A claims handler who became a chief information officer

On the morning of 17 January 1995 I was asleep in a company dormitory in Osaka, nine months into my first job as a claims handler at Nishikaze Fire and Marine. By the following week I was working from a rented warehouse in Amagasaki, sorting paper claim files into cardboard boxes marked by district. Most of the households I visited in Kobe that spring had no earthquake cover, and their fire policies excluded fire caused by an earthquake. I spent weeks explaining a contract to people who had signed it without understanding it, in streets where the houses had burned. I have never since been able to think of an insurance system as anything other than the means by which a handler answers a frightened person quickly and honestly.

I studied law at Ikutama University, graduating in March 1994, and joined Nishikaze straight afterwards, as most Japanese graduates of my generation joined their employers. In 1999 I moved to large-loss property claims in Hiroshima, adjusting factory fires and typhoon damage to warehouses and schools. I became a claims supervisor in Osaka in 2003. That was when I noticed that the claims system decided how my handlers spent their days: which fields they had to complete before they could pay, which approvals queued behind which, and how often they printed a screen to carry it to another desk. When Nishikaze began a new claims system in 2004, I volunteered to represent the claims department, and I studied information management in the evenings at the Hinokage Institute of Information Studies to understand what the project team was telling me.

In 2007 I crossed over properly, as manager of claims systems planning. The defining test of that period came on 11 March 2011. Nishikaze had several thousand household policies in the tsunami areas of Tohoku. The industry agreed to treat whole districts as total losses on the evidence of aerial photographs, which spared families from waiting for an adjuster to visit a house that no longer existed. My team built the means to apply those area decisions to individual policies, check them against addresses and pay without a site visit. It was the first time I saw a systems team shorten a family's wait by weeks.

Nishikaze merged into a larger domestic group in October 2011, and the claims systems I had helped to build were scheduled for retirement in favour of the acquirer's. I led the migration of our claims records and then left, because I wanted to build rather than decommission. In July 2012 I joined Suzukaze Direct, the Tokyo-based subsidiary of a fictional European insurance group, which sold motor cover online and by telephone. Everything I had learnt in agencies and warehouses had to be relearnt for a business whose only branches were a website and two contact centres. I rebuilt its claims and telephony systems, and from 2017 I led its whole technology function. When the September 2018 typhoon flooded car parks at the Kansai ports, we settled thousands of vehicle claims from photographs, and I learnt how much a claims process can be shortened when the evidence arrives with the first report.

In 2020 our parent decided to run its Asian subsidiaries' systems from Europe. At the same time Kotonami's president asked whether I would join as CIO, the first executive officer in its technology function hired from outside the company. I accepted because Kotonami carried everything Suzukaze Direct did not: household catastrophe exposure, an agency network, older customers and decades of systems built by layers of contractors.

I lead about 760 people: 120 in the CIO's office and 640 at Kotonami Information Systems, our subsidiary, where I chair the board. About 900 more work for us through system integrators and their subcontractors. I report to the president, present to the board whenever a system decision is material, and report twice a year to our Audit and Supervisory Committee on IT controls, outsourcing and cyber risk. I ask my teams for evidence rather than reassurance: test results rather than status colours, and how long the oldest open problem has waited rather than how many were closed.

Inside the industry I am probably known as a claims person's CIO, which I take as a compliment. I hold three views that have cost money. A system programme should be stopped as soon as the evidence says it will fail, whatever has been spent. Catastrophe capacity in an agency insurer is mostly people with the right permissions, not servers. And an insurer that lets policyholder data leave its systems has not protected it, whatever its rules say. I am now preparing for a second responsibility beside my executive one: an outside directorship, where I would ask the questions I have spent thirty years learning to answer.

Open to Board appointments · Speaking invitations

Languages / Japanese · English