Date: 15.12.2025 (2025-12-15)
Christoph Frank (15.12.2025) Context: • https://www.linkedin.com/in/christoph-frank-phd-955338b4/ • Director Group Financial Reporting at gategroup (10k+ employees) • Zug, CH • My message: • Ich bin auf Ihr Profil gestossen, da Sie bei gategroup Finance Reporting leiten. • • Wir entwickeln proprietäre KI-Modelle für Controlling & Accounting. Aktuell automatisieren wir vor allem das Verbuchen von Rechnungen. • • Der nächste Schritt ist die Automatisierung des Monthly Closing & Controlling, mit dem Ziel, den Aufwand um etwa 50% zu reduzieren. • Reply: • Klar wir können uns gerne kurz unterhalten - ich hätte am nächsten Montag Nachmittag 15.12. Zeit oder dann am Mittwoch Morgen 17.12. Senden Sie mir doch einfach eine Teams Einladung auf • CFrank@gategroup.com
Notes: • Notes: • He is very interested in AI • • Automation: • • Biggest objections / challenges in his daily life? • Forecasting: • How to get better in forecasting? • • Automation • How can the existing process be improved? • A lot of potential • A lot of tools are in experimentation phase • Many finance people are not very tech savvy • Reconciliation: • There are quite a few SW provider that can do that already -> Abstimmung Vorsystem und Hauptbuch - e.g. Anlagebuch vs. Hauptbuch; auch Bankenabstimmung • Tool: they do not have a tool - but he is not sure (he thinks SAP, Dynamics can do it) • Data quality and structure is important -> then algorithms can be used -> is missing • He is interested in time series -> he built something - a lot is dependent on tuning of parameter -> he wants to have a solution by press of a button • Also challenge numbers for sanity • He does mostly Konzernrechnung -> abschätzen ob Bandbreite stimmen kann (is it within expected value) • There are not real provider as far as he knows • If data is in cloud one can build it together on their own • People see problems but can not imagine that technology can solve it • Tools he is using? • No AI tools atm. • He has a team that develops ML models (he works himself with Python) • They made some automations -> most valuable: • he built a time series data model (Prophet framework of Meta) -> uses it for revenues • Seasonalities, trends -> additional variables he can give to the model • not usable for 1 year, but 3 to 4 months it is quite usable • Konsolidierungssysteme: • Konzerne haben Validierungen -> Abschreibungen von P&L compare with Bilanz • Sanity checks • -> they built tools to check sanity - developed within Finance team (one person in IND with IT background) • His job: • Compliance / are numbers correct? (IFRS konform etc.) • Models that make that faster - quality maybe not necessary better - maybe a bit - but focus on speed • FP&A • Other department • Forecasts one time per quarter for 12 months ahead • Dream: whole forecast automatically done (press of a button) -> utopisch -> a lot of work to gather all relevant data etc. • He offered in the end that i can always contact him - via phone or even just send invite • Most valuable flows for him? • • Context: team setup? internal and external? • • Which tasks are the biggest pain? With magic wand -> what would he solve? • Number 1: Per press of a button -> Finanzabschluss der publiziert werden muss automatic • Number 2: he tried it but failed • Jahresabschluss is very long -> Check that EBITDA is always the same throughout the document • Put Geschäftsbericht in a tool -> report with errors -> a lot of Konzerne would be interested • Which tasks would he focus on? • • Do they already have automations? Which ones? How well do they work? • • Which companies would he target? • • General feedback to pitch? •