Notes from the Google Webmaster Hangout on the 22nd of September, 2017.
Location Name High in URL Path Raises Suspicion About Doorway Pages
Location names high up in the URL path raises suspicions that there are going to be a lot of doorway pages on the site, where the full content is being duplicated across a large number of cities. The web spam team might take action on these if there is nothing unique across these city pages.
Move All Pages at Once For Migrations
John recommends moving all pages over at once for migrations. If the migration is done in steps, it tends to take a lot longer and there’s more room for error with Google struggling to understand if you’re migrating part of the site or the whole site.
Google Fixing Discrepancy Between Mobile-Friendly Results & Message in Search
Google are working to improve discrepancies between mobile-friendly test and messages in search results regarding mobile friendliness. When Google can see a webmaster is logged in and looking at one of their own search results, they will add a message if they think one of the pages isn’t mobile friendly. The problem comes when the algorithms haven’t reprocessed that URL in a while and they aren’t certain that page is mobile friendly, in which case they will flag it in the search results. Go by the results of the mobile-friendly test.
Knowledge Panel Won’t Change With Mobile-first Indexing
Elements of search results, like the knowledge panel, won’t change with Mobile-first indexing as ideally Google will have the same information just from mobile pages instead.
Google Has No Documented Character Limit For Mobile & Desktop For Titles
Google has no documented recommended title character length limits for mobile and desktop. John recommends working out what works best for your website. Sometimes really short titles work better than long ones.
Hash Symbols Are Ignored by Google
When Crawling and indexing pages Google ignores everything that comes after a hash symbol. Using these is fine as long as they don’t lead to unique content not available within the normal URL as this would likely be missed by Google.
Use Jargon if That’s What Users Search For
If your users search using specific jargon (e.g. Singlish) then it makes sense to use that on your site, as Google should be able to pick that up and recognise these words. Match what users are searching for.
Google Found Examples of Searchers Using One Language But Expecting Content in Another
Google has seen quirks where people search in one language but expect to see another e.g. People searching using English characters but expect content in Hindi.
Low Proportion of Indexed Pages Points to Technical Issue
If a site has a low proportion of indexed pages, this usually points to a technical issue than a quality issue. Compare the sitemap index counts and index status report for differences. Try splitting up sitemap file, checking indexed pages using info: query, that rel canonicals match those in sitemap file, hreflang and internal linking. Also, uppercase, lowercase, trailing slashes all matter. Then check crawl stats to get idea of crawl rate and if it’s reasonable.
Not One Best Option For Dealing With Multi Regional & Lingual Sites
There is no one best option to dealing with multi-regional, mutli-lingual sites but John recommends listening to Aleyda Solis’ Edge of the Web Podcast for excellent coverage of the international issues.
Include Shared Content Block For Pages That Vary Dependent on Location
If content served varies dependent on location then John recommends having a shared content block across all variations as Google primarily crawls from IP addresses geo-located to San Francisco.
Fine to Noindex Paginated Pages But Ensure Google Can Access Linked Content on These Pages
It is fine to noindex paginated pages (e.g. after page 2) but you need to make sure that Google can still get to the content linked on these pages via other routes in your internal linking structure (e.g. related products/content, category pages).
Parameters Are Hint to Google Not to Crawl as Frequently
Parameters in Search Console aren’t a directive, more of a hint to Google not to crawl these as frequently. Blocking URLs in Robots.txt is a strong directive to say Google should not be looking at these pages e.g. if Google’s crawls are crashing your site.