Skip to content

Latest commit

 

History

History
53 lines (28 loc) · 2.25 KB

README.md

File metadata and controls

53 lines (28 loc) · 2.25 KB

This repo is the new home for the RGoogleAnalytics library migrated from Google Code SVN

What is it

RGoogleAnalytics is a R Wrapper around the Google Analytics API. It allows fast and easy data extraction in R so that further statistical analysis can be run on the data

Key Features

  • Provides Access to v3 of the Google Analytics Core Reporting API

  • Ability to pull more than 10,000 rows of data in batches via pagination of queries

  • Ability to mitigate the effect of Query Sampling by splitting the date-range of queries and hence extract (nearly) unsampled data

  • Ability to cache data fetched from Google

  • Supports authorization via OAuth 2.0

  • In cases where queries are sampled, the output also returns the percentage of sessions that were used for the query

Installation

To get the current development version from github:

# require(devtools)
devtools::install_github("Tatvic/RGoogleAnalytics")

Dependencies

  • httr handles the underlying OAuth2.0 Authorization flow and the API requests

  • lubridate handles the date manipulation logic underlying Query Partitioning

Background

Work on RGoogleAnalytics was started by Michael Pearmain at Google. He was supported by Nick Mihailowski (Google) and Vignesh Prajapati (Tatvic).

Tutorials and Use-cases

  • Basic tutorial to get started linking Google Analytics to an API pull
  • Under development

Important Links

  • List of Valid Dimension/Metric Combinations from the Google Analytics API Reference Guide

  • Query Feed Explorer allows you to test your queries for syntatical correctness. Once verified, the query parameters can then be copied to your R Script

  • Demo link on how to use this package to extract data from Google Analytics.