SALURBAL Google Analytics Ingestions

This notebook ingests google analytics data for our datawarehouse we will leverage the googleAnalyticsR package.

Properties

Lets get a list of all of our google analytics accounts - focusing on SALURBAL

## All SALURBAL Properties
salurbal_ga = ga_account_list("ga4") %>% 
  filter(str_detect(property_name, "salurbal"))

## Portal Property
ga_portal = salurbal_ga %>% 
  filter(property_name == 'salurbal-data-portal')
property_id = ga_portal$propertyId

salurbal_ga

Here is a table of available dimensions and metrics available for us to analyze.

df_metrics_available = ga_meta(
  version = c("universal", "data"),
  propertyId = property_id,
  cached = TRUE,
  no_api = FALSE
) %>% 
  as_tibble()%>% 
  filter(status != "DEPRECATED")

df_metrics_available %>% 
  select(type, name = uiName, description) %>% 
  arrange(desc(type), name) %>% 
  reactable(searchable = TRUE)

Portal

Lets get some basic monthly information about SALURABL portal. Lets see what is available for this property.

Trends

Download

Let’s just get basic metrics.

if (!file.exists('metadata_cache/df_daily.parquet')){
  df_daily =  ga_data(
    property_id,
    metrics = c(
      "sessions", 
      "newUsers",
      "activeUsers", 
      "screenPageViews",
      "bounceRate",
      "averageSessionDuration",
      "eventsPerSession"
    ),
    dimensions = c('date'),
    date_range = c(start_date, end_date),
    limit = -1
  ) %>% 
    arrange(desc(date))
  
  df_daily %>% 
    clean_names()%>% 
    write_parquet("metadata_cache/df_daily.parquet")
} else {
  df_daily = read_parquet("metadata_cache/df_daily.parquet")
}

df_daily %>% 
  reactable()

Metrics

New users by Month

This is the one metric we have on the usage dashboard. Let’s summarize new users by month.

df_monthly_new_users = df_daily %>% 
  select(date, new_users ) %>% 
  mutate(year = year(date),
         month = month(date),
         year_month_str =  glue::glue("{year}-{month}")) %>% 
  group_by(year_month_str) %>%
  summarise(new_users = sum(new_users)) %>%
  ungroup() %>% 
  mutate(date = ymd(glue::glue("{year_month_str}-01")))  %>% 
  ## format date as MM dd, YYYY
  mutate(Name = format(date, "%B %d, %Y")) %>% 
  arrange(desc(date))%>% 
  select(Name, `New Users` = new_users, Date =  date)


df_monthly_new_users %>% 
  write_csv("metadata_cache/df_monthly_new_users.csv")
df_monthly_new_users %>% 
  jsonlite::write_json("metadata_cache/df_monthly_new_users.json", auto_unbox = TRUE, pretty = T)

Let’s preview this

df_monthly_new_users %>% 
  reactable()