Problem Solving › Data Analysis
La previsione di domanda ha sbagliato del 35% · Data Analysis · Advanced
The demand forecast model, reliable for two years, was 35% off last quarter: it predicted 12,000 units, 8,000 sold, leaving 200,000 euros of unsold stock. Management wants to 'fix the model' and demands explanations. That quarter saw a competitor enter and a price change. Before randomly tweaking parameters, how do you diagnose why it failed and what to do for next quarter?
Il modello di previsione della domanda, affidabile per due anni, ha sbagliato del 35% l'ultimo trimestre: ha previsto 12.000 unità, ne sono state vendute 8.000, lasciando magazzino invenduto per 200.000 euro. La direzione vuole 'aggiustare il modello' e chiede spiegazioni. Nel trimestre c'è stata l'entrata di un concorrente e un cambio prezzo. Prima di ritoccare parametri a caso, come diagnostichi perché ha fallito e cosa fare per il prossimo trimestre?
Read the case, write how you would solve it — in free text, no prompt required — and Grace's AI judges whether your solution actually works: solved or not solved, with a 0-100 score and coaching feedback that points at what is still uncovered without revealing the answer. Your first 5 evaluations each day are free.
It is a advanced-level Data Analysis case based on a real workplace situation. You read the problem, write how you would solve it in free text, and Grace's AI judges whether your solution actually works — or what is still uncovered.
No. Grace judges the substance of your reasoning, not similarity to a script. Different approaches pass, as long as they genuinely solve the problem within the constraints described in the case.
Grace's AI checks four things: whether you address the core problem (not just the symptoms), whether your plan is feasible within the constraints of the case, whether you consider risks and consequences, and whether you avoid the typical mistakes. You get a solved/not-solved verdict, a 0-100 score and coaching feedback.
Anyone who wants to train structured problem solving: managers, professionals and candidates preparing for case interviews in Data Analysis. The advanced difficulty makes it a demanding test of real-world judgment.
All 150 problem-solving cases → · Prompt training scenarios →
From our network
AGORÀ Intelligence — Enterprise AI Governance Platform
Govern AI at scale: policies, adoption and measurable results on your data. Built for boards and C-suite.
Visit agora-intelligence.com →From the Agora Intelligence blog
📱 Download the Android app (beta) iOS coming soon
Say what you mean. Get what you need.
Grace Certified — the AI coach that trains and certifies your prompt engineering — by Agora Intelligence.