Welcome to the Q&A Forum

Hi Dan,

The robots must be getting the urls from somewhere so it is worth finding out where. If you download the moz report in csv and open in excel you can control-windows-L to get a filterable list. If you filter for duplicates and find these urls on the left then on the far right it should reference where they are being linked from. I suspect you will find pages in the site that have these images in them and are linking to the attachment_id urls (often it is from gallery pages).

Once you have found the pages, then try applying the yoast redirects and see if they work as expected (ie redirect the attachment_id links to the relevant gallery page for example). Ideally you would get rid of the links completely from the code - this will probably need a bit of dev work on the template but should be pretty straightforward since you are likely just removing the A tag from around the images.

Hi Dan,

Is the site running wordpress? If so it sounds like maybe a badly coded template which is showing links somewhere in the code to the attachments (if you click on the image in its normal page does it take you to the duplicate url you mention?). It would be best to find out where the linking is happening and correct it so the links are removed if at all possible. The Yoast plugin also has a setting where you can redirect attachment ids to their related post (its in the permalinks settings of the yoast plugin) - that might help solve the problem.

Hi Tim,

Are you using the Yoast Seo plugin? If so a lot of things including canonical tag settings can be sorted out there. I don't think much has changed in terms of wp optimisation in 2015 - best practices from the past continue to be best practices now. From your description it sounds like maybe the problem isn't the canonical tag so much as your url (permalink) structure and/or the indexing of your archive and tag pages. Yoast has a good rundown on best practices in regards permalinks and general wp seo here and Dan Shure wrote a good blog post here on Moz which includes a good explanation of why and how to noindex certain pages (like archives and tags). The blog post is from a few years back, but all of the best practices he mentions remain valid.

HI Thomas,

You could use a wordpress multi site installation for this which allows for setting up different wp installs on subdomains or sub directories as Ryan suggests. This will allow you to have separate 'sites' while all still being controlled from one wp admin panel and with a bit of coding/setup work to tweak things if needed. You can also use a domain mapping plugin to run multiple domains (.com, .com.au, .co.uk etc) from the subfolders if that is in the plans.

It looks like it is spinning off 404s based on your main site's html pages (which is not in wordpress right?) which makes it look like not so much a wordpress problem as a linking problem. The blog is using relative urls (like this: ../../../../../clients.html) in its header and footer links back to the main site. This can potentially cause problems and even though in a browser the links all seem to be working fine the 404 you mention looks like it could be created by this setup. I would put absolute urls in the main menu and footer links and see if that clears up the 404s.

Hi,

If you can do as recommended above and get rid of the page completely then this would be the way to go. Since it is on a national newspaper site and a couple of years old I am guessing this might be tough to do (although worth a shot).

Otherwise you can try to boost your own rankings and grab more real estate by:

1. Making sure all your on page data is branded (page titles, company schema etc).

2. Use verified social channels and link to them from your site (especially google+ but fb, twitter, linkedin, youtube are all good).

3. Get a few more branded links into the site (I assuming you are in 2nd position, it is a bit unusual for a branded search to not return the brand as the first option - you might just need a few more (honest) links).

5. Short term solution: Adwords on your brand's search. You should get a 4 pack of site links + descriptions in it also so should give a fair amount of real estate above the newspaper article and should be inexpensive.

Is all that new traffic organic? Is your sale/conversion rate increasing at the same rate?

If the answer to either of those questions is yes then you should be moving cautiously. If the main aim is to make the site mobile friendly then do that first and monitor traffic/rankings for a couple of weeks. After that depending on what you see try implementing the other changes you have planned in order and give a week or two in between each change to monitor how things are going.

Changing urls, titles, metas, internal linking and content all at the same time on a site with that much traffic (and I am assuming at least some profit).... is crazy. If rankings/traffic plummet, which change caused it? Where do you even start diagnosing with that many changes at once? Nightmare!

HI,

If you have the subdomain set up in GWT then you could do a url removal request (Google Index -> Remove URLs). Just put in the root admin subdomain (ie https://admin.naturalworldsafaris.com/). This usually works pretty quickly and given your description will likely be the quickest way to drop admin urls that are showing up in the serps.

It is not an ideal solution and the tool will tell you that for permanent exclusion you should do it through your robots.txt file. in regards your question about that though - the robots.txt file for the subdomain has to be placed on the subdomain, adding the subdomain to the main domains robot.txt file wont work.

Hi,

I don't think you can specify profile and cover pics through the bulk upload tool (I never managed to) so you are likely going to have to go the manual route. Even so, with 80+ locations it only took an hour or two to upload them all. The bulk tool certainly saves some time uploading a lot of the needed data, but in my experience has a few bugs still.

As to whether the pics should be unique or not depends a bit on the circumstances. We were dealing with a franchised chain and it made more sense to standardize the profile and cover pics across all the stores (so... I would have quite liked to automate this to be honest!).

