moratorium decisions—pushing current cat data directly into their underwriting system without manual intervention. What began as a proof of concept moved to production in approximately two months through joint development with EigenRisk’s team. Spencer shares that moratoriums run on autopilot: “I don’t need to think about that. I just know it’s happening.”

Real-Time CAT Event Response 

Vave’s portfolio data refreshes overnight, every night. When a CAT event strikes (or is approaching) the team can immediately identify exposed assets and evaluate accumulations without scrambling to prepare data. During the January 2025 California wildfires, this meant the ability to monitor fire extent changes daily and assess portfolio impact in near real-time. 

The value compounds when analyzing historical scenarios. Because EigenPrism retains expired policies, Vave can quickly answer questions like “What would this event have looked like on our portfolio six months ago?” Normally, this kind of work in a manual environment requires a heavy data lift. Now, everyone’s time is spent on analysis instead of information gathering.  

Pre-Event Confidence When It Matters Most 

During the 2024 hurricane season, Vave used EigenPrism to generate loss projections 24-48 hours before storms made landfall. Instead of scrambling to build scenarios manually, the team ran analyses that produced clear probability ranges they could immediately share with leadership and capacity partners. 

As Spencer put it: “There’s a 5% chance this thing could be quite bad, but there’s a 95% chance it’s going to be all right.” That kind of clarity, delivered before an event hits, gave stakeholders confidence and the business time to act.

While EigenPrism sits with the analytics team, Vave has provided access across the business. Product managers use the platform to evaluate why they’re winning certain risks and whether pricing should be adjusted based on hazard layers like Fathom flood data. This creates a more integrated approach between exposure management and pricing decisions.

I’d definitely recommend EigenPrism. I think it’s somewhat unique in the market in terms of the ability to interface hazard layers on top of exposure data. The new functionality that has come in the last six months has been really valuable – you’re filling the gaps before I’m recognizing them.

– Tim Spencer, Head of Analytics, Vave

90% of time goes to analysis, not data prep: What was 90% data preparation and 10% analysis at past companies is now “almost zero” on data prep and over 90% on generating actionable insights. 

Accurate loss estimates: During the active 2024 season, Vave’s initial estimates required no material reserve adjustments, making them fast to market and accurate. 

Confidence in automation: Moratoriums, exposure refreshes, and hazard layer updates run continuously without manual oversight, freeing the team to focus on strategic analysis. 

Faster communication to capacity partners: The ability to generate insights quickly strengthens Vave’s relationships with capacity providers during event windows. 

Reduced build-vs-buy risk: Spencer notes that building an equivalent solution in-house, and maintaining relationships with hazard data providers, would be a significant ongoing overhead: “You’d never fully claw back what you’ve invested upfront.”

Vave operates lean by design. But, Spencer estimates that for a traditional London market insurer with a 30-40 person CAT team, the manual workflows EigenPrism eliminates could easily represent two or more FTEs of effort: “When you work out what people are doing across their full year, there’s probably 5-10% of people’s time spent doing that sort of thing. You could probably save multiple FTEs with a technology-first, integrated exposure and hazard approach.”