Moz Q&A is closed.
After more than 13 years, and tens of thousands of questions, Moz Q&A closed on 12th December 2024. Whilst we’re not completely removing the content - many posts will still be possible to view - we have locked both new posts and new replies. More details here.
URL path randomly changing
-
Hi eveyone,
got a quick question about URL structures:
I'm currently working in ecommerce with a site that has hundreds of products that can be accessed through different URL paths:
2)www.domain.com/category/productx
3)www.domain.com/category/subcategory/productx
4)www.domain.com/bestsellers/productx
5)...
In order to get rid of dublicate content issues, the canoncial tag has been installed on all the pages required. The problem I'm witnessing now is the following:
If a visitor comes to the site and navigates to the product through example 2) at time the URL shown in the URL browser box is example 4), sometimes example 1) or whatever. So it is constantly changing.
Does anyone know, why this happens and if it has any impact on GA tracking or even on SEO peformance.
Any reply is much appreciated
Thanks you
-
If that was the final product page, then yes, you should be using a standard htacess rewrite command to ensure that the final product urls are always www.domain.com/productx But in saying that, the way you had it is totally fine if all the other url possibilities have a canonical tag that points back to optimised version (original product url - www.domain.com/productx)
The htacess rewrite it's not something you should be handling manually. Magento has that option inbuilt into it. It would be a fair amount of work if you had to do that manually. and I would just run with the canonical option if that were the case. Any good eCommerce platform should have the inbuilt ability to automatically remove the category folders and other search queries from the final product url.
Sometimes it's ok to leave the category folders in the url, it just depends on the products being sold. Below would be an example where I would leave the category folders in the url if I was selling different colored soccer balls.
www.sports.com/soccer-balls/black-white/
www.sports.com/soccer-balls/blue-white/
www.sports.com/soccer-balls/red-yellow/ -
Hi Richard,
thanks a lot for your reply.
Let me clarify: productx is a final product page (not a category page with a variety of products). This means that my productx page basically corresponds to your /final-product example.
According to your post and the htaccess command mention, I assume, that it does not cause problems, if the URL shown in the browser does not correspond to the one, the user actually took?
So no matter if a customer comes to the final product page through 1) 2) 3) 4). The URL shown could always be 1) and thats fine. Is that what you ment?
Thanks in advance
-
Those paths all seem fine as they are all legitimate ways of getting to that bunch of products. I'm also assuming that the final page on each of the below urls is a page that contains a selection of products and you can still click on an individual product from the list and go to its url.
1)www.domain.com/productx
2)www.domain.com/category/productx
3)www.domain.com/category/subcategory/productx
4)www.domain.com/bestsellers/productxIt's even debatable whether you need canonical tags on any of those above urls, it all depends on how different those pages are from each other with regards to the content on the page before the products. If all of those above urls had different H1 tags and different content before the product feed then they are all stand alone legitimate pages that don't need canonical tags. But if they're all virtually the same and not much customization has been done to the auto generated pages then yes, they should all have a canonical tag back to www.domain.com/productx perhaps, or the most suitable page.
A bigger issue I think you may have is the url of the final single product page. It shouldn't be like this:
www.domain.com/category/subcategory/productx/final-product/
or like this either:
www.domain.com/bestsellers/productx/final-product/Optimally, it should read like this no matter how the visitor got there:
www.domain.com/final-product/In most cases, for optimal Seo, an extension, plugin, or htaccess command should rewrite the final product url to strip out all of the category url paths or best seller url paths from the urls so your final product page url is short and clean like this: www.domain.com//final-product/ even though the path is really ths: www.domain.com/category/subcategory/productx/final-product/
It's pretty hard to tell what the optimal solution for your site is without having a look at it and understanding your product range and categorization a bit better, but I hope that helps a bit.
-
Thanks for your reply Hector,
The way, the pages are structured on our site is the way 99% of ecommerce business have them structured so that is not the issue here. It's more the path itself that concerns me a bit.
Cheers,
-
It is a technical issue with your ecommerce platform. Definitely it is not good to have that kind of different URLs.
Canonicals are helpful with pages where you cannot do anything but having two similar pages on your site, or when there are almost identical pages. But when dealing with such an important page on an Internet project like the product page on a ecommerce site, you should definitely take action and manage to have unique URLs for every product, not depending of the path the visitor follows to reach that page.
It will become difficult to measure conversion rates or any other KPI on Analytics, and also will become a problem in SEO, with so many different pages to link.
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
SERP result (URL) doesn't change after a 301
A couple of months ago there was a result in Google for our branded search term which wasn't the 'official' URL, actually the result shown in the SERP was www.mycompany-ip.nl. We've applied a 301 redirect of this URL to the 'official' URL which is a subdomain: department.mycompany.nl. From Google the redirect is obviously working, but up until now, I don't see Google replacing the incorrect URL by the correct URL. I am wondering what to do to make the result correct. André
Technical SEO | | ConclusionDigital0 -
Resubmit sitemaps on every change?
