Imagine if you could simply ask your most complex financial data: “Why are our personnel expenses in the South region above budget?” – and within seconds, you would receive not only an answer, but a complete executive report including a root cause analysis.
What was previously reserved for experienced TM1 experts and power users is now accessible to everyone thanks to the combination of IBM Planning Analytics and AI models such as Claude Code. We are leaving the age of tedious copy-paste behind and entering the era of “agentic finance.”
The invisible link: How Claude learns to understand TM1
The technological breakthrough that makes this connection possible is the so-called MCP server (Model Context Protocol). It acts as a universal interpreter between Claude‘s language intelligence and the multidimensional computing power of the TM1 engine.
Instead of a person exporting data from IBM Planning Analytics and copying it into an AI, the AI agent accesses the data source directly. It understands the cubes, dimensions, and hierarchies of your model in real time.
This “direct access” means no manual errors, no outdated data, and, above all, an end to technical barriers to entry. An AI agent works directly with a real enterprise planning engine.
Democratization of data: Making PA accessible to non-experts
Until now, IBM Planning Analytics (PA) has been a tool for specialists. Anyone who wanted in-depth insights either had to be able to write MDX queries themselves or wait for the IT department to do the work. Claude Code fundamentally changes the rules of the game.
1. Natural language analysis
A specialist user without prior technical knowledge can now ask complex questions. The AI agent translates this query in the background into the necessary query language (such as MDX), retrieves the data, and prepares it in an understandable way.
2. Automated variance analysis
Instead of spending hours searching for discrepancies, the agent scans the data models, identifies outliers, and provides the reasoning right away. The process from question to finished deliverable is reduced from hours to minutes.
3. Reporting without copy-paste
The biggest time waster in controlling is manually transferring values into presentations. Claude Code eliminates this step. The agent creates reports directly from the TM1 engine’s “single source of truth.”
Why this is the development of the future
Is this “agentic” approach just a trend or the new standard? The answer is clear: it is the necessary evolution of financial planning.
AI models are powerful when viewed in isolation, but it is only when combined with the right engine, the right data, and the right context that they become transformative. The combination of a stable, high-performance planning engine (TM1) and an agile AI interface (Claude) is exactly what companies will need in 2026.
For FP&A teams, this means a shift in roles: away from “data collectors” and “report creators” and toward strategic advisors who evaluate the insights provided by the agent and translate them into action.
Conclusion: Set the course now
The integration of AI agents into IBM Planning Analytics is no longer a science fiction scenario. Open standards such as MCP are bringing the vision of seamless, intelligent financial planning within reach.
If you want to open up Planning Analytics to non-experts, you need to rely on Agentic AI. Not only does it increase efficiency, but it also makes your most valuable data available where it is needed: in the hands of decision-makers.
Would you like to learn more about how to integrate MCP into your IBM landscape?
Contact us for a strategy consultation.
