Welcome to the Q&A Forum

Excellent thanks Dana I have a premium subscription to Screaming Frog so I'll get on this straight away - thought there must be an easy way to find them but forgot about the custom settings in Screaming Frog Will get on this right away Kind Regards Liam

Ha! Hi Liam - as you can see STILL no technical knowledge from me

I'd agree with the other guys here: just because GWT does not find an error does not mean there is one. What Google usually does in these situations is choose one version over another, so if you Google your site - even using the site: search modifier - you'll only see the one they've chosen. On the other hand  Moz's crawler goes through your site and reports on everything it finds, which in this case means it's a factual statement that there's a way to move from the www. version of your site to the non-www. version. The first thing you need to do is find out which page on the wrong version (which I'll assume to be the non-www. versions or versions with trailing slashes ) is being linked to from the right version (which I'll assume to be www. without a trailing slash) and where that link is. If you use a link crawler like Xenu's Link Sleuth you should be able to do that. Then you can go and remove the troublesome link(s) and all your errors will disappear.

Hey Liam, Thanks for following up. Unfortunately, we use thousands of dynamic IPs through Amazon Web Services to run our crawler and the IP would change from crawl to crawl. We don't even have a set range for the IPs we use through AWS. As for throttling, we don't have a set throttle. We try to space out the server hits enough to not bring down the server, but then hit the server as often as necessary in order to crawl the full site or crawl limit in a reasonable amount of time. We try to find a balance between hitting the site too hard and having extremely long crawl times. If the devs are worried about how often we hit the server, they can add a crawl delay of 10 to the robots.txt to throttle the crawler. We will respect that delay. If the devs use Moz, as well, they would also be getting a 403 on their crawl because the server is blocking our user agent specifically. The server would give the same status code regardless of who has set up the campaign. I'm sorry this information isn't more specific. Please let me know if you need any other assistance. Chiaryn

Some fantastic Replies, thank you very much. I think I will be removing the exclamation marks simply due to the number of titles that have too many characters anyway. The point about the positioning of the keyword was also good, that had completely skipped my mind on this occasion! I'll see what difference removing them makes to the CTR and weigh up the options. Thanks

I think that this is the video that was mentioned by Rich. It goes back to 2011. Matt does say that he can't account for other search engines, but Google is cool with using a rel="canonincal" tag to point to "itself."

Hi Ladies & Gents It was a canonical tag error - we went from 6000 errors to 11500 errors and now after the issue was fixed we have a handful of non related errors to sort. Really appreciate all the feedback. Kind Regards Liam