Welcome to the Q&A Forum

If the products are all but the same besides color, would you be able to implement something like amazon has, where each product is only one page and you select the color/size option within the the product framework?  This way instead of having a bunch of pages that are basically the same product, all of your effort can go to one page per product.

Have you tried adding in the automatic cross domain tracking to pass your cookie to paypal and epdq.co.uk?

https://developers.google.com/analytics/devguides/collection/analyticsjs/cross-domain#autolink

If you are seeing a break then something is causing it between your sub-domains, is there an auto-loading interstitial page or redirect of some kind being used?

I recommend Tag Manager as it makes defining click based events easier and the Tag Assistant tool for Google Chrome can be very useful in figuring out what is causing weird problems with your Google Analytics and other on site tools.

Ben,

I assume you set up your GA account awhile ago and are still using the old code, you have two options to fix this.  The first is you can keep your old code but change it to also have cross domain tracking, the other, and better option is that you upgrade your tracking to use the most current version as you will need to do this eventually.

I also recommend looking into Google Tag Manager as it allows for easy advanced tracking implementations.

Cross domain guide for analytics.js:  https://developers.google.com/analytics/devguides/collection/analyticsjs/cross-domain

Pixel tags do not effect your SEO. The only thing that could cause a problem is if they are loading slowing and causing load time issues with your site, otherwise back end code for tracking does not have anything to do with how you are ranked in search.

I would argue against Fiverr, not just on the grounds that it cheapens the landscape, but mostly due to the quality.  I tried using it multiple times from semi-technical and click bait style articles and every time the quality was poor to the point that editing them took more time and effort then just writing them would have been.

I agree with Alberto about Odesk.com, it is a great resource to find freelance writers with.

Once you remove it in GWT make sure to update both your robots.txt to not index along with adding the no index tag to all pages on your test site.  Be careful that you don't accidentally push either the robots or noindex tag onto your live site.

The reason that you want to do both is, while the robots is the best option, if somehow a page on your test site gets linked it can show up in Google for a few days until Google re-crawls your whole site and sees that the robots says not to index anything.  You can tell if this problem is happening in your analytics if you see traffic to test pages spike up for a few days every couple of months.

Ok, the problem is your pages are all canonical to themselves, the canonical tag should point at the main page for the content, not to every page.  For your first example, all pages that get their content from http://www.spain-internship.com/en need to have canonical tags to that page, instead the copy page has this:

href="http://www.spain-internship.com/fi/etusivu/186-all-fi/home-page-fi" rel="canonical" />

it should have

href="http://www.spain-internship.com/fi/" rel="canonical" />

Have you tried searching the account number to see what other websites (if any) are using that account? If you client's UA number does not end in a -1 then there is a good chance whomever made the site used the same account for all of the work they did, including their own site.

Do you have an example of one of these generated pages as well, everything looks fine on the main page.

Do you have a link to one of these pages so we can look at how it is deploying the canonical onto the page.

Bought better code, the old galleries had been a hack job by a previous agency, we bought new clean code, customized it for our needs.  We also upgraded from using on-page analytic code to Google Tag Manager at the same time which made all future tracking updates much easier and cheaper.

As a best practice you should, and while I do not think it will fix your data, who knows it might. The problem will be, if it does not give you better insight now, all it will do is split up this unknown data into 2 separate groups in your data instead of just one.

I had a similar problem on a site awhile ago, the issue was not with the analytics code but with the site which had a large amount of java galleries and procedurally generated content which would somehow create small interstitial pages when generating pages and redirect from them to the actual pages.

It will depend on how/why it is saying you are a referral source to yourself.  There most likely is a larger issue with how your site is functioning that is causing a disconnect to happen that just removing yourself as a referral will not solve.  If there is still a break in your tracking then it will see it as a new session with no source information and attribute conversions to (not found) or something similar.  To fix the issue you are going to need to understand why your analytics program thinks you are sending yourself traffic and fix that problem first, then you will be able to get real and actionable data on what your main conversions drivers are.

This is exactly right, I have them but the only value outside of what I learned to pass them is if you are trying to be a Google partner (http://www.google.com/analytics/partners/apply).

Bas is 100% right, you do not have to but each social media outlet's audience act differently. To get the most out of your social media postings you should tailor your message delivery and content based on each platforms strengths and what your customers respond best to on each independently.

