| 7:06 am on Jul 7, 2008 (gmt 0)|
I'm currently working a clean-up for a site, and I'm finding that the duplicate title information is both accurate and useful. It's a bit of a tangle in this case because the site is in a frameset, but the titles that WMT lists and the URLs shown are accurate for me.
|when i check the content on many identified pages they have the same amount of unique content |
I don't see any report in WMT for duplicate or near duplicate content, Whitey - only title elements and meta descriptions.
| 7:50 am on Jul 7, 2008 (gmt 0)|
Something weird's happening - not sure if it's G or just WMT - but there appears to be a correlation.
More issues on WMT URL's when logged in :
- click on link* and it takes me to a URL that is different;
- then click on the link shown in the cached snapshot of the page expecting it to show the corresponding page
The cache says " These terms only appear in links pointing to this page: and quotes the referring URL words without backslashes" - now that's double weird because the only page that could refer to it, doesn't.
- but when the link on the cache is clicked it shows the page of the original link*
I thought i was seeing things, so i got on the phone to someone to view if it from a different IP address. Same thing.
The cache continues to be inaccurate when i log out from WMT.
What's interesting is that WMT is picking up the same error shown on the normal cache search and recording it in WMT as a dupe when it shouldn't!
Just one of the other errors I'm trawling through.
[edited by: Whitey at 7:57 am (utc) on July 7, 2008]
| 7:57 am on Jul 7, 2008 (gmt 0)|
It's not just you Whitey,
i'm seeing a mixture of correct and incorrect dupe issues in GWT.
| 8:01 am on Jul 7, 2008 (gmt 0)|
Thank goodness you're seeing it too.
What's more it seems to indicate that WMT and the caching system is up the creek. Surely this is flowing on to results .... I'll do some more digging on the results.
Here's some more errors :
- URL's designated as duplicate are sometimes refering to a truncated URL
All the duplicate meta tag and description notification have been generated recently, probably in the last 10 days since i last looked [ sorry i don't know the exact date ].
[edited by: Whitey at 8:08 am (utc) on July 7, 2008]
| 8:06 am on Jul 7, 2008 (gmt 0)|
It does sound like there's an indexing issue with the site, Whitey, and WMT (which is only a report) is picking up the strange data. I'd say really dig into that cache page issue to see what you can uncover.
| 8:20 am on Jul 7, 2008 (gmt 0)|
When I take part of the URL with the wrong cache and place it into Google "the web" like this [ without the www.URL.com ] :
It returns a result with an incorrect title of the referring WMT URL that i reported, the corresponding incorrect description, but the correct URL
But here's the news , when i click on the incorrect title link, it takes me to the correct page !
There are similar pages across the site, but they vary to the above with their titles
[edited by: Whitey at 8:27 am (utc) on July 7, 2008]
| 11:14 am on Jul 7, 2008 (gmt 0)|
The main WMT problem that I noticed about a month ago, was in the "short meta description" report (as that was the only issue the site had).
Sometimes, the same URLs were listed more than once, and with the same meta description for each of the duplicates. These are static URLs for pages that do not use parameters. There are NO canonical issues of any sort on the site.
When we cleaned up the tags, Google cleaned up their list at the next update.
The update of that list is really quite slow. When all of the pages of the site had been cleaned up, it took Google over two weeks to update those reports to show that all was OK (this was for a site with less than 100 pages, and only a very small number of errors).
We noticed that it almost always updated the 'meta description reports' about one to two days after the "homepage last accessed" message changed, and that the 'links report' was usually updated the day after that.
Nice pointer about meta description problems in the site:domain.com search: [webmasterworld.com...]
| 11:38 am on Jul 7, 2008 (gmt 0)|
|Is anyone noticing a glitch with the content analysis tool in WMT |
When checking " Pages with duplicate title tags" a large number of links mentioned resolve to the wrong URL.
Yes, I saw this a couple of days ago with META tags in WMT. Couldn't figure out what was going on, went and checked my site and everything was fine there, but the different links in WMT all went to the same page.
| 12:31 pm on Jul 7, 2008 (gmt 0)|
>>but the different links in WMT all went to the same page.
That's exactly what I'm seeing, and it's a formatting/html glitch. If there's a list of 3 pages that have dup titles (or descriptions) - making these up as an example:
Then clicking on any of them gives only the first - kitchen.htm
| 12:45 pm on Jul 7, 2008 (gmt 0)|
hmm what WMT
| 4:25 pm on Jul 7, 2008 (gmt 0)|
>>hmm what WMT
Google's Webmaster Central.
| 4:54 pm on Jul 7, 2008 (gmt 0)|
ohh - well I dont use that anyway
| 8:07 pm on Jul 7, 2008 (gmt 0)|
The only way I have been able to effectively use this data is to download it and cut and paste the URLs into a browser.
Would be nice if they fixed this.
| 10:33 pm on Jul 7, 2008 (gmt 0)|
|But here's the news , when i click on the incorrect title link, it takes me to the correct page ! |
So how do we explain that when clicking on the incorrect meta title relating to another URL in the Google SERP's it returns to the correctly intended URL.
I'm not sure if my words are clear enough, but this is what i find most significant.
Why are the caches incorrect ?
