Use Generative AI to summarise commentary from bottom of hierarchy upwards to top of hierarchy

TristanS

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:

Different roles enter and consume "contextual" data (commentary) at different levels of a product / customer hierarchy. It is the contextual data (story) that provides users a better understanding of the forecast numbers. E.g. revenue forecast has dropped by 10% compared to previous year. But why? If a comment is provided advising this was due to discontinuing a non profitable product then the drop in revenue isn't such a bad thing. Need a generative AI capability to rollup/summarise commentary entered at the bottom of the list hierarchy up to the top of the list hierarchy. This facility will truly enable connecting users at different levels with generative AI by contextualising the information at correct level of detail/summary required by the level of the role consuming the information.

For example, in a demand planning application. Demand planners would be entering their forecast commentary at a SKU level. But a CFO or CEO reviewing the forecast would be reviewing it at a product category or business division level.

How often is this impacting your users?

During planning reviews.

Who is this impacting? (ex. model builders, solution architects, partners, admins, integration experts, business/end users, executive-level business users)

Business end users. Middle / Senior and Executive management team that need to consume information entered at lower levels at a much higher level.

What would your ideal solution be? How would it add value to your current experience?

Refer to screenshots provided below for ideal solution. This will improve collaborative and connected planning

Please include any images to help illustrate your experience.

Example below shows a 3 level product hierarchy

When demand planner enters comment at SKU level (product 2). The comments box on the card is highlighted with a blue dot.

When a user select the higher level ancestors (segment 1), the comment box highlight disappears. User who would normally view information at this level loses visibility of the comments in the lower list nodes.

In my suggestion, the highlight on the comment box will still appear. When the user clicks on the comment box, it will have information from the lower leaves summarised. Similar to how chatgpt users can configure how answers are presented, model builders / users should have the ability to configure parameters for comment summarisation. e.g. to less than X number of words, provide context to quantitative impact, etc … . E.g. 700 Units and $500,000 revenue increase due to project X in July 25 and $300,000 decrease due to discontinuation of product sub segment 2 products from Sep 25. Having said that, users should still be able to add a comment at the Segment 1 level (as per current functionality supporting multiple comment entries) to augment additional information to what Gen AI has already provided.

To improve useability of data entry of comments at the lowest level (SKU) for demand planners, it would be best if the comments box can be linked to a lineitem from a module. So demand planners would enter comment against a lineitem in module which would then flow through to the comments card via a linking facility.