Honestly I would not worry about it to much, their yelp page has a lot of positive reviews so it would hurt them to change that.  You could try adding a CAPTCHA onto the contact form on the site to dissuade sales people and stop auto fill bots, but that could inconvenience potential customers so you will need to weigh the pros and cons. You could try adding their number to no-call lists to see if that helps, but I don't think that there will be a ton of calls in the long run, things like this tend to happen a bunch for a week or two then drop off.

Ben,

Your image did not link properly.  As I cannot see the data, does the majority of the CA traffic come from one or a few places that are not major cities? Have you disabled bots in the view settings?  It seems to me that there might be a bot crawling Yelp listings within that area collecting and verifying phone numbers to sell.

How do you have your goal set up? it seems like your goal is firing twice, or you have an automatic page reload somewhere, unfortunately problems like this are hard to diagnose without being able to actually see your analytic's information. The best you can do is look at the rules defining your goals and see if they are to broad, if you are basing them off on an event that is hard coded onto your site then you should also look into your code to make sure it is not firing twice during a purchase.

Meta descriptions do not effect your page optimization in the sense that a search engine is effected by it, but it will effect how often people click to your site as the meta description is the copy a person is shown. You should make your descriptions informative and truthful so that it accurately portrays what each page has to offer giving a potential visitor the best ability to judge what content will be on the page (and thereby lowering bounces/re-searches). This being said you should write custom meta descriptions, not just let it truncate content already on your page.  I would start with products that are relevant to the most used keywords and work backwards, if you dedicate yourself to doing at least 20 descriptions a week you will have all of your legacy products done in a year.

Are you sure that the URL's are the same, even small changes to URLs can effect your site.  Look at your Google Analytics to see if there is a change at the page name level and when it occurred.  This could also be a coincidence but doubtful) check your GWT to make sure that you have not been penalized at all.  Lastly, while your URL may not have changed, did the page layout and/or information on each page remain the same or was it changed?

Awesome, I hope it fixes your problem.

I would guess it has to do with a lack of any actual content on the main page, and all of the other pages "content" being aggregated from other sites where it was originally posted.  A site like this would need either a large amount of advertising to drive traffic or in my opinion a complete overhaul and reevaluation of the content strategy.

I would recommend getting Yoast SEO for Wordpress, then in the Yoast options for XML sitemap under taxonomies tell it to exclude Categories.  Most likely after the migration somewhere an option that creates or references pages based on categories and/or tags was left on.  You can simple exclude these pages from your sitemap while you check your other plug-ins and settings to make sure that one of them is not trying to reference these pages which have been turned off.

How long ago did they make the updates?  Did you also update your sitemap to no longer have uppercase versions in it?

Write as much copy as you need to convey the information that is needed for your visitor on that page.  If you can create a good user experience on one page with 100 words (by answering the pain that they would be going to that page to solve) and then you need 1,000 words for your next page, then do that to.  Don't make arbitrary copy limits or goals based on SEO, make your pages user friendly and your copy useful and your pages will rank well for the appropriate content.

While you should not fear change, your comment "However, the current setup is sending a worthwhile traffic volume to the site." says all you need to know, if the service + city landing page model is working, don't change it to change it or because you read something that says you should.  If you do however believe that targeting the broad keywords instead will net you better local results, set up a few landing pages for the broad keywords and A/B test them against the local landing pages.

CID= is a campaign code for Omniture I believe, but you use it for tagging external campaigns that link back to your site.  You should find out more as to why your team wants to do this.

If you have canonical tags on each page to themselves the query string should not effect your SEO (they shouldn't anyways, but it is safest to do it).

This is done for pages that may have additional information added to the url in a query string. While I prefer to have all pages canonical to themselves, I highly recommend doing it to any home page or landing page that can have multiple external links pointing to it.

You may want to speak with your team as to why they want to add query strings, your analytics program should not need internal links to have query strings added, and if you are using Google Analytics adding their UTM query strings to internal links will cause major errors with your data.

If they do have legitimate reasons to query tag internal links, just make sure that you have all of your pages canonical tagged to themselves and then you won't have to worry about it messing up your URL structure.

If you have separate pages for mobile and desktop that contain different information then yes, you should make sure that your meta and title reflect the information that is displayed on the page per device.

