How To Implement Pagination Properly? Important and Urgent!
-
I have seen many instructions but I am still uncertain. Here is the situation
We will be implementing rel prev rel next on our paginted pages.
The question is:
- Do we implement self referencing canonical URL on the main page and each paginated page?
- Do we implement noindex/follow meta robots tag on each paginated page?
- Do we include the canonical URL for each paginated page in the sitemap if we do not add the meta robots tag?
- We have a view all but will not be using it due to page load capabilities...what do we do with the viewl all URL? Do we add meta robots to it?
- For website search results pages containing pagination should we just put a noindex/follow meta robots tag on them?
- We have seperate mobile URL's that also contain pagination. Do we need to consider these pages as a seperate pagination project? We already canonical all the mobile URL's to the main page of the desktop URL.
Thanks!
-
Hello SEO32,
I apologize for the delayed response. There are several good questions here. They're also complicated questions, which don't really always have a single "correct" answer. So much revolves around the specific situation, and without seeing your website it is difficult to say what is best for you. Also, much of what we think we know about this kind of stuff is either based on what Google tells us (which isn't always the truth) and what we've observed and deduced from our own experiences (which aren't always the same). True "testing" of this stuff one variable at a time isn't always possible so we rely on best practices and our own experience.
That said, I will attempt to answer your questions with what I would probably do in most situations, including links to more information when possible.
Do we implement self referencing canonical URL on the main page and each paginated page?
Here's what Rand says, and he's probably seen way more data than I have:
"Whatever you do, DO NOT:
- Put a rel=canonical directive on paginated results pointing back to the top page in an attempt to flow link juice to that URL. You'll either misdirect the engines into thinking you have only a single page of results or convince them that your directives aren't worth following (as they find clearly unique content on those pages).
- Add nofollow to the paginated links on the results pages. This tells the engines not to flow link juice/votes/authority down into the results pages that desperately need those votes to help them get indexed and pass value to the deeper pages.
- Create a conditional redirect so that when search engines request paginated results, they 301 redirect or meta refresh back to the top page of results."
Keep in mind that post is from 2010, and I think before Google said a "View All" canonical <a>was their preference</a>.
I have seen plenty of sites do well ranking the canonical category page, and with indexing most of the product pages, while all paginated pages had a rel canonical that referenced the first page in the series (i.e. .com/category/ or .com/categry1/category2/). It probably helps that they had good XML sitemaps for product pages, and plenty of internal linking, unique content on category pages, etc.
I have also seen sites do well using rel next/prev without rel canonical, or rel next/prev with self-referencing canonicals on paginated category pages.
I think where you run into problems is when you also allow the facet/filter/sort versions to have self-referencing rel canonical tags.
Here is what I advise in most cases:
Use rel next/prev (not because I think it works, but because Google says to and I don't think it hurts) along with self-referencing rel canonical tags, and "follow,noindex" robots meta tags on paginated pages.
Always include a followable link to the first page in the series from every subsequent page. For example:
<previous>first...1...25...26...27...last...</previous>
I recommend always having a first and last page link. The first is obvious because it means pagerank is going to flow into it from every other page in the set, giving it the most internal links of all. The last is more of a crawlability and usability thing. For users it helps us figure out how much further we have to go. It does the same thing for search engines. Instead of blindly following a path that may or may not have an end, a message is sent that tells a spider how much further it has to go. I don't know if Google takes advantage of that signal or not, but it just makes sense to include it. If you want to get fancy you can try making the 'last' link flash or javascript or something so it doesn't pass (as much?) page rank.
The category root pages usually have links from site-wide navigation, unlike the paginated versions, which further establishes it as the page that should be ranked highest.
Make sure the first page in each series is indexable, and has content that does not appear on the paginated versions. Also, make sure that ?p=1 doesn't have a self-referencing canonical tag, but references the root page for that series (e.g. /category1/category2/).