Why are the results not matching to the correct page?
These URL's originate in the incorrect WMT reporting
|not sure if it's G or just WMT - but there appears to be a correlation |
The errors seems to be extending beyond WMT into the SERP's.
Is anyone else seeing wrong meta titles, wrong caches , then clicking through to the correct URL?
[ The place to source these URL's appears to be in WMT.]
There are also issues with wrong results in the UK SERP's :
|http://www.webmasterworld.com/google/3687429.htm On results from google.co.uk [ filtered with "sites in UK" ] we are showing both our .COM and .co.uk site in positions 45 and 46 for the same term. |
Neither site is hosted in the UK or has any Whois information for the UK. The only relationship with the UK is the TLD of the 2nd one.
They also have different content.
Is there a Google UK glitch on at the moment ?
I wonder if this a sign of some major changes ahead.
Any other reports ?
[edited by: Whitey at 10:41 pm (utc) on July 7, 2008]
| 7:16 am on Jul 8, 2008 (gmt 0)|
|Is anyone else seeing wrong meta titles, wrong caches , then clicking through to the correct URL? |
Still interested to hear from anyone on this, but is anyone seeing their sitemap pages showing for "duplicate" content, when they wouldn't expect them to.
| 10:44 pm on Jul 8, 2008 (gmt 0)|
Just for reporting this:
I have over 6200 duplicate titles where almost all of them are what you describe. Very odd. It is always a pair, one with the correct data. It looks like:
"this is a duplicate title"
Link : someurl-111.html
Link : someurl-111.html
Only the LinkTXT (the displayed url) is wrong, there is no duplicate.
My site is doing very well atm, but I fear this issue will mess up everything. This could be a major issue. I have checked for duplicate titles a few days ago with no problems and now such stuff is rising.
I HOPE GOOGLE IS AWARE OF THIS - OR THIS MAY HIT THE FAN SOON!
| 11:41 pm on Jul 8, 2008 (gmt 0)|
rocco - What % of your overall page total [ reported with site:yoursite.com ] are the reported duplicates and home page TBPR [ tool bar page rank ].
Also, what do you show for site:yoursite.com/*
6200 duplicates could be dangerously high
[edited by: Whitey at 11:43 pm (utc) on July 8, 2008]
| 8:49 am on Jul 9, 2008 (gmt 0)|
Whitey about 1-2%, pr6, site:yoursite.com/* drastically fell lately
| 11:24 am on Jul 9, 2008 (gmt 0)|
Your PR may keep you out of danger with this low % while you tidy what you can in WMT
| 11:26 am on Jul 9, 2008 (gmt 0)|
This is a Google bug, Jonathan Simon confirmed the issue over here [seroundtable.com], it should be fixed shortly.
| 12:16 pm on Jul 9, 2008 (gmt 0)|
I appreciate the heads up, rustybrick - thanks!
For those who may not know the name, Jonathan Simon has been on the Google team since 2006, and he participates in the Webmaster Help area of Google Groups. Jonathan was at PubCon last year, and he wrote a long post addressing some of the key questions he picked up in Vegas [groups.google.com].
So we do have an answer. The correct pages are flagged in Webmaster Tools as having duplicate titles, but the underlying links are currently wrong. It will be fixed soon, but in the mean time the data in the downloadable spreadsheet is not problematic.
| 4:13 pm on Jul 9, 2008 (gmt 0)|
Cool to know, thanks. In the meantime I have found some real possible issues on my site that I have never thought of while looking into this issue.
| 10:13 pm on Jul 9, 2008 (gmt 0)|
Likewise ... rustybrick .... much appreciate the feedback & thanks to Jonathan Simon for keeping an eye out on the threads.
| 10:04 am on Jul 12, 2008 (gmt 0)|
Good to see that someone is on the case.
| 11:10 pm on Jul 15, 2008 (gmt 0)|
Hmm. The "Crawl Stats" graphs updated today. Is this a weekly event now? Seems a bit like it to me.
Anyway, on the busiest days, according to the graphs, Google pulled up to 20 pages from the site per day.
I know for a fact, that on many days they actually pulled about 50 to 80 pages. On some days they did pull less than ten pages.
| 5:27 am on Jul 16, 2008 (gmt 0)|
Just for the record [ in case folks here are relying on this feature ] , the duplicate content reported on the site:tool is much more than shown on the internal reporting.
On one of our sites, I noticed we had 88 reports of duplicate meta titles on the site:tool and only 20 in WMT for our "image gallery" pages.
[edited by: tedster at 12:13 am (utc) on July 18, 2008]
| 11:53 pm on Jul 17, 2008 (gmt 0)|
Cross referred to other problems currently reported in WMT : WMT - Web crawl glitch [webmasterworld.com]
Are there any more?
[edited by: Whitey at 11:53 pm (utc) on July 17, 2008]
| 11:18 pm on Jul 19, 2008 (gmt 0)|
Yes, I can confirm having seven "short meta description" errors in the report in WMT, but about double that on a site:domain.com search, where the snippet consists of the first words from the page instead of Google using the meta description for the snippet.
| 11:05 pm on Jul 20, 2008 (gmt 0)|
Yes, we do have some parallel threads running - let's take further discussion here: