I just saw a post on linkedin claiming that the ranking starts to drop when the publication date appears because Google treats the blog as outdated and doesn’t pick up the modification date.
Three months ago, I tested this theory. The website I was working on only displayed the publication date, with no place for a modification date on the page.
Curious about the impact, I updated the schema to include a modification date to see if Google would recognize and display it.
Now, after three months, Google still hasn’t shown the modification date in the SERP. It continues to display only the publication date.
This made me dig deeper into how Google picks dates.
Unfortunately, Google’s current documentation provides no clear answers—just vague “best practices” and the usual “Google does not guarantee” disclaimer. Typical Google!
Luckily, I stumbled across two older resources that explained how Google used to pick dates.
One was from Google’s own documentation, which they’ve since removed. According to these sources, Google picks dates in the following order:
1-URL (e.g., /2024/11/20/article-title)
2-Title Tag (e.g., Updated on 2024-11-20)
3-Body of the Document (e.g., a visible “Last Updated” date near the content header)
4-Meta Tags (e.g., <meta name=”last-modified” content=”2024-11-20T10:00:00+00:00″>)
5-HTTP Response Header (e.g., the Last-Modified date returned by the server)
From my research, I learned a few important things:
Google may ignore dates embedded in tags in the HTML, so be careful with how you structure visible dates.
Google checks the server’s modified date. If there’s a conflict between the server response, structured data, or visible page content, it can send confusing signals to Google.
Google caches dates to save bandwidth. It sends conditional requests using the If-Modified-Since header:
If the server responds with 304 Not Modified, Google skips re-downloading the resource.
If the server responds with 200 OK, Googlebot fetches the updated resource and refreshes its cached data.