Google Analytics graphs

options(digits = 2)
## Set knitr options
knitr::opts_chunk$set(comment = "#>", collapse = TRUE, dev = "png", dpi = 150, warning = FALSE, message = FALSE)

## Set locale
if (.Platform$OS.type == "unix") {
    Sys.setenv(LANG = "en_US.UTF-8")
    Sys.setlocale("LC_ALL", "en_US.UTF-8")
} else if (.Platform$OS.type == "windows")
    Sys.setlocale("LC_ALL", "English")

## [1] "LC_CTYPE=en_US.UTF-8;LC_NUMERIC=C;LC_TIME=en_US.UTF-8;LC_COLLATE=en_US.UTF-8;LC_MONETARY=en_US.UTF-8;LC_MESSAGES=ru_RU.UTF-8;LC_PAPER=ru_RU.UTF-8;LC_NAME=C;LC_ADDRESS=C;LC_TELEPHONE=C;LC_MEASUREMENT=ru_RU.UTF-8;LC_IDENTIFICATION=C"

Prerequirements

Load R pakcages

library("RGA")
library("ggplot2")
library("scales")
library("reshape2")
library("dplyr", warn.conflicts = FALSE)
library("DT")

Authorisation

authorize()

Get Profile ID

# Profiles (profiles) list
profiles <- list_profiles()
# Site URL
site_url <- "balancer.ru"
# Get Profile ID
id <- profiles[profiles$website.url == site_url, "id"]

Heatmap plot of users activity

Setup dates range

# Get first date
start.date <- firstdate(id)
# Get the next monday
start.date <- as.Date(cut(start.date, "weeks")) + 7
# Get the last sunday
end.date <- as.Date(cut(Sys.Date(), "weeks")) - 1

Get data

ga_data <- get_ga(id, start.date = start.date, end.date = end.date,
                  metrics = "ga:pageviews", dimensions = "ga:dayOfWeek,ga:hour")

Prepare data

ga_data$day.of.week <- replace(ga_data$day.of.week, ga_data$day.of.week == 0, 7)
ga_data$day.of.week <- ordered(ga_data$day.of.week, levels = 1:7,
                         labels = c("Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"))

Draw plot

ggplot(ga_data, aes(x = day.of.week, y = hour)) +
    geom_tile(aes(fill = pageviews), colour = "white") +
    scale_fill_gradient(name = "Page views", low = "steelblue", high = "red") +
    scale_y_continuous(breaks = unique(ga_data$hour)) +
    labs(title = "Heatmap daily user activity", x = "Day of week", y = "Hour") +
    theme(axis.ticks = element_blank())

Time-Series Calendar Heatmap

Setup dates range

# Set first date
start.date <- "2012-01-01"
# Set the last date
end.date <- "2014-12-31"

Get data

ga_data <- get_ga(id, start.date = start.date, end.date = end.date,
                  metrics = "ga:sessions", dimensions = "ga:date",
                  filters = "ga:sessions > 0")

Prepare data

ga_data <- ga_data %>%
    mutate(year = as.numeric(format(date, "%Y")),
           month = as.numeric(format(date, "%m")),
           yweek = as.numeric(format(date, "%W")),
           mday = as.numeric(format(date, "%d")),
           wday = as.numeric(format(date, "%u")),
           yday = as.numeric(format(date, "%j"))) %>%
    mutate(month = ordered(month, levels = 1:12,
                          labels = c("Jan", "Feb", "Mar", "Apr", "May", "Jun", "Jul", "Aug", "Sep", "Oct", "Nov", "Dec")),
           wday = ordered(wday, levels = 1:7,
                         labels = c("Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun"))) %>%
    group_by(year, month) %>% mutate(mweek = yweek - min(yweek) + 1) %>% ungroup

Draw plot

ggplot(data = ga_data, aes(x = mweek, y = wday)) +
    geom_tile(aes(fill = sessions), colour = "white") +
    geom_text(aes(label = mday)) +
    facet_grid(year ~ month) +
    scale_fill_gradient(name = "Sessions", low = "steelblue", high = "red") +
    labs(title = "Time-Series Calendar Heatmap", x = "", y = "") +
    scale_x_continuous(breaks = NULL) +
    scale_y_discrete(limits = rev(levels(ga_data$wday)))

Year by year comparison

Setup dates range

# Set first date
start.date <- "2012-01-01"
# Set the last date
end.date <- "2014-12-31"

Get data

ga_data <- get_ga(id, start.date = start.date, end.date = end.date,
                  metrics = "ga:sessions", dimensions = "ga:date,ga:year",
                  filters = "ga:sessions > 0")

