Markel

Markel, a leading specialty insurer recognized for its multi-channel delivery and one of only four UK insurers with a positive broker NPS score, identified a structural bottleneck in its underwriting operations that was constraining GWP growth. Senior underwriters — the most experienced capacity in each team — were required to manually triage every incoming submission, assessing appetite fit before routing or declining. This created a significant opportunity cost, diverting skilled talent from complex risk decisions and broker relationships. Compounding the issue, underwriters spent over 30% of their time on low-value administrative tasks: re-keying risk data into CRM and policy administration systems, and manually retrieving third-party data. Capacity was further consumed processing out-of-appetite submissions that would never convert to quotes, slowing turnaround times for genuinely quotable business and damaging customer experience.

Markel partnered with Cytora to redesign their underwriting operations as end-to-end digital risk flows. Cytora's platform applies NLP and AI-supported automation to handle the pre-underwriting layer that had previously consumed expert capacity. Incoming broker submissions are automatically ingested and digitized, with NLP extracting and structuring risk data from unstructured documents. The platform enriches each submission with third-party data, applies Markel's appetite rules to triage and score risks, and routes decision-ready submissions to the appropriate specialist underwriter — bypassing the manual senior-underwriter triage step entirely. Downstream system population, including CRM and policy administration, is handled automatically, eliminating the re-keying burden. The approach was explicitly designed to decouple GWP growth from headcount by concentrating underwriter attention on evaluated, in-appetite risks rather than raw submission volume.

The Cytora implementation delivered measurable gains across productivity, speed, and strategic recognition:

• 113% uplift in underwriting productivity (GWP per FTE), effectively more than doubling output per underwriter without proportional headcount growth.
• Quote turnaround time for strategic broker partners reduced from 24 hours to 2 hours — a 12x improvement that directly supports Markel's broker NPS position.
• 30%+ of underwriter time previously spent on low-value tasks was reclaimed and redirected to complex risk assessment and relationship management.

The initiative was recognized internally at Markel's Annual Shareholder event, reflecting the business impact at the executive level. Following these results, Markel deepened its partnership with Cytora to extend the digital risk flow model further across its operations.

• Start with process design, not tooling. Markel's results came from redesigning the underwriting workflow end-to-end before deploying AI — the technology enabled a target operating model, not the reverse.
• Triage automation yields disproportionate returns. Removing senior underwriters from routine submission filtering frees the highest-value capacity for complex risks and broker relationships.
• NLP on unstructured submissions is the practical entry point. Automating data extraction and enrichment from broker submissions eliminates the re-keying burden that silently absorbs 30%+ of underwriter time in legacy workflows.
• Broker SLA performance is a competitive differentiator. Compressing quote turnaround from 24 hours to 2 hours has direct NPS implications — speed of response is measurable and meaningful to distribution partners.