Is SeoMOZ Crawl Diagnostics wrong here?
-
We've been getting a ton of critical errors (about 80,000) in SeoMoz' Crawl Diagnostics saying we have duplicate content in our client's E-commerce site. Some of the errors are correct, but a lot of the pages are variations like:
www.example.com/productlist?page=1
www.example.com/productlist?page=2
However, in our source code we have used rel="prev" and rel="next" so in my opinion we should be alright.
Would love to hear from you if we have made a mistake or if it is an error in SeoMoz.
Here's a full paste of the script:
-
Just a minor clarification - you can use both rel=prev/next and rel=canonical, IF you have something like search filters. Then, the canonical would point to the unfiltered current page and the rel=prev/next would point to the filtered paginated pages. Yeah, I know, that made a lot of sense. Let's say your page is:
http://example.com/stuff?page=2&sort=price
...then you might have
It's more than a little confusing.
Definitely check out that JavaScript issue, though - it might be that bots aren't seeing what people are seeing, and that could be very dangerous.
-
Hi,
In regards the rel=next you are absolutely right, I must have overlooked it or just searched for the prev tag. So yes as far as proper implementation of the prev/next in that respect it is correct and please ignore that last part of my first post!
Turning of javascript is instructive to see all those tags on their individual page and helps clarify what exactly is being outputted and when without the dynamic loading, providing you don't miss a rel=next tag that is really there
-
Hi Lynn,
Thank you very much for your answer / analysis! As you said "It is a bit confusing" and I will just read your answer a couple of times...
I will grant your answer "Good answer" for you thorough analysis! I think it is spot on with the double "next/prev" and "rel=can" tags. I do have one remark. You said: When I turn off javescript, I get this:
In my opinion this is alright, because it shouldn't have a "prev" as this is the initial page.
-
Hi,
I had a look at what I assume is the site and I think you have a combination of things going on that is likely causing confusion (to you, to the moz bot, probably to google too!)
Firstly, it is not recommended to use rel prev/next and rel canonical on the same page. With that what you are effectively doing is only indexing the first page of the results since all the other pages rel canonical back to the first one. If you have a 'view all' type page then you could rel canonical all of the paginated pages back to this one and you would not need to use the prev/next tags at all. It is also possible that your use of relative canonical links in combination with the above is also causing confusion, usually best to use absolute urls if possible.
Beyond that, the site dynamically loads more products as you scroll down the page which also results in the url changing to hoeretelefon/? for ALL the pages. If that is a problem or not depends on how it is coded and how the google and seomoz bots are deciding to parse the page, but it certainly adds another potential area of complexity to the issue.
Lastly, if you browse the site with javascript turned off you can see something odd in that the initial page /elektronik/baerbar-lyd/hoeretelefon has no prev/next OR canonical tag but has a link to /elektronik/baerbar-lyd/hoeretelefon?page=1 on which you find prev/next and canonicals back to the non paginated version. So you are basically skipping the pagination setup that goes from the original to the page=1 (but also giving a canonical back to the original page).
Phew! It is a bit confusing. I would recommend deciding on if you want to go with prev/next or canonical in the first place and take it from there. I would think that if you have the ability to canonical to a 'see all products page' then this might be the best way to go since it should theoretically take care of any issues the dynamic loading is causing also.
Hope that helps!
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
-
1 page crawled - again
Just had to let you know that it happend again. So right now we are at 2 out of the last 4 crawls. Uptime here is 99,8% for the last 30 days, with a small downtime due to an update process at the 18/5 from around 2:30 to 4:30 GMT In relation to: http://moz.com/community/q/1-page-crawled-and-other-errors
Moz Pro | | alsvik0 -
Question #3) My last question has to do with Some SEOmoz crawl diagnostics -
I recently fixed (or well, I am asking to make sure that this was the right thing to do in my first question posted a few minutes ago), a problem where all of my internal main sidebar category pages were linking using https://, which to my knowledge means SECURE pages. anyways, OSE, and google seem to be not recognizing the link juice. but my rank fell for one of my main keywords by 2 positions about a week after i made the fix to have the pages be indexable. Making my pages properly linked can't be a bad thing right? That's what I said. So I looked deeper, and my crawl diagnostics reports showed a MASSIVE reduction in warnings (about 3,000 301 redirects were removed by changing the https:// to http:// because all the secure pages were re-directing to http:// regular structure) and an INCREASE, in Duplicate Page Titles, and Temporary redirects... Could that have been the reason the rank dropped? I think I am going to fix all the Duplicate Page Title problems tonight, but still, I am a little confused as to why such a major fix didn't help and appeared to hurt me. I feel like it hurt the rank, not because of what I did, but because what I did caused a few extra re-directs, and opened the doors for the search engine to discover more pages that had problems (which could have triggered an algo that says hey, these people have to much duplicate problems) Any thoughts will be GREATLY appreciated thumbed, thanked, and marked as best answers! Thanks in advance for your time, Tyler A.
Moz Pro | | TylerAbernethy0 -
What user agent is used by SEOMOZ crawler?
We have a pretty tight robots.txt file in place to only allow the major search engines. I do not want to block SEOMOZ.ORG from being able to crawl the site so I want to make sure the user agent is open.
Moz Pro | | eseider0 -
How reliable is SEOMOZ data?
I have been using the Mozbar data to locate Auth websites, basically the higher the DA/PA the better it is. If I then check the site out and they have good load times, lots of social sharing etc this makes it extra special and now a good site to try and get a link on. I found 3 blogs to guest post on which all have DA/PA 80+ some of the blogs have a PR7. Now this is great but what scares me is have I just wasted 3 guest posts because the data is incorrect. These blogs are run by 1 person and the kind of work involved to achieve those stats would be very heavy. Maybe this blogger has just hit the nail on the head with SEO and link building. Input anyone?
Moz Pro | | activitysuper0 -
Did anyone else see "Rel Canonical" drop to zero after their latest SEOmoz crawl?
In the Crawl Diagnostics section of the SEOmoz reports, we get errors in red, warnings in yellow, and notices in blue. After my latest crawl, I saw the "Rel Canonical" part go from about 300 down to 0. Obviously, this isn't right, so I'm wondering if this is a bug that everyone is experiencing. U9W5I
Moz Pro | | UnderRugSwept0 -
How can i get seomoz to crawl a campaign on demand
hi how can i get seomoz to crawl a campaign on demand instead of on a weekly basis? For example i have corrected some error warnings and on page elements and would like it to re crawl the site sooner to see how the corrections have worked? thanks
Moz Pro | | Bristolweb0 -
Can you set-up a manual SEOmoz crawl?
I received a crawl report yesterday, made some site changes, and would like to see if those changes were done correctly. Rather than wait a week for my automatic crawl to be generated, is there anyway to initiate a manual crawl on a single subdomain as a PRO member? As a PRO member, you can schedule crawls for 2 subdomains every 24 hours, and you'll get up to 3,000 pages crawled per subdomain. When we've finished crawling, your reports will be sent to your PRO email address, which is currently From here... http://pro.seomoz.org/tools/crawl-test
Moz Pro | | ICM0 -
SEomoz slow to crawl?
Hello - I am just trying out the trial and it said the next crawl was nov 1st but I see no change in any of the errors since the initial crawl... so just waiting to find out if what I changed was fixed or not. Is this normal ?
Moz Pro | | Bethany_BabyBrowns0