The easiest way to do this is to export the data and work with it outside of Google Analytics. I reccomend using http://www.nextanalytics.com/ to automatically export data from GA into excel.

The idea of redirecting a user to a car that might not match their search does not seem like a very user friendly option, if they wanted a mustang and clicked a search listing for it but none are available and were then redirected to a camaro page the user would not be happy, and that flow is built only for the site and not the customer, IMO.

My thoughts are instead of worrying about what is best for Google, think of what will give a user the best experience and go with that.  While it is nice to have a lot of pages index, if by the time they get to Google they are gone, what good does that do a visitor who was searching for a specific term that your site no longer offers? They are much more likely to leave which will effect your whole site negativity as bounces from search go up.

You would need to use a series of filters and/or regular expressions to draw out only your job pages based on the domain and then place information into the URI.  Your filters can be found in the admin section under your view.  It takes a good amount of testing to get filters like this right, and a small mistake can destroy a lot of data so make sure not to use your main account and instead create a testing account to fine tune this.

Using universal is it not hard to, you can just place the code onto the 3rd party site.  To make it easy to look at you will want to make a find and replace filter to target only the traffic to your 3rd party site and change the URI so that it will make more sense when looking at it (if you do not then it will see your home page and the main page of the jobs site as just "/"). Make sure that you set up a test profile for your filters so that you do not cause problems with your data.  This is the more complex way of looking at it, if you only care how many people click off to your jobs section and not how they interact with it you can place event tracking code onto any link that goes to the job section.

All of this is much easier if you use Google's Tag Manager as the cross domain linking or adding custom event calls become even easier to implement.

Does your 3rd party service have the ability to add GA code to include the necessary cross domain tracking code or does it use universal analytics?

You may want to be careful keeping pages live that are basically useless to visitors, if the ads are expired and it makes people leave your site (bounce) it will hurt your entire site, not just those pages.

Like most social media buttons, any link involved is tagged nofollow.  There are really no SEO effects to adding a button like that onto a site.

It will not remove a page just because the link is gone. Your best bet would be to set up your back end to automaticly add a noindex meta tag into the of ads once they expire.

Example of tag

If they are doing a lot of campaigns where they would have many urls with various query strings pointing to the pages then it would make sense to just have each page generated with the tag on it.

You put all the work into those articles, why would you throw away all the search relevance that they have earned? I would never ever move content and not redirect it, no matter how many 301s I would have to add.

While you will always loose some ranking with large site wide restructures, you can minimize it by having your redirects in place as soon as you make the shift.  When I do restructures or site re-launches I create a large CSV that has all of my current URLs and the corresponding URL that it will update to.  Double check the list and use it as a checklist to make sure all of your old content properly redirects to the new.  On a side note, when doing something like this it is a good time to update your site to secure as you are already redirecting the majority of your pages, so redirected them to the https: version is not much extra work.

As for if you should do this, that will depend on your site goals and if adding the date in the url would benefit your visitors or corresponds with another digital part of your digital strategy.

You are correct in your assumption, if you are not allowing Google to index those pages then Google image search will likewise not index images from those pages.  While you could then create gallery pages for your images that did not have noindex meta tags, without having other information on those pages the images will not rank well (and at that point you would be better spending your time just creating unique content for your similar product pages).  Unfortunately, if you want your images to searchable you will need to make your pages as well

Dave,

Your best bet to keep these pages indexed but not have duplicate content issues is to add code that will place a canonical tag on search pages based on the original search query so that even when additional information was caused to load into the URL, such as your example, it would point back to the base page.  If you do not want these pages indexed at all you can do what Keszi said or block all search pages using your robots.txt.

I would not block your product pages or search pages but I agree with blocking your shopping cart funnel.

Google official explanation of canonical and some common mistakes Google sees when they are implemented.

https://support.google.com/webmasters/answer/139066?hl=en

http://googlewebmastercentral.blogspot.com/2013/04/5-common-mistakes-with-relcanonical.html

They are different things for used for different reasons. By using the robots to block any page behind a log in you will not have to worry about them trying to access that information at all. You should also have canonical tags pointing to themselves on all pages especially product pages and landing pages.

What do you mean "pages that have nothing to do with my site" are these not on your domain or are they on your domain but you are not familiar with them?