import pandas as pd
import matplotlib.pyplot as plt
from functions import return_ga_data, save_df_to_excel, clean_columns, VIEW_ID
# Get data from Analytics
df = return_ga_data(start_date='30daysAgo',
end_date='yesterday',
view_id=VIEW_ID,
metrics=[{
'expression': 'ga:sessionsPerUser'
}],
dimensions=[{
'name': 'ga:channelGrouping'
}])
# Creat pandas dataframe
df = df[['ga:channelGrouping', 'ga:sessionsPerUser']]
# rename columns
clean_columns(df)
# Save to csv
save_df_to_excel(df, 'spu.csv')
# Display chart
df.head(10).sort_values('sessionsPerUser',
ascending=False).plot(kind='bar',
x='channelGrouping',
y='sessionsPerUser')
plt.show()
# run in loop dates
# start_date = sys.argv[1]
# end_date = start_date
# base dims and all metrics
pages = return_ga_data(
start_date = start_date,
end_date = end_date,
view_id = VIEW_ID,
metrics = [
{'expression': 'ga:pageviews'},
{'expression': 'ga:uniquePageviews'},
{'expression': 'ga:timeOnPage'}
],
dimensions = [
{'name': 'ga:dimension1'}, # session ID
{'name': 'ga:dimension3'}, # timestamp
# page dimensions
{'name': 'ga:hostname'},
{'name': 'ga:pagePath'}
],
group_by = ['ga:dimension1', 'ga:dimension3', 'ga:hostname', 'ga:pagePath'],
dimensionFilterClauses = [],
segments=[]
)
# for console logs
print('start preprocessing pageviews data')
pages.fillna('na', inplace = True)
https://developers.google.com/analytics/devguides/reporting/core/dimsmets#cats=user,session,traffic_sources,adwords,goal_conversions,platform_or_device,geo_network,system,social_activities,page_tracking,content_grouping,internal_search,site_speed,app_tracking,event_tracking,ecommerce,social_interactions,user_timings,exceptions,content_experiments,custom_variables_or_columns,time,doubleclick_campaign_manager,audience,adsense,ad_exchange,doubleclick_for_publishers,doubleclick_for_publishers_backfill,lifetime_value_and_cohorts,channel_grouping,related_products,doubleclick_bid_manager,doubleclick_search
"""
from functions import return_ga_data, clean_df_columns, format_all_dates
import constants
#
#
#
#
# Do Stuff
#
#
#
#
sessions_df = return_ga_data(start_date=constants.START_DATE,
end_date=constants.END_DATE,
view_id=constants.VIEW_ID,
metrics=constants.SESSIONS_METRICS,
dimensions=constants.SESSIONS_DIMENSIONS,
segments=constants.SEGMENTS,
split_dates=True,
group_by=[])
# Rename columns
sessions_df = clean_df_columns(sessions_df)
if 'date' in sessions_df:
sessions_df['date'] = format_all_dates(sessions_df.date)
from functions import return_ga_data, save_df_to_excel
### Returns sessions and tranactions by channel for a specific Analytics View Id
# be sure to add correct view ID
df = return_ga_data(start_date='7daysAgo',
end_date='yesterday',
view_id='ADD VIEW ID',
metrics=[{
'expression': 'ga:sessions'
}, {
'expression': 'ga:transactions'
}],
dimensions=[{
'name': 'ga:channelGrouping'
}])
# generate dataframe
df = df[[
'ga:channelGrouping', 'ga:sessions', 'ga:pageValue', 'ga:transactions'
]]
#cleans up column names
old_columns = list(df.columns)
for old in old_columns:
df.rename(columns={old: old[3:]}, inplace=True)
# print out dataframe
print(df.sort_values('sessions', ascending=False))
df_new_users = return_ga_data(
start_date=query_start_date,
end_date=query_end_date,
view_id=ga_view_id,
metrics=[
{
'expression': 'ga:goal1Completions'
},
],
dimensions=[
{
'name': 'ga:isoYear'
},
{
'name': 'ga:isoWeek'
},
],
split_dates=False,
dimensionFilterClauses=[{
'operator':
'OR',
'filters': [{
'dimensionName': 'ga:userType',
'not': False,
'expressions': ['new visitor'],
'caseSensitive': False
}],
}],
)
yesterday = datetime.strftime(datetime.now() - timedelta(1), '%Y-%m-%d')
start_date = 'YYYY-MM-DD'
start_date = pd.to_datetime(start_date)
temp_start_date = start_date
df = pd.DataFrame()
end_date = yesterday
step = 20 # This is a relatively arbitrary value to try to reduce the incidence of API errors
end_date = pd.to_datetime(end_date)
while temp_start_date <= end_date - timedelta(step):
for n in range(0, 5):
try:
df = df.append(return_ga_data(
start_date= datetime.strftime(temp_start_date, '%Y-%m-%d'),
