AI Assistant for Business Analytics
Budget: ₹600 – ₹1,500 INR
AI Solution Design Round (Self-Paced)
________________________________________
Context
You are part of the Data & AI team at a Consumer Goods (FMCG) organization working on the Beverages category.
Business users across functions frequently request:
• Promotional performance summaries
• Inventory movement insights
• Regional sales comparisons
• Product-level campaign impact
Currently, these requests are handled through manual dashboarding and ad-hoc analysis by data teams, leading to:
• Delayed decision-making
• Repetitive analysis
• Dependency on analysts
Leadership wants to explore whether an AI-powered assistant can help business users obtain such insights conversationally.
________________________________________
Your Task
Design a solution that enables business users to:
Ask natural language questions about weekly promotion effectiveness and receive structured analytical responses.
Example:
• “Did last month’s campaign improve sales in the South region?”
• “Which categories saw inventory reduction during promotions?”
________________________________________
What You Need To Submit (Max 3–4 Pages)
________________________________________
Section A: Problem Decomposition
1. What kinds of user questions should the assistant support?
2. What structured analytical outputs should the assistant generate?
________________________________________
Section B: AI Solution Approach
Describe what methods you will use
Explain:
• Why your chosen approach is suitable
Define an Agentic construct
________________________________________
Section C: Data Interaction Strategy
How would your AI system:
• Access analytical datasets?
• Perform aggregations?
________________________________________
Section D: Limitations & Risks
Mention:
• One risk related to:
o Incorrect query generation
o Hallucinated responses
o Data misinterpretation
How would you mitigate this?
________________________________________
Section E: Design Trade-Off
Describe:
One place where you chose:
• Simplicity over flexibility
or
• Accuracy over latency
________________________________________
Estimated effort: 60–90 minutes
________________________________________
________________________________________
Context
You are part of the Data & AI team at a Consumer Goods (FMCG) organization working on the Beverages category.
Business users across functions frequently request:
• Promotional performance summaries
• Inventory movement insights
• Regional sales comparisons
• Product-level campaign impact
Currently, these requests are handled through manual dashboarding and ad-hoc analysis by data teams, leading to:
• Delayed decision-making
• Repetitive analysis
• Dependency on analysts
Leadership wants to explore whether an AI-powered assistant can help business users obtain such insights conversationally.
________________________________________
Your Task
Design a solution that enables business users to:
Ask natural language questions about weekly promotion effectiveness and receive structured analytical responses.
Example:
• “Did last month’s campaign improve sales in the South region?”
• “Which categories saw inventory reduction during promotions?”
________________________________________
What You Need To Submit (Max 3–4 Pages)
________________________________________
Section A: Problem Decomposition
1. What kinds of user questions should the assistant support?
2. What structured analytical outputs should the assistant generate?
________________________________________
Section B: AI Solution Approach
Describe what methods you will use
Explain:
• Why your chosen approach is suitable
Define an Agentic construct
________________________________________
Section C: Data Interaction Strategy
How would your AI system:
• Access analytical datasets?
• Perform aggregations?
________________________________________
Section D: Limitations & Risks
Mention:
• One risk related to:
o Incorrect query generation
o Hallucinated responses
o Data misinterpretation
How would you mitigate this?
________________________________________
Section E: Design Trade-Off
Describe:
One place where you chose:
• Simplicity over flexibility
or
• Accuracy over latency
________________________________________
Estimated effort: 60–90 minutes
________________________________________