Nobody blames a 2CV for losing a Formula 1 race

Nobody blames a 2CV, the little French post-war economy car, for losing a Formula 1 race. They ask who entered it.

Finance teams get entered in the wrong race all the time. The logic is easy to follow: headcount is the most visible cost in the function, and a large low-cost team looks cheaper than a small team of experts. On the budget line, it is. The cost comes back somewhere else. In the close that still takes weeks, in reconciliations nobody can explain, in a process run exactly as it was taught, because nobody in that seat is measured on changing it.

Expertise is the investment that lowers the running cost.
The order matters:
1.      clean the data,
2.      design and document the processes,
3.      standardise the work, automate what repeats,
4.      and only then bring in AI.
On billing and R2R, that order took automation past 90%, and the business absorbed twice the invoice volume without adding anyone. A €600M+ migration across 14 entities landed with 100% accuracy because 9,000+ vendor and customer records were standardised before anything moved.

AI sits last on that list for a reason. In my own work, the processes, the controls and the direction were set before any model was involved. I used AI the way finance people have always used a search engine, to get the code that put the information in the format I needed, only faster and better tailored. Left alone, it overengineers. Given a clear design, it is an excellent assistant.

When the repetitive work goes, the job changes rather than disappears. Someone still has to confirm the output is right and explain what the numbers mean for the business. That is where finance moves from processing to partnering. Paying a large team to press the same buttons every month and calling it done is a model whose time is running out.

Which race have you entered your finance team in?