Hello Mozers, Our sitemaps were submitted to Google and Bing, and are successfully indexed. Every time pages are added to our store (ecommerce), we re-generate the xml sitemap. My question is: should we be resubmitting the sitemaps every time their content change, or since they were submitted once can we assume that the crawlers will re-download the sitemaps by themselves (I don't like to assume). What are best practices here? Thanks!
Technical SEO | | yacpro131 -
Special characters in URL
Will registered trademark symbol within a URL be bad? I know some special characters are unsafe (#, >, etc.) but can not find anything that mentions registered trademark. Thanks!
Technical SEO | | bonnierSEO0 -
Spaces (actual spaces) in URL
Hi all, Is there a huge loss of SEO performance if a URL shows spaces with an actual space (i.e. %20) in the URL rather than a "-" (or indeed a "_")? I know the preferred option is to have a "-", but I am just wondering if it is worth our effort to manually change the "%20" to a "-" in all the instances? Thanks 🙂 Diana
Technical SEO | | Diana.varbanescu0 -
Should the date be included in news URLs
My website is not a news or magazine site, but we do have a news section updated 2-3 times a week with industry related news. We are working on a new structure for the URLs.
Technical SEO | | theLotter
Should the date be included in the URL? From this article from Google I understand that as long as we submit a news sitemap it doesnt matter whether or not numbers are included in the URL, correct? https://support.google.com/news/publisher/answer/68323?topic=116650 -
URL - Well Formed or Malformed
Hi Mozzers, I've been mulling over whether my URLs could benefit a little SEO tweaking. I'd be grateful for your opinion. For instance, we've a product, a vintage (second hand), red Chanel bag. At the moment the URL is: www.vintageheirloom.com/vintage-chanel-bags/2.55-bags/red-2.55-classic-double-flap-bag-1362483150 Broken down... vintage-chanel-bags = this is the main product category, i.e. vintage chanel bags 2.55-bags = is a sub category of the main category above. They are vintage Chanel 2.55 bags, but I've not included 'vintage' again. 2.55 bags are a type of Chanel bag. red-2.55-classic-double-flap-bag = this is the product, the bag **1362483150 **= this is a unique id, to prevent the possibility of duplicate URLs As you no doubt can see we target, in particular, the phrase **vintage. **The actual bag / product title is: Vintage Chanel Red 2.55 classic double flap bag 10” / 25cm With this in mind, would I be better off trying to match the product name with the end of the URL as closely as possible? So a close match below would involve not repeating 'chanel' again: www.vintageheirloom.com/chanel-bags/2.55-bags/vintage-red-2.55-classic-double-flap-bag or an exact match below would involve repeating 'chanel': www.vintageheirloom.com/chanel-bags/2.55-bags/vintage-chanel-red-2.55-classic-double-flap-bag This may open up more flexibility to experiment with product terms like second hand, preowned etc. Maybe this is a bad idea as I'm removing the phrase 'vintage' from the main category. But this logical extension of this looks like keyword stuffing !! www.vintageheirloom.com/vintage-chanel-bags/vintage-2.55-bags/vintage-chanel-red-2.55-classic-double-flap-bag Maybe this is over analyzing, but I doubt it? Thanks for looking. Kevin
Technical SEO | | well-its-1-louder0 -
404 errors on non-existent URLs
Hey guys and gals, First Moz Q&A for me and really looking forward to being part of the community. I hope as my first question this isn't a stupid one but I was just struggling to find any resource that dealt with the issue and am just looking for some general advice. Basically a client has raised a problem with 404 error pages - or the lack thereof- on non-existent URLs on their site; let's say for example: 'greatbeachtowels.com/beach-towels/asdfas' Obviously content never existed on this page so its not like you're saying 'hey, sorry this isn't here anymore'; its more like- 'there was never anything here in the first place'. Currently in this fictitious example typing in 'greatbeachtowels.com/beach-towels/asdfas**'** returns the same content as the 'greatbeachtowels.com/beach-towels' page which I appreciate isn't ideal. What I was wondering is how far do you take this issue- I've seen examples here on the seomoz site where you can edit the URI in a similar manner and it returns the same content as the parent page but with the alternate address. Should 404's be added across all folders on a site in a similar way? How often would this scenario be and issue particularly for internal pages two or three clicks down? I suppose unless someone linked to a page with a misspelled URL... Also would it be worth placing 301 redirects on a small number of common mis-spellings or typos e.g. 'greatbeachtowels.com/beach-towles' to the correct URLs as opposed to just 404s? Many thanks in advance.
Technical SEO | | AJ2340 -
Changing DNS -- SEO implications?
Hey Moz, We're migrating an old site on an old server over to a new server/DNS. The plan is to keep the same URL structure and reuse our existing URL's. As long as we make minimal changes to each page's content, we should be able to update our DNS entry and get all the pages recreated and assigned to their correct URLs without any reduction in SEO rankings. Is this correct? This site gets a lot of organic traffic and ranks highly on some challenging keywords, so it's key that we retain our rankings as much as possible. I've read that it's wise to lower the DNS time-to-live to one hour, about a day before the move, to help Google crawl the DNS a little quicker. Are there any other recommendations you guys can offer or past experiences?
Technical SEO | | stephen_reply0