reporting_ga

Simple framework using Google Analytics API with python+pandas to do quick visualization focused on goals.

Setup

• Follow this tutorial to get creds: https://developers.google.com/analytics/devguides/reporting/core/v4/quickstart/installed-py

• Dependencies (don't know if it works with py2):

  pip3 install -r requirements.txt

• Create config.ini at the root

[Creds]
VIEW_ID: ID_GA
CLIENT_SECRETS_PATH: client_secret.json

[ConfProject]
folder: FOLDER_NAME/

Example

• Launch notebook:

jupyter notebook

• Create object:

d = g.AnalyticsData(start_date="", end_date="", dimension="", metric="", goal="", version="", filter_data='')

Name	Description	Example
start_date	Starting date	"2016-08-05"
end_date	Ending date	"yesterday"
dimension	Dimensions	"date,deviceCategory,userType"
metric	Metrics	"sessions"
goal	If you have goals	"goalXCompletions"
version	String	"dimensionXX"
filter_data	String	ga:dimension5=~^(New)

All the parameters have to follow this guide: https://developers.google.com/analytics/devguides/reporting/core/dimsmets without the "ga:" except for filter_data parameter.

• Retreive data

df = d.get_full_data()

• Graph

input:

d.ts(df)

output:

Documentation

.ts()

Name	description
df	data to pass (pandas dataframe)
filter_dim	Filter data by one of the dimension's name requested
filter_val	Filter data by one of the dimension's value requested
ylim	min and max limit of the graph
title	Title of the graph

ts example

d.ts(df, filter_dim='deviceCategory', filter_val='desktop', title='My time series graph')

.bar()

Name	Description
filter_dim	Filter data by one of the dimension's name requested
filter_val	Filter data by one of the dimension's value requested
ylim	min and max limit of the graph (tuple)
title	Title of the graph

bar example

d.bar(df, filter_dim='deviceCategory', filter_val='desktop', title='My time series graph', ylim(30,40))

.diff()

Name	Description
ref_dim	the version to compare
filter_dim	Filter data by one of the dimension's name requested
filter_val	Filter data by one of the dimension's value requested
title	Title of the graph

diff example

d.diff(df,ref_dim='Test version', filter_dim='deviceCategory', filter_val='desktop', title='My time series graph')

.cumulative()

Name	Description
filter_dim	Filter data by one of the dimension's name requested
filter_val	Filter data by one of the dimension's value requested
ylim	min and max limit of the graph
title	Title of the graph

cumulative example

d.cumulative(df, filter_dim='deviceCategory', filter_val='desktop', ylim(30,40), title='My time series graph')

.diff_cumul()

Name	Description
filter_dim	Filter data by one of the dimension's name requested
filter_val	Filter data by one of the dimension's value requested
ref_dim	the version to compare
title	Title of the graph
ylim	min and max limit of the graph

diff_cumul example

d.diff_cumul(d,filter_dim='deviceCategory', ref_dim='Test version', filter_val='desktop', ylim(30,40), title='My time series graph')

Next steps

• Add special functions for e-commerce websites.

• Add new api?

  ​