Google Search Console Tips

An ‘other error’ occurs in Google Search Console live tests when it doesn’t get around to fetching the URL in that particular test, for example a resource file. When crawling a page for indexing purposes Google will spend more time fetching and caching the embedded URLs in order to be able to render them properly.

Google may sometimes display a result in the right sidebar, which is a mix between a knowledge panel and featured snippet. John expects this page will be seen by Google Search Console as position 10 or 11, because it is classed as being in the second column of search results.

GSC uses the top-most position of a URL to show as the average in the Performance reports. This hasn’t changed despite the update to how featured snippets are displayed, so if your page is shown in a featured snippet, the position shown will remain the same. However, if you no longer have the featured snippet then the position of the traditional search result will be used instead.

Unnatural searches like site: queries can be removed from Google Search Console performance reports, and often wouldn’t appear there anyway due to the small amounts of people that use them.

The average position in Google Search Console is calculated on both a per-URL and per-query basis. This data can seem inconsistent in some cases, for example, URLs in sitelinks within the top result will be counted as position 1 from a query perspective, but not from a URL perspective.

Search Console tracks things differently than Google Analytics does, so John expects there will be some difference between the metric counts in both. This is because GSC tracks what users click on when visiting a website, while Analytics tracks site visits differently, in particular when redirects are used.

The aggregate reports in search console, for example mobile usability, AMP and rich results, focus on a sample of URLs from a site. In comparison, the coverage report includes all of the indexed URLs and this means it can be difficult to compare the total numbers seen across all of the reports. E.g. the coverage report may show 4,000 valid indexed pages, while in the mobile usability report the total may only be 2,000 valid pages, as this is the sample size taken.

John recommends two approaches for understanding a site’s crawl budget, one of these is looking at the speed at which Google was able to download individual pages, which can be found in GSC. If this is high it may indicate that Google have crawled as much as they could, but may have missed some pages. The other aspect is server errors, as they can impact the ability to crawl the site. Reviewing server logs will allow you to identify if server errors are occuring.

In the Google Search Console crawl stats report, all of the access calls which go through the infrastructure that Googlebot uses will be included. This includes Googlebot crawling and rendering as well as robots.txt and sitemap access. John informed that while the crawl stats are useful, it can be difficult to compare them with log files.

Search Console speed reports are based on Chrome User Experience data, which is real-world data from users. Therefore, for smaller sites, Google may not have enough data to make a meaningful judgment on speed metrics.