Draw plot

ggplot(data = ga_data, aes(x = as.Date(date), y = sessions)) +
    geom_line() + geom_point(size = 1.5) +
    facet_wrap(~ year, ncol = 1, scales = "free_x") +
    scale_x_date(labels = date_format("%b"), breaks = date_breaks("month")) +
    labs(title = "Time-Series years comparison", x = "")

Year by year comparison

Setup dates range

# Set first date
start.date <- "2012-01-01"
# Set the last date
end.date <- "2014-12-31"

Get data

ga_data <- get_ga(id, start.date = start.date, end.date = end.date,
                  metrics = "ga:sessions", dimensions = "ga:yearMonth",
                  filters = "ga:sessions > 0")

Prepare data

ga_data$date <- as.Date(paste0(ga_data$year.month, "01"), "%Y%m%d")

Draw plot

ggplot(data = ga_data, aes(x = ga_data$date, y = sessions)) +
    geom_line() + geom_point(size = 1.5) +
    scale_x_date(labels = date_format("%b %Y"), breaks = date_breaks("month")) +
    theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
    geom_vline(xintercept = as.numeric(as.Date(levels(cut(ga_data$date, breaks = "year")))),
               alpha = 0.5, linetype="dashed") +
    labs(title = "Time-Series years comparison", x = "")

Time series

Setup dates range

# Set first date
start.date <- "7daysAgo"
# Set the last date
end.date <- "yesterday"

Get data

ga_data <- get_ga(id, start.date = start.date, end.date = end.date,
                  metrics = "ga:users,ga:sessions,ga:pageviews", dimensions = "ga:dateHour")

Prepare data

ga_data <- melt(ga_data, id.vars = "date.hour", variable.name = "metrics")

Draw plot

ggplot(ga_data, aes(x = date.hour, y = value)) +
    geom_line(aes(colour = metrics)) +
    ylim(c(0, NA)) +
    scale_x_datetime(labels = date_format("%d %b"), breaks = date_breaks("day")) +
    labs(title = "Time series", x = "", y = "") +
    scale_color_discrete(name = "Metrics")

Time series

Setup dates range

# Set first date
start.date <- "31daysAgo"
# Set the last date
end.date <- "yesterday"

Get data

ga_data <- get_ga(id, start.date = start.date, end.date = end.date,
                  metrics = "ga:users,ga:sessions,ga:pageviews", dimensions = "ga:date")

Prepare data

ga_data <- melt(ga_data, id.vars = "date", variable.name = "metrics")

Draw plot

ggplot(ga_data, aes(x = date, y = value)) +
    geom_line(aes(colour = metrics)) +
    ylim(c(0, NA)) +
    scale_x_datetime(labels = date_format("%d %b"), breaks = date_breaks("week")) +
    labs(title = "Time series", x = "", y = "") +
    scale_color_discrete(name = "Metrics")

Trafic sources

ga_data <- get_ga(id, start.date = "2014-03-01", end.date = "2014-12-31",
                  metrics = "ga:sessions,ga:pageviews,ga:avgSessionDuration,ga:bounceRate",
                  dimensions = "ga:source,ga:medium",
                  sort = "-ga:sessions,ga:medium",
                  max.results = 20)

datatable(ga_data) %>% formatPercentage("bounce.rate")

Prepare data

ga_data <- get_ga(id, start.date = "2014-03-01", end.date = "2014-12-31",
                  metrics = "ga:sessions,ga:pageviews,ga:avgSessionDuration,ga:bounceRate",
                  dimensions = "ga:medium", sort = "-ga:sessions,ga:medium",
                  filters = "ga:medium!@(not set)")

datatable(ga_data) %>% formatRound("avg.session.duration", 0) %>% formatPercentage("bounce.rate", 2)

Google Analytics graphs

Artem Klevtsov

19 января 2015 г.

Prerequirements

Load R pakcages

Authorisation

Get Profile ID

Heatmap plot of users activity

Setup dates range

Get data

Prepare data

Draw plot

Time-Series Calendar Heatmap

Setup dates range

Get data

Prepare data

Draw plot

Year by year comparison

Setup dates range

Get data

Draw plot

Year by year comparison

Setup dates range

Get data

Prepare data

Draw plot

Time series

Setup dates range

Get data

Prepare data

Draw plot

Time series

Setup dates range

Get data

Prepare data

Draw plot

Trafic sources

Prepare data