Check 0 pageviews by comparing sitemap.XML URLs with Google Analytics visits.

compareXMLtoGA.R

library(googleAnalyticsR)
library(xml2)
library(dplyr)

ga_auth()

## date range of URLs to test
dates <- c(Sys.Date() - 30, Sys.Date())

##GA View ID
id <- 11111111

## function to get sitemap URLs
get_sitemap <- function(sitemap, field = "loc"){
  
  sm <- as_list(read_xml(sitemap))
  
  out <- try(Reduce(rbind, 
                vapply(sm, function(x) Reduce(rbind, x[[field]]), character(1))
  ))
  if(inherits(out, "try-error")){
    message("Problem with sitemap:", sitemap)

    return(NULL)
  }
  as.vector(out)
}

## make google SEO filter
google_seo <- filter_clause_ga4(
                              list(
                                dim_filter("source", "EXACT", "google"),
                                dim_filter("medium", "EXACT", "organic") 
                              ),
                              operator = "AND")
## get the pages
pages <- google_analytics_4(id,
                            date_range = dates,
                            dimensions = "pagePath",
                            metrics = c("pageviews","totalEvents"),
                            dim_filters = google_seo,
                            max = -1,
                            anti_sample = TRUE)



## get the sitemap index file
url_si <- "http://www.example.com/sitemap.xml"
sitemap_index <- get_sitemap(url_si)

## get all the sitemaps (maybe you only need the call above if you have no sitemap index)
many_sitemaps <- lapply(sitemap_index, get_sitemap)

## all the urls in all the sitemaps
all_urls <- Reduce(c, many_sitemaps)

## Compare and get the URLs that are in XML but not in Google Analytics

## dplyr transformations
sitemap_urls <- as.tbl(as.data.frame(all_urls, stringsAsFactors = FALSE))
sitemap_urls <- sitemap_urls %>% mutate(path = paste0("/",urltools::path(all_urls)))
sitemap_not_in_ga <- anti_join(sitemap_urls, pages, by = c(path = "pagePath"))

## write out to CSV
write.csv(sitemap_not_in_ga, file = "./data/sitemap_urls_not_in_ga.csv", row.names = FALSE)

Just saw this, sorry. It won't work in Wordpress, which is PHP. This is a script to run in R, locally on your laptop, perhaps.