end_date = datetime.strftime(temp_start_date + timedelta(step), '%Y-%m-%d'),
view_id=<YOUR_GA_VIEW_ID>,
metrics=[
{'expression': 'ga:transactionRevenue'},
{'expression': 'ga:transactions'},
{'expression': 'ga:transactionTax'}
],
dimensions=[
{'name': 'ga:transactionId'},
{'name': 'ga:dateHourMinute'},
{'name': 'ga:channelGrouping'},
{'name': 'ga:source'},
{'name': 'ga:medium'},
{'name': 'ga:campaign'},
{'name': 'ga:keyword'},
{'name': 'ga:deviceCategory'},
from functions import return_ga_data, save_df_to_excel
df = return_ga_data(start_date='2017-09-13',
end_date='2017-09-21',
view_id='100555616',
metrics=[
{
'expression': 'ga:sessions'
},
],
dimensions=[
{
'name': 'ga:source'
},
],
split_dates=True,
group_by=['ga:source'],
dimensionFilterClauses=[{
'operator':
'OR',
'filters': [{
'dimensionName': 'ga:userType',
'not': False,
'expressions': ['new visitor'],
'caseSensitive': False
}],
}],
segments=[])
print(df)
sessions1 = return_ga_data(
start_date=start_date,
end_date=end_date,
view_id=VIEW_ID,
metrics=[{
'expression': 'ga:sessions'
}, {
'expression': 'ga:bounces'
}, {
'expression': 'ga:sessionDuration'
}, {
'expression': 'ga:pageViews'
}, {
'expression': 'ga:goal1completions'
}, {
'expression': 'ga:goal2completions'
}, {
'expression': 'ga:transactions'
}, {
'expression': 'ga:transactionRevenue'
}],
dimensions=[
{
'name': 'ga:dimension1'
}, # session ID
{
'name': 'ga:dimension2'
}, # client ID
{
'name': 'ga:date'
},
{
'name': 'ga:hour'
},
{
'name': 'ga:minute'
},
{
'name': 'ga:landingPagePath'
},
{
'name': 'ga:deviceCategory'
}
],
group_by=[
'ga:dimension1', 'ga:dimension2', 'ga:date', 'ga:hour', 'ga:minute',
'ga:landingPagePath', 'ga:deviceCategory'
],
dimensionFilterClauses=[],
segments=[])
ecom = return_ga_data(
start_date=start_date,
end_date=end_date,
view_id=VIEW_ID,
metrics=[
{
'expression': 'ga:itemRevenue'
},
{
'expression': 'ga:itemQuantity'
},
{
'expression': 'ga:productListViews'
},
{
'expression': 'ga:productDetailViews'
},
{
'expression': 'ga:productAddsToCart'
},
{
'expression': 'ga:productRemovesFromCart'
},
{
'expression': 'ga:productCheckouts'
},
{
'expression': 'ga:uniquePurchases'
},
],
dimensions=[
{
'name': 'ga:dimension1'
}, # session ID
{
'name': 'ga:dimension3'
}, # timestamp
{
'name': 'ga:productName'
},
# custom dimensions
{
'name': 'ga:dimension11'
}, #bundle
{
'name': 'ga:dimension15'
}, # shipping method
{
'name': 'ga:dimension17'
}, # fullfillment type
{
'name': 'ga:dimension18'
} # add to cart context
],
group_by=[
'ga:dimension1', 'ga:dimension3', 'ga:productName', 'ga:dimension11',
'ga:dimension15', 'ga:dimension17', 'ga:dimension18'
],
dimensionFilterClauses=[],
segments=[])
events1 = return_ga_data(
start_date=start_date,
end_date=end_date,
view_id=VIEW_ID,
metrics=[{
'expression': 'ga:totalEvents'
}, {
'expression': 'ga:uniqueEvents'
}, {
'expression': 'ga:eventValue'
}, {
'expression': 'ga:goal1completions'
}, {
'expression': 'ga:goal2completions'
}, {
'expression': 'ga:transactions'
}, {
'expression': 'ga:transactionRevenue'
}],
dimensions=[
{
'name': 'ga:dimension1'
}, # session ID
{
'name': 'ga:dimension3'
}, # timestamp
# event parameters
{
'name': 'ga:eventCategory'
},
{
'name': 'ga:eventAction'
},
{
'name': 'ga:eventLabel'
},
{
'name': 'ga:dimension4'
}, # account id
{
'name': 'ga:dimension5'
} # plan
],
group_by=[
'ga:dimension1', 'ga:dimension3', 'ga:eventCategory', 'ga:eventAction',
'ga:eventLabel', 'ga:dimension4', 'ga:dimension5'
],
dimensionFilterClauses=[],
segments=[])
transactions = return_ga_data(
start_date=start_date,
end_date=end_date,
view_id=VIEW_ID,
metrics=[{
'expression': 'ga:transactionRevenue'
}, {
'expression': 'ga:transactionTax'
}, {
'expression': 'ga:transactionShipping'
}, {
'expression': 'ga:itemQuantity'
}, {
'expression': 'ga:totalRefunds'
}, {
'expression': 'ga:refundAmount'
}],
dimensions=[
{
'name': 'ga:dimension1'
}, # session ID
{
'name': 'ga:dimension3'
}, # timestamp
{
'name': 'ga:dimension4'
}, # account id
{
'name': 'ga:transactionId'
}
],
group_by=[
'ga:dimension1', 'ga:dimension3', 'ga:dimension4', 'ga:transactionId'
],
dimensionFilterClauses=[],
segments=[])