Hey! This post is part of my Google Webmaster Hangouts Notes. I cover them regularly to save you time.
Before you get down to the answers, I want to ask you something. Will it be valuable for you to get all the notes (based on 40+ hours of video!) broken down by categories for easier navigation in the form of an eBook? Leave your replies in the comments!
If 2 URLs have the same content, Google tends to pick the shorter URL as a canonical (1:14)
Google uses URLs to identify content, so it doesn’t really matter if the URL is long or short.
But if there are 2 different URLs that have the same content, Google tends to pick the shorter URL as a canonical URL.
If you have an interstitial, make sure it doesn’t redirect to a different page (5:16)
When an interstitial redirects to a different URL, Googlebot follows that redirect and will only have this content to index.
What happens in this case is that Google thinks that all the website pages are interstitial. As a result, you basically remove your website from search.
You can use a full-page interstitial but make sure a user remains on the same URL and that Google can still see the main page content.
Don’t point to AMP cached URLs in your internal linking (13:18)
Pointing to cached URLs is a bad practice as the pages can sometimes change.
That’s ok if a linked page swaps up and shows the cached version. But the initial linking should not point to the cached URLs.
Moreover, cached URLs are often blocked in robots.txt. This is also a good reason not to use them in the website’s internal linking structure.
The number of valid DMCA complains can influence a website ranking (15:31)
If your website has many valid DMCA complaints, Google might take into account and decrease your rankings.
If your mobile site version is missing some markup compared to the desktop version, Google won’t be able to index this missing markup (16:21)
If you have some kind of markup or information only on the desktop website, Google won’t see it. The reason is mobile-first indexing: Google considers only the content of your mobile website and doesn’t look at its desktop version.
It takes Google a significant amount of time to pick up a website migration if it involved significant URL changes and redesign (17:58)
If you migrate a website and change its structure, Google needs to re-estimate the website and understand the new structure, context of the pages, internal linking, etc. All this takes a significant amount of time (even if the migration happens within the same domain).
Paradoxically, a migration to a different domain but with the same URL structure takes Google much less to pick it up as Google takes everything one-to-one and copied it over.
A good way to diagnose a website migration is to compare the period of time before and after the change (ideally, the same weekdays).
Compare the quires and URLs ranking before and after. This helps to quickly diagnose the possible issues like the old URLs not being redirected or not showing in the new timeline at all. Analyzing the quires helps to see what kind of searches bring less traffic (e.g. branded vs non-branded), are they happening across the board or applicable to only some queries.
All this helps to diagnose any issues with your migration.
It’s OK that Google can try to crawl really old pages even if they return errors (28:30)
It’s common for Google to request pages even if they’re old, even 4-5 years old. This is fine and there’s nothing you need to worry about here.
If such URLs waste craw budget, you can adjust the crawl rate setting in Google Search Console to tell Google to crawl less. In this case, Google will focus more on the URLs it cares about and if there’s no time for the old pages left, Googlebot just won’t request them.
If Google is unsure about the user intent, it can rank multiple pages from a single website (32:03)
If Google is unsure what the intent of the query is, it can rank multiple pages from a single website. For example, a user might be searching for a particular product (-> a product page) or a particular kind of product (-> a category page). So the fluctuations might happen: one day Google can show the category page, next day – the product page or both pages.
Google structured data will become even more complicated over time (34:20)
Structured data can become even more complicated in the future: there will be more types of it and lots of requirements for different search features.
Googlebot doesn’t use crawl budget for the URLs blocked in robots.txt (41:01)
If a URL is blocked by robots.txt, Googlebot doesn’t use the crawl budget for it.
A reminder: for most of the websites, the crawl budget is not an issue.
Fresh content is not always better (44:12)
Google uses different signals to figure out the date of a page. The URL is one of the signals. Other things include article markup, the visible date on the page, etc.
An important thing to note here is that for evergreen content, the date doesn’t really influence the rankings.
While ranking pages, Google looks at the current state of all signals (46:28)
Google ranks the pages based on the information it has at the time when it’s checking the current state of the pages.
So if a website A linked to a website B with a dofollow link but then added a nofollow attribute to that link, Google will take into account the nofollow link (the current vs past situation).
That’s it for today!