In our latest episode of Unstructured Unlocked, Andrew Holdway, Head of Partnerships at Swiss Re Risk Data Solutions, joins us to unpack a persistent challenge in insurance: carriers have more data than ever, yet underwriting decisions are still slowed by inconsistency and fragmentation.
Andrew’s perspective is clear. The issue isn’t access to data. It’s how, and when, that data enters the underwriting workflow.
Listen to the full podcast here. >>>
Inconsistent data use is undermining underwriting and portfolio confidence
Andrew explains that many insurers enrich data at one stage of the process, often during triage, but fail to carry that same data consistently through pricing, underwriting, and portfolio management. Different teams rely on different sources. Manual steps creep back in. Secondary checks introduce new versions of the truth.
The result is operational drag and weakened feedback loops. Portfolio leaders are asked to steer performance without full visibility into what data informed upstream risk decisions. When data use isn’t consistent from ingestion through bind, confidence suffers.
Bringing enrichment and CAT data upstream changes the economics of underwriting
According to Andrew, the biggest shift happens when risk data is applied at submission ingestion and remains attached throughout the lifecycle of the risk.
Upstream triage aligned to appetite creates focus and discipline. Underwriters spend less time reconstructing submissions and more time evaluating risk. Actuarial teams price against consistent exposure data. Portfolio managers gain clearer insight into performance drivers.
Andrew also highlights how granular catastrophe and property data enable precision instead of blunt exits from entire regions. When carriers can differentiate risk at a building level, not just a ZIP code, they can price more accurately, protect capacity, and compete more effectively in softening markets.
Related content from ITC London: Why Insurance Modernization Must Start at the Front Door >>>
AI only works when the inputs are disciplined
AI is part of the conversation, but Andrew takes a pragmatic stance: AI is an enabler, not a solution in itself. Predictive models and automation create value when applied to governed, contextualized inputs. Applied to inconsistent data, they simply scale inconsistency faster.
The common thread across the discussion is simple. If insurers want sharper pricing, faster decisions, and more confident portfolio management, they must start upstream. Modernizing underwriting begins at the point where data first enters the enterprise.
Subscribe to the Unstructured Unlocked podcast to get the latest episodes on your favorite platforms, including:
