I learnt to look for the work behind the variance

I learnt to distrust a comfortable margin before I worked in technology. At Mersavel Components, my first employer, the monthly cost report could remain reassuring even when supervisors were dealing with more changeovers, rework or waiting time than the standard assumed. I was a junior analyst, not a production manager. My useful work was to visit the process, understand the difference and bring it back into the account without pretending that a revised spreadsheet had repaired the factory.

I joined in September 1992 after completing a management-accounting degree at Averielle Institute that June. The role involved cost preparation and variance analysis under the plant controller. I learnt to ask what a measure included before comparing it with another. A direct production margin could be useful without being the whole company's profit, and a favourable labour variance could coincide with work postponed into a later period. Those distinctions shaped how I later approached services, where the unfinished work is often less visible than a physical stock of components.

In March 1998, I moved to Corvessa Software as a project controller. The business implemented enterprise systems, and the critical resource was experienced people's time. A signed contract did not tell me how many days remained, which customer dependency could delay acceptance or whether the delivery estimate had been updated since the proposal. I worked with project managers to make those questions part of the recurring review. I did not direct the technical solution. I wanted its financial account to reflect the solution the team actually expected to deliver.

Promotion to regional finance director in July 2005 gave me responsibility for a larger portfolio and an eighteen-person team. We supported delivery across France and Belgium. I became more aware of how an apparently sensible local incentive could produce a poor enterprise choice. Keeping every engineer billable could leave too little time to learn a new platform; accepting an attractive fee could create a delivery bottleneck elsewhere. I began separating the use of capacity from the contribution earned by using it. A busy team and a well-run business are related, but they are not identical measures.

I joined Avelrune Digital Services as group controller in January 2012. The founders wanted a stronger connection between managed-service accounts, implementation work and the group result. At the end of that appointment, the company had 430 employees and EUR61 million of annual revenue. My team reconciled different operating measures and established a consistent account of contract changes. I learnt to preserve the detail that mattered to delivery while giving owners a view they could compare from one reporting period to the next.

I became CFO in April 2019. A minority investment partner joined the ownership in 2020, while the founders retained a majority. The change increased the demand for a clear account of growth quality, cash and investment, but it did not change the need to understand the work behind a project estimate. I lead forty-four finance colleagues through five direct reports and report to the chief executive. I attend relevant board discussions as management. I have no current directorship, and I would not describe those presentations as independent board service.

One project brought the role into focus. A EUR2.8 million implementation still carried its original EUR2.14 million delivery-cost estimate even though the team expected EUR2.59 million of cost. I asked for the revised view before discussing how to recover the difference. A later scope agreement added EUR380,000 of fees and EUR100,000 of associated work. The resulting EUR490,000 forecast direct contribution was a less comfortable figure than the original proposal, but it was a better basis for deciding how to finish the assignment. It was also not the group's net profit.

I have made similarly deliberate choices about learning time. Protecting 12,400 paid training hours for 155 engineers cost EUR620,000 in internal labour and reduced available billable capacity during the period. I supported the programme because the delivery capability mattered, while refusing to declare that every hour would return a predictable amount of revenue. I wanted managers to explain the skills and assignments the learning would support, and to review what actually changed after people returned to project work.

My next chapter should preserve this connection between the account and the operating reality. I am interested in a first external board role where a substantial people-cost base, long assignments or recurring service commitments require careful financial judgement. I would bring questions learnt in a factory and developed in technology delivery, without assuming that one industry's measures fit every business. I want a chair and management team who can discuss an unwelcome estimate early, while there is still a practical decision to make about the work.