All subsequent variations (e.g. color, size) should rel canonical back to their root page. For example:
/category1/category2/?page=2&size=s&color=blue would have the following URL in the rel canonical tag:
/category1/category2/?page=2
Which happens to be followable, but not-indexable, and has a self-referencing rel canonical tag.In this way you give search engines a strong signal about which URL in the whole set is the strongest (i.e. /category1/category2/) because it is indexable, has its own content, has the most internal and external links, is the simplest version of this URL pattern, and is at the root of the directory. You're telling search engines which page is next in the series, and that this page is first in the series. You're telling search engines which page is last in the series, as well. Google usually does an awesome job figuring it out from there. There are always exceptions.
Do we implement noindex/follow meta robots tag on each paginated page?
I would. Consider this from Google's perspective, or from that of a searcher. Someone types "Blue Flower Dress" into Google. Is the best page to return a deep category page full of blue dresses, one of which happens to have flowers? Or would it be the Blue Flower Dress product page? I can't think of any reason why I would want to land on page 3, where what I'm looking for is listed among dozens of other things, when I could just go straight to the thing I'm looking for.
Likewise, if someone searches for "Blue Dresses" is the best page /dresses/blue/?page=3 (paginated page in the Blue Dresses category), OR /dresses/blue/ (the very first page of the Blue Dresses category), which also has useful content about blue dresses?
Long story short, when it comes to transactional eCommerce queries, they're usually either looking for a product page or the first page of a specific category or sub-category. Or sometimes the home page. Therefore, I don't see any reason for allowing paginated URLs to be indexable in most cases. Non-transactional eCommerce content is different (e.g. buying guides, comparison charts, reviews...) but I still wouldn't allow paginated pages to be indexed in most cases.
Slightly Off Topic - Filters/Facets/Sorts
Or perhaps the category is "casual dresses" and "blue" is specified in the "color" attribute. In this case, would the best page be /dresses/casual/?color=blue , /dresses/casual/ or /dresses/casual/?color=blue&page=4 for someone who Googled "blue dresses"? I've bolded the one I'd prefer as a searcher.
Here again, as with the internal search results, there is an opportunity to use real data to inform your decision. Pay attention to the facet/filter/sort URLs most accessed by shoppers and consider turning those into category or collections pages with their own URL pattern (e.g. /dresses/casual/blue/). One example I come across all the time is when "Brand" is a filter instead of its own limb in the category structure. If people are shopping by brand, as they do with most consumer products, then you should have a brand subcategory under each major top-level category. If I search for Levi Jeans Google doesn't want to send me to a "pants" page where I have to set a filter to see only Levis. I should go to pants/brand/levi/ . If I Google Chefmate Pots I want to see cookware/pots/brands/chefmate so I don't have to set a filter after I get there.
This doesn't mean all filter pages should be turned into category pages either. Use your best judgement based on the pages most of your users are accessing from the navigation and filters.
Do we include the canonical URL for each paginated page in the sitemap if we do not add the meta robots tag?
I would add the robots meta tag. Please let me know if I've misunderstood the question.
We have a view all but will not be using it due to page load capabilities...what do we do with the view-all URL? Do we add meta robots to it?
I would add a meta robots "index,nofollow" tag, and would also use the canonical page's URL (e.g. /category1/category2) in the rel canonical tag.
For website search results pages containing pagination should we just put a noindex/follow meta robots tag on them?
This is one of those situations involving crawl budget potentially being eaten up by an infinite amount of pages. I would consider blocking the internal search result URLs in the robots.txt file. They are of no use to Google, as they consider a search engine returning search results with links to more search results somewhere else a bad user experience. This is also what Google recommends in their Webmaster Guidelines:
"Use robots.txt to prevent crawling of search results pages or other auto-generated pages that don't add much value for users coming from search engines."
However, I would also make use of those pages internally. Rather than relying on a search result page for things people often look for, track what is being searched for and create static, indexable pages. For example, try "Collections" pages on eCommerce sites, as well as FAQ pages, or "Industries" or "Use Case"-type pages on lead generation sites. This is a much better user experience for someone arriving on that page from a search engine.
