In his book Co-Intelligence, Ethan Mollick suggests firstly to invite AI to the table in everything you do. And secondly to make sure to be the human in the loop that works with AI.

Working with AI/GenAI is a skill that can be developed. Prompting GenAI certainly takes practice and Mollick provides some interesting and well thought through prompt examples.

If you’re an IBP manager and you’re making your Powerpoint slides (really, should we still be doing that?) for the coming Exec IBP meeting, it is not always easy to quickly understand all changes from one month to another. What are all the moving parts across product portfolio, demand, supply, inventory, strategic projects and financial projections we need to take note off? How are they related?

Taking MolIicks advice, I searched up two of my last monthly IBP meetings I facilitated about a decade ago. I uploaded October F14 and asked my own GPT (IBP recommendation guide) “Provide me a summary for this IBP meeting deck highlighting risk & opportunities”.

Within second ChatGPT answered with below executive summary (which was correct after I checked this deck I hadn’t seen for ten years) with additional risk, opportunities, strategic actions and next steps I don’t show here.

Executive Summary

• Financial Performance:
  • EBIT Projections: H1 shows a $4.3M unfavorable deviation from the Annual Operating Plan (AOP), and the full year (F14) is projected to be $6.9M unfavorable.
  • Demand & Supply:
    • F14 demand plan is on target with AOP at 62,723 tons. However, there is a 1199-ton gap versus AOP in H1.
    • Net Sales Value (NSV) gap is $5.1M for F14, with a $8M gap in H1.
    • Gross Profit at standard (GP@std) shows a $3.9M gap versus AOP.

Thanks GenAI. Now imagine that ten years ago I was working on the November F14 deck. I upload the last month IBP deck (October) and ask ChatGPT: “Read this October F14 IBP deck and remember it”. To which it answered

“The contents of the October F14 IBP deck have been read and retained. How would you like to proceed with this information? If you need specific insights or actions based on the deck, feel free to ask!“

My next prompt: “Now read this November F14 deck and call out the key changes in financial projections versus October F15. What has changed?” See the typo (F15 instead of F14)?

ChatGPT answered

Here’s a summary of the key changes in financial projections between the October F14 and November F14 IBP decks:

Financial Projections:

• EBIT Projections:
  • October F14: H1 projected a $7.3M unfavorable deviation from AOP, with a full-year (F14) projection of $9.8M unfavorable to AOP.
  • November F14: H1 projected a $3.1M unfavorable deviation from AOP, but full-year EBIT is now projected to be $564K favorable to AOP.
• Net Sales Value (NSV):
  • October F14: A $5.4M gap in NSV versus AOP was reported.
  • November F14: The NSV gap reduced slightly to $4.7M versus AOP.
• Gross Profit at Standard (GP@std):
  • October F14: The GP@std gap was $3.9M versus AOP.
  • November F14: This gap increased slightly to $4.7M versus AOP.

Conclusion:

The financial outlook improved slightly from October to November, primarily due to better yield management and supply chain cost controls. However, the GP gap widened slightly, and NSV challenges remained significant. The November deck also emphasized more granular changes in operational plans, especially regarding capacity adjustments and product-specific strategies.

Thanks GenAI. Now I understand (some of) the changes between IBP cycles on a financial level. Of course I could prompt for other changes as well and ask how they could be related. ChatGPT’s numbers were correct. As this is ‘real’ data its reading, I think there might be less chance to hallucinate.

I then prompt: “What would you recommend to the executive based on these finances?” to which it answered six pieces of advice for which I only show the first two.

Based on the financial insights from the October and November F14 IBP decks, here are several recommendations for the executive team to consider:

1. Focus on Yield and Cost Management Improvements

• Recommendation: Continue to emphasize yield improvements and cost management, which have already positively impacted EBIT. The November data shows a $4.3M favorable yield improvement and better conversion costs. Prioritize initiatives that can further enhance these areas, such as investing in technology or process optimizations that reduce waste and increase efficiency.
• Action: Set up cross-functional teams to identify additional yield improvement opportunities, particularly in areas like Categeory1 and Category2 where the November results show capacity constraints.

2. Address NSV and GP Gaps with Targeted Strategies

• Recommendation: The NSV gap remains a concern, with a $4.7M gap versus AOP. Given the slight widening of the GP gap, focus on strategies that can close these gaps, such as price adjustments, product mix optimization, and enhancing sales efforts, especially in underperforming categories like Pre-pack.
• Action: Implement targeted promotional campaigns or pricing strategies for underperforming categories, especially in the Retail1 and Retail2 channels, where pricing challenges have been identified. Also, consider renegotiating terms with key customers to improve NSV and GP margins.

Thanks GenAI. Are these groundbreaking insights? No! Are these ground breaking recommendations? No!

But the whole exercise took me about ten minutes. And it could provide input to an interactive Integrated Reconciliation meeting, where the IBP manager prompts GenAI with other functions to ideate solutions for gaps, risks & opportunities in the business and the market environment.

Now imagine we reference this to any IBP deck we ever created, all promotional programs we activated, all the NPD we ever launched, all the IBP decision we ever made, and cross reference it with our financial statements and even external market or competitor data. It is easy enough to upload the financial statement of a publicly listed competitor (Ah, you haven’t done that yet?) and compare it.

GenAI powered, but human-in-the-loop facilitated recommendations for IBP decisions are a no-brainer!

Mollick is right. Always invite AI to the table!!!

Disclaimer: I changed financial numbers, categories and retail names in order not to breach confidentiality.