Page_funnel_visits.py

import pandas as pd
checkout = pd.read_csv('checkout.csv',
                       parse_dates=[1])
checkout.drop_duplicates(['user_id'],inplace=True)
checkout.to_csv('checkout-dedupe.csv',index=False)

visits = pd.read_csv('visits.csv',
                     parse_dates=[1])
cart = pd.read_csv('cart.csv',
                   parse_dates=[1])
                   
checkout = pd.read_csv('checkout.csv',
                       parse_dates=[1])
purchase = pd.read_csv('purchase.csv',
                       parse_dates=[1])

print(visits.head(5))
print(cart.head(5))
print(checkout.head(5))
print(purchase.head(5))

visits_cart = visits.merge(cart, how = 'left')

visits_cart.head()

total_visits = len(visits_cart)

print(total_visits)

null_cart_times = len(visits_cart[visits_cart.cart_time.isnull()])

print(null_cart_times)

visited_not_cart = float(null_cart_times) / float(total_visits)

print(visited_not_cart)

cart_checkout = cart.merge(checkout, how = 'left')

print(cart_checkout.head())

null_checkout_times = len(cart_checkout[cart_checkout.checkout_time.isnull()])

cart_not_checkout = float(null_checkout_times) / float(len(cart))

print("Cart but not checkout:", cart_not_checkout)

all_data = visits_cart\
.merge(cart_checkout, how = 'left')\
.merge(purchase, how = 'left')

print(all_data.head(5))

reached_checkout = all_data[~all_data.checkout_time.isnull()]

checkout_not_purchase = all_data[(all_data.purchase_time.isnull()) & (~all_data.checkout_time.isnull())]

checkout_not_purchase_percent = float(len(checkout_not_purchase)) / float(len(reached_checkout))

print("% of users who got to checkout but did not purchase:",checkout_not_purchase_percent)

print("{} percent of users who visited the page did not add a t-shirt to their cart".format(round(visited_not_cart*100, 2)))
print("{} percent of users who added a t-shirt to their cart did not checkout".format(round(cart_not_checkout*100, 2)))
print("{} percent of users who made it to checkout  did not purchase a shirt".format(round( checkout_not_purchase_percent*100, 2)))


all_data['time_to_purchase'] = \
    all_data.purchase_time - \
    all_data.visit_time

print(all_data.time_to_purchase)

print(all_data.time_to_purchase.mean())