Amazon: Define what makes a promotional deal successful and rank product categories by their success rate.
The Deals team ran about sixty time-limited promotions in the second quarter and wants to know which product categories the program works best for before planning the next quarter. They have exported one row per deal and asked you to decide how a successful deal should be defined, apply it to every deal, and rank the categories by their success rate.
Note: for each deal, the export records the sales target set when the deal was planned, the units sold during the deal, and the item's baseline — the average number of units it sold per day in the weeks before the deal, at its list price.
amazon_promo_deals.csv has one row per deal:
deal_idcategoryproduct_namestart_dateend_date - last day of the dealstatus - ended or activelist_price - the item's regular pricedeal_price - the price during the dealunits_target - the sales target set when the deal was plannedunits_sold - units sold during the deal so farbaseline_daily_units - average units sold per day in the weeks before the dealYou can solve this in Python, SQL, or a mix of both — StrataScratch notebooks support both languages in the same notebook.
Assume this data comes directly from the deals system and hasn't been cleaned, so review it carefully before relying on the numbers. When something looks off, work out what it is and decide how to handle it; when a definition or a final decision is genuinely ambiguous, choose a reasonable option and explain why. There's often more than one valid way to compute a result, so explain the approach you take and the reasoning behind your choices.
What's assessed: whether your definition of a successful deal fits the Deals team's purpose, whether you apply it correctly and consistently to every deal, and how clearly you justify it. There is no single correct ranking: a different, well-argued definition is a valid answer.