vignettes/googleAnalyticsR-table-filters.md

In jdeboer/ganalytics: Interact with 'Google Analytics'

title: "Advanced Filters" author: "Johann de Boer" date: "2018-09-09" output: html_vignette: keep_md: true md_document: variant: markdown_github vignette: > %\VignetteIndexEntry{ADVANCED FILTERS} %\VignetteEngine{knitr::knitr_notangle} %\VignetteEncoding{UTF-8}

ganalytics provides functions that makes it easy to define filters using natural R language operators. This example shows how to use ganalytics to define dimension or metric filters that can be used by the googleAnalyticsR package. The current development version of googleAnalyticsR supports filters defined with ganalytics.

Setup/Config

Once installed, load these packages. Please refer to the googleAnalyticsR package documentation on configuration steps you may need to complete in order to use the Google Analytics APIs.

library(googleAnalyticsR)
library(ganalytics)
library(dplyr)
library(tidyr)
library(ggplot2)
library(purrr)
library(knitr)

ga_auth(file.path("~", "ga.oauth"))

## Token cache file: ~/ga.oauth

view_id <- "117987738"
start_date <- "2018-05-01"
end_date <- "2018-06-30"

Pull the Data

In this example, we'll define the following filters: Device category is desktop or tablet - a dimension filter using an OR condition. New visitors using either a desktop or tablet device - a dimension filter involving both an AND and an OR condition. * At least one goal completion or transaction - a metric filter using an OR condition.

The above list of filters will be defined using ganalytics expressions as follows:

# Device category is desktop or tablet - a dimension filter using an OR condition.
desktop_or_mobile <- Expr(~deviceCategory == "desktop") | Expr(~deviceCategory == "tablet")

# New visitors using either a desktop or tablet device - a dimension filter involving both an AND and an OR condition.
new_desktop_and_mobile_visitors <- Expr(~userType == "new") & desktop_or_mobile

# At least one goal completion or transaction - a metric filter using an OR condition.
at_least_one_conversion <- Expr(~goalCompletionsAll > 0) | Expr(~transactions > 0)

We can now use googleAnalyticsR to

results <- google_analytics(
  viewId = view_id,
  date_range = c(start_date, end_date),
  metrics = c("users", "sessions", "goalCompletionsAll", "transactions"),
  dimensions = c("deviceCategory", "userType"),
  dim_filters = new_desktop_and_mobile_visitors,
  met_filters = at_least_one_conversion
)

kable(results)

|deviceCategory |userType | users| sessions| goalCompletionsAll| transactions| |:--------------|:-----------|-----:|--------:|------------------:|------------:| |desktop |New Visitor | 962| 933| 777| 0| |tablet |New Visitor | 39| 39| 38| 0|

jdeboer/ganalytics documentation built on May 18, 2019, 11:30 p.m.