🍕 Plato's Pizza Sales

You are working as a 📈 Data Analyst 📈 at Plato's New York Style Pizza Kitchen, the most 🔥 lit 🔥 pizza joint in Brooklyn. Sales are slumping, and while we've been collecting transactional data, your boss, Plato 🎓, is a data boomer 🧙‍♂️. Plato needs you to grab a bigger slice of the market and find a recipe for success in the data.

SELECT * FROM 'platos_pizza.csv'

Well, this is sample data that Deepnote gives you to work with, as soon as you set up an account. I'm going to assume a few things about these columns and this data, and try to answer the following questions. (These questions also come with the data.)

What are the best and worst selling pizzas?

pizza_dataframe.column09.value_counts()

It looks like Classic pizzas sell the most, and Veggies sell the least.

Average total pizzas per order?

pizza_dataframe.column01.value_counts().head()

It looks like column01 has the Order ID, and the dataframe has one record per order item. Grouping by order ID and averaging the number of rows should be good enough.

(pizza_dataframe.groupby("column01") # grouping by Order ID .column00 # selecting unique identifier for order item in the dataframe .count() # taking count of items in each order .mean() # taking the mean to get avg num of items per order )

There are just over 2 pizzas on average per order. This, however, does not distinguish between pizza sizes (column08), which might be interesting to look at.

What value of the largest order?

select column03, column06, column07 from 'platos_pizza.csv' where column03>1

So, it seems column07 has the total cost or value for each item in that order. We can sum up column07 to get the total sales for each order.

(pizza_dataframe.groupby("column01") .agg({"column07":'sum'}) #.sort_values("column07", ascending=False) # see all orders by value (desc) #.idxmax() # get the Order ID for biggest order .max() # get the value for biggest order )

It looks like Order 144 was the largest order by value (order value of 238.45).

What percentage sold pizzas had 🍍pineapple🍍 on them?

"column10" seems most appropriate for this. It seems to list the ingredients in each pizza, so we can check how many of them contain pineapples.

pizza_dataframe[ (pizza_dataframe.column10 # column10 has ingredients in the pizza .str.lower().str.contains('pineapple') ) ].column00.count()

pizza_dataframe.column00.count() # also equal to # pizza_dataframe.column00.nunique()

f"{48 * 100 / 500}% of sold pizzas have pineapple on them."

That's 48 more pizzas sold than ideal, which have pineapples on them.

Anyway, that's about it. Deepnote seems cool :)

Thanks for checking this out. Please share any feedback you might have, I'd be glad to hear from you.