Hmm, was something done to fix the googlebot redirect issue or did it just fix itself? Here it states that googlebot will often identify itself as mozilla and your fetch/render originally seemed to indicate that at least some of the time that was the page google was getting. It is a bit murky technically what exactly is going on there but if google is getting redirected some of the time then as you said you are getting into a circular situation between the redirect and the canonical where it is a bit difficult to predict what will happen. If that is 100% fixed now and google sees the main page all the time then I would wait a day or two to see if the page comes back into the index (but be 100% sure that you know it is fixed!). I still think that is the most likely source of your troubles...

Hi,

Not sure why there would no h1 tag and then two, but to override the default category h1 you are likely going to have to do a bit of coding to add an extra field to your category cms pages so you can override the default. This link runs through a way to apply it (I haven't tried it personally so test thoroughly!).

How does that second version of the homepage work and how long has it been around for? I get one version of the homepage in one browser and the second in another, what decides which version is served and what kind of redirect is it? I think that is the most likely source of your troubles.

Hi Neil,

As Don says there is nothing technically wrong with serving the site from a subdomain, google will simply see all the content as coming from the www domain, so no issues with indexing, getting page authority etc.

The issue is likely to be the 'keeping some of the old links live' - I cannot think of a way you are going to be able to do this unless you recreate the old joomla content on the new wordpress install and either make the wordpress install serve the same urls as joomla did or 301 redirect the old urls to the new ones. When you say that you have advised that google will not be able to crawl them this means also that users will not be able to see them, that YOU will not be able to see them. So - they are not really live in any real sense (unless I misunderstood something!).

HI,

bmw -intitle:bmw

Should work (if you ignore the google rewritten titles in the serps).

Hi,

Having a self referencing canonical tag on your own pages is not a problem. The canonical tag needs to go into the head of the page though (it is not valid if in the body of the html) so just make sure that the 3rd party syndication service actually provides this - it might - but it might not I am guessing. Even with the canonical I would still include a clean text link back to the original page if this is possible (both as a second indication of origin but also for the visits it might send).

HI,

There was a recent WBF about syndicated content which runs down the various technical ways you can attribute your listings as the original source - check it out here. It would probably also help to make sure your listings are the first ones into the index which can be done by internally linking to the new jobs (obviously), quickly adding them to your sitemap, sharing them through social channels (especially twitter) - all of which should help make sure your content is indexed quickly and ideally before it is replicated on other sites.

If the other sites have stronger authority than yours you would still really want to get one of the 3 options discussed in the video implemented. It sounds like you already have the link back to your site (option 3 in the video) so perhaps the link is not 'clean' ie a straight text link that leads to the exact job on your site and is not no-followed?

It might also depend on what kind of long tail you are looking at ranking for. Individual job ads might not pull a lot of organic traffic by themselves if they are not aggregated by type or location for example - at which point the higher authority domains are likely to show an advantage (just a thought).

HI Neil,

The joomla installation will in effect be 'deactivated' when you switch the htaccess rules over to serving the root from the wp install (you will also be replacing index.php in the root folder if memory serves).  Google will not have a problem with this BUT the old joomla content will not be available anymore (to users or bots). You will be recreating the old content on the new wp install also right?

You will need to make sure your 301s are setup properly from the old joomla urls to the new wp urls and you will need to wait until after the switch to the wp site to verify these are working properly so a bit of preplanning and verification afterwards will be needed there. Other than that all the usual best practices for switching to a new site are all still of importance, check out this guide for a good rundown. Hope it helps!

HI,

If you are seeing these warnings in your crawl reports then 410 pages would be a high priority issue since it means you are still linking to these pages from within your own site. In that respect there is no difference between 404s and 410s, they both declare a page is no longer available and if you are still linking to these pages in your site then you should try to find the pages that are linking to them and update the links. If you download your crawl report, open in excel and then filter for 410 errors one of the columns on the far right should tell you which page was linking to the 410 page(s) in question.

That is a pretty good example you have there of a slightly ... unexpected ... knowledge box!

The knowledge box can draw data from a number of places, as Ron mentions schema markup and google plus pages are used as are other online sources like wikipedia (which is the source in your example) and freebase listings. Dr Pete has written a number of posts about this which might help (here, here and here for example) and another good article on 'hacking' the knowledge graph here.

The particular case you mention is a definition box and is driven from wikipedia and might be pretty difficult to change, if you are looking to influence knowledge graph data about a company or an entity the above links should help.

Hi,

The moz crawler should respect canonical urls and will not flag them as duplicate content as long as they are set up correctly. There should be no time lag in this process, if the tags are set up correctly then the next time you get a crawl report the canonicals should be recognized. If you are still seeing duplicate content issues then there might well be a problem in the canonical setup so it would be good to double check and make sure the canonicals are setup and working the way you expect them to be (Magento can be a bit of a nightmare with these kind of things!).