Not automating fund reporting costs more than the hours that go into the factsheets each month. The larger cost is three burdens that grow with every cycle: knowledge that exists only in spreadsheets and in people's heads, dependence on a small number of people, and a process that has to absorb every new requirement by hand. For investment foundations and pension funds as much as for fund management companies and asset managers, FINMA-regulated or not, that is a question of operational risk, not only of efficiency. Waiting is therefore not a neutral decision, but a decision for a process that becomes harder to replace every month.
The first burden is knowledge that is written down nowhere. Manual reporting runs on small corrections: a rounding adjusted by hand, a product with a special rule, a benchmark that has been linked differently since a changeover. Each of these corrections is reasonable on its own, but rarely documented. Every month, the number of rules that exist only in a spreadsheet or in one person's memory grows. Whoever automates later first has to reconstruct that knowledge, and the longer the wait, the more of it there is.
The second burden is dependence on people. In many organisations, one or two people know the reporting process end to end. If they are absent in the week of the month-end close, go on holiday or change jobs, the publication date is at risk. This dependence appears on no budget line, but it becomes visible in every cover plan. An automated pipeline moves the knowledge out of people's heads into traceable logic that anyone who has been introduced to it can read and check.
The third burden is expansion by hand. Reporting requirements rarely shrink: a new share class, an additional language, a further ESG metric, a new group of recipients. In a manual process, every addition raises the monthly effort permanently, and with it the number of places where an error can occur. In an automated pipeline, the same addition is built and checked once and then runs deterministically in every cycle, with the same result on every run.
The three burdens do not add up, they reinforce each other. Every new requirement creates new special rules, every special rule deepens the dependence on the people who know it, and the more dependent the process, the riskier the switch appears. That is how reporting reaches the familiar state of being too important to touch and too laborious to leave as it is. The best moment to automate is therefore usually not after the next expansion, but before it.
What the alternative looks like in operation can be seen at IST Investmentstiftung: Amula AI's reporting pipeline has been live there since October 2025, with 39 investment groups in scope and being rolled out group by group, rollout ongoing. The first production pipeline was live about two months after the strategy phase began. The factsheets are produced in German and French in IST's own design, and monthly effort is down from 20+ to 3–4 hours (an internal estimate based on first measurements).
The honest comparison is therefore not "automation versus nothing", but "automation versus a process that every year ties up more knowledge, creates more dependence and demands more expansion by hand". That is where Amula AI's reporting automation starts. You own the pipeline, templates and logic; it runs in your own environment, on the systems you already use. If you want to judge the output first, you can start with a proof of concept: up to 10 key pages of your report rebuilt in your design with dummy data, before committing to a full build. Know your bottleneck? Start with reporting. Not sure where to start? Start with the AI Assessment.
