I agree that something is wrong here.
Say 20 different people ordered a product 1 time each (20 distinct orders, 20 distinct people). This would mean 100% first-time orders.
The number of people who made orders can't be greater than the number of orders (20 people, 30 orders for example). So the only other option is more orders than people (say, 20 orders, 5 people), and this could mean one of two pretty opposite cases:
- 1 person ordered 16 times and the other 4 only once each (few people reordering (1 of 5))
- each person ordered 4 times (many people people reordering (5 of 5))
Both of these cases would have the same "reorder_ratio", yet they're opposites based on my understanding of the ratio. So I'm missing something or the exercise is wrong.