In as much detail as possible, describe the problem or experience related to your idea. Please provide the context of what you were trying to do and include specific examples or workarounds:
In planning and forecasting processes, large volumes of textual commentary are captured at detailed levels (e.g., line items, entities, business units). As these comments move up the organisational or planning hierarchy, they become increasingly difficult to review, synthesize, and interpret consistently.
The proposed idea is to use Large Language Models (LLMs) to automatically summarise and aggregate commentary as it “bubbles up” through the hierarchy—from granular inputs to executive‑level views—while preserving key messages, risks, and themes.
How often is this impacting your users?
Every plan cycle.
Who is this impacting? (ex. model builders, solution architects, partners, admins, integration experts, business/end users, executive-level business users)
Without automated bubble‑up commentary summarisation, organisations face information overload at lower levels and information loss at higher levels. This results in reduced efficiency, lower narrative consistency, and weaker decision support—particularly for management and executive stakeholders who rely on concise, high‑quality summaries aligned with aggregated data.
What would your ideal solution be? How would it add value to your current experience?
How It Would Work (Conceptually)
Input level
Users provide written commentary on drivers, assumptions, risks, or deviations at a detailed level (e.g., cost line item, country, product).
Aggregation logic
As data rolls up (e.g., country → region → group), the associated text is aggregated.
LLM‑based summarisation
The LLM generates concise summaries at each level that:
- Identify recurring themes and key drivers
- Highlight material risks, changes, and exceptions
- Remove redundancy while preserving intent
Viewer‑oriented output
The summary is presented for consumption, giving decision‑makers a clear, readable narrative aligned with the aggregated data.
Please include any images to help illustrate your experience.
