Table of Contents
Key takeaways
- ReBuy's breaking point was a board deck where the numbers in Google Sheets no longer matched the screenshots already in the presentation, the night before a board meeting.
- ReBuy shortlisted six vendors (IBM, Board, CoPlanner, Anaplan, Jedox, and Pigment) and chose Pigment for flexibility and the ability for finance to own modeling itself.
- Siemens Mobility Software built its Pigment application, covering workforce, cost, revenue, and reporting, in under five months with implementation partner Valantic.
- Siemens cut its forecast cycle from four to six months down to two to three weeks and now runs about 10 forecasts a year.
- Both companies underestimated how much structure they'd need to define themselves rather than getting out of the box.
- The shared advice for anyone starting a selection process: understand your data and processes before you implement, not during.
The moment two finance teams decided enough was enough
ReBuy and Siemens Mobility Software came to Catalyst Berlin with different businesses but the same starting problem: planning processes that had outgrown Excel and Google Sheets.
Mandy Kramer, head of controlling at ReBuy, a European recommerce company for consumer electronics and media that handles roughly 5 million items in stock and reported €240 million in revenue last year, described the moment clearly. Before a board meeting, her team had spent a late night getting a slide deck aligned with the numbers. The next morning, the Google Sheet showed different figures than the screenshots already in the deck. A broken data connection had quietly changed the numbers overnight. "This is not the situation we can accept anymore," she told her CFO.
Marvin Persing, team lead for FP&A at Siemens Mobility Software, described a slower-burning version of the same problem. The business unit, built from acquisitions including HaCon (founded 1984, about 450 employees), had a consolidation system but nothing connecting planning across teams. Salesforce pipeline data lacked proper versioning, so after every forecast cycle, people showed up to alignment meetings with different versions of the truth.
How each team evaluated the market
Both ran structured, multi-vendor selections:
- ReBuy worked with the analyst firm BARC and consultancy firm CloudSpot, shortlisted six tools (IBM, Board, CoPlanner, Anaplan, Jedox, and Pigment), and ran a cross-functional project team spanning tech, accounting, and controlling.
- Siemens Mobility Software compared Anaplan, OneStream, Jedox, and Pigment.
Both ended up doing a deep dive between Jedox and Pigment, and both chose Pigment, but for slightly different reasons. Kramer said flexibility was the deciding factor: the team didn't want to hit a point where "we cannot plan this in detail because the tool doesn't allow us," and wanted finance, not IT, to own the modeling. Persing said Pigment had fewer features than competitors at the time his team decided (late 2024), but the team's engagement and willingness to evolve the platform built trust that outweighed the feature gap.
What surprised them during implementation
Neither team got the out-of-the-box experience they expected.
Kramer said she assumed basic structures like P&L and balance sheet layouts would come pre-built. Instead, ReBuy's team had to define most of that structure itself, which took longer up front but gave them a model with no technical limitations once in place. ReBuy split the rollout into two phases: financial reporting first (five legal entities in Navision plus one in Comarch, all previously consolidated by hand in Google Sheets), with planning logic as phase two.
Persing's team hit a different wall: integration. Siemens Mobility started connecting SAP (via Snowflake), Personio, and Salesforce four to five months before the implementation deadline and still didn't finish in time, forcing a temporary CSV workaround and a model rebuild once the real connections landed. His advice for anyone starting a similar project:
- Clean and harmonize your data layer before you start building in the platform, not during.
- Treat integration timelines as longer than they look, especially in an enterprise environment that won't wait for your project schedule.
- Reuse your existing team where you can, but budget for additional hires once demand outpaces what the current team can deliver.
What they'd tell someone starting today
Kramer's advice to other controllers in the room: block planning capacity on your calendar and defend it against competing priorities, because the time invested up front in mapping out the structure "pays off tenfold down the road." She also drew a clear line on AI: the Modeler Agent and Analyst Agent speed up modeling and analysis, but AI can't fix bad data or flawed logic, and accountability for the numbers stays with the controlling team, not the algorithm.
Viviana Guzmán Yacamán, planning and analytics manager at Valantic, who worked with Pigment both as a client and now as an implementation partner, gave the same core advice from the other side of the table: understand your data and your processes before you implement, because teams that try to fix those problems mid-implementation run into trouble.
Persing's closing line summed up both experiences: "You can start small, and then scale smart."
Request a demo to see how a connected model handles your own reporting structure.
