So today I discovered that the pages from my website are listed with an unexpected date value. I specify the schema.org properties dateCreated and dateModified for most of my content pages. I'd expect that search results show me when a page was last updated, to get a sense of the currency of the page. But it's showing the date of first publishing which may be years ago. That's a bit unsatisfying but I don't want to misuse the metadata because Google probably reads it wrong.

Some search terms for you to try it out: "gitrevisiontool"; "easyxml"; "multiselecttreeview" (look for the results on dev.unclassified.de; the human- and machine-readable dates come at the end of the page)

Does anybody know more about what's wrong here? Or does it work as designed? (What a stupid design that would be.)


Okay. I have been looking at this for a while and there seems to be little information on how Google gets date/time stamps except for:

The created date is the date that Google fetches the resource known as the inception date.

It is possible to make a HEAD request and get the last modified date of an index.html, however, some CMS or other systems may block this. I seem to remember Google making HEAD requests in the past.

It appears that Google will examine a recognizable date format that exists just below the first H1 tag though have not tested this.

If you can supply an example of the markup you use, perhaps there is an answer. It may be that the markup is just informational. It may be that this date is not integrated into the index date(s) since it is user supplied.

If you use a sitemap, that may be a better solution for you.

  • I do use a sitemap but it doesn't contain any dates. Also I've read elsewhere that dates (<lastmod>?) are ignored there.
    – ygoe
    Mar 10 '14 at 9:44
  • You may be right. The only thing I have heard that influences Google SERP dates is putting a recognizable date just below the first h1 tag. I read something about this years ago so seeing it again in a different format, makes me think there is truth to that statement though I have not tried it. I might implement this idea on one of my sites to see if it works! ;-) As well, the code I wrote for my site does collect the update date for my content, but does not do anything with it in the HTTP header. I will have to look at how I can change some code.
    – closetnoc
    Mar 10 '14 at 16:00
  • I'm now going to try out reversing the order of the two dates. Updated first, created last. Maybe it'll improve the situation sometime.
    – ygoe
    Mar 10 '14 at 21:13

Your microdata seems to be correct, at least the ones I tested.

According to About microdata and Rich snippets - Reviews. Google doesn't recognize the Article element. That has at least two implications, first is that it may not use all of the information provided on their rich snippets, and second that it may support/recognise it in the future.

Since your code is valid, at least on the microdata side, and your use is correct, not stuffing the page or hiding content, I'd say that the information you are providing is still useful. Even if it's not presented on the SERPs for now.

Also, remember that Google considers valuable content that has been on Internet for some time, specially if it has visits and links back from other sites.

One recommendation, is to move the statistic data that you provide to the top, at least the modified data, that way, regardless og what google shows, the user will see immediately, that the content is still valid and relevant. That may, or may not, help with the rich snippets part.

As a side comment, you may consider adding/improving the author/publisher microdata.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.