We have separate mobile URL's that also contain pagination. Do we need to consider these pages as a separate pagination project? We already canonical all the mobile URL's to the main page of the desktop URL.
I think you should if that's the way you're handling it. Here is a post I did on mobile best practices. It covers some other options. I would also add a rel=”alternate” tag in the HTML header of the desktop page, which alerts search engines to the corresponding mobile URL and helps define the relationship between the two pages.
The bottom line for me is to always think about what would be the best experience for someone searching from Google for something, and to try and use all of the various technical options to ensure that is the page I'm telling Google they should rank for that query, or those types of queries. The 'best practice' changes, depending on the situation.
I hope others will join the discussion with their own experiences and findings.
-
Hi,
Please check this moz article on this @ SEO Guide to Google Webmaster Recommendations for Pagination
Hope this helps you.
Thanks
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
-
Pagination dilemma
We are about to migrate an ecommerce site to a whole different platform and we are facing a pagination dilemma. The new platform has a view all page for each category where all the individual pages point with a rel canonical to the view all page. What would be best to to as for the redirect. should we do the 301 redirect to the view all page ? Considering that the view all page takes quite a time to load, wouldn’t it be good to drop the view all an implement the rel-next, rel-prev and point the 301 redirect to the first page. If we do the view all page, wouldn't be a problem if all the natural backlinks that we'll get will not be for the page view all page, but for the page1, page2 etc?
Intermediate & Advanced SEO | | lvt0 -
I have implemented rel = "next" and rel = "prev" but google console is picking up pages as being duplicate. Can anyone tell me what is going on?
I have implemented rel="next" and rel = "prev" across our site but google console is picking it up as duplications. Also individual pages show up in search result too. Here is an example linkhttp://www.empowher.com/mental-health/content/sizeismweightism-how-cope-it-and-how-it-affects-mental-healthhttp://www.empowher.com/mental-health/content/sizeismweightism-how-cope-it-and-how-it-affects-mental-health?page=0,3The second link shows up as duplicate. What can i do to fix this issue?
Intermediate & Advanced SEO | | akih0 -
Ecommerce SEO: Title Tags for pagination
Here's a specific question about title tags for ecommerce website... We've got lists of products (category list pages) that stretch for many pages... is there any benefit to added a something to make the title tag unique. For example: Page 1: <title></span><span class="html-tag" data-mce-mark="1">Category List Page Example</span><span class="html-tag" data-mce-mark="1"></title> Page 2: <title></span><span class="html-tag" data-mce-mark="1">Category List Page Example - Page 2</span><span class="html-tag" data-mce-mark="1"></title> Page 3: <title></span><span class="html-tag" data-mce-mark="1">Category List Page Example - Page 3</span><span class="html-tag" data-mce-mark="1"></title> FWIW, we've got the pagination and canonicalization nailed down tight. Moz crawl actual brought a dupe content issue based on title tags.
Intermediate & Advanced SEO | | 19prince0 -
Importance of minimal markup on a page
Moz.com's 2013 SEO study gave some value to "Total # of Characters in the HTML Code" (https://moz.com/search-ranking-factors/2013). Is having minimal HTML on a page still a ranking factor, even if not a huge one?
Intermediate & Advanced SEO | | A2890 -
Is it a problem that Google's index shows paginated page urls, even with canonical tags in place?
Since Google shows more pages indexed than makes sense, I used Google's API and some other means to get everything Google has in its index for a site I'm working on. The results bring up a couple of oddities. It shows a lot of urls to the same page, but with different tracking code.The url with tracking code always follows a question mark and could look like: http://www.MozExampleURL.com?tracking-example http://www.MozExampleURL.com?another-tracking-examle http://www.MozExampleURL.com?tracking-example-3 etc So, the only thing that distinguishes one url from the next is a tracking url. On these pages, canonical tags are in place as: <link rel="canonical<a class="attribute-value">l</a>" href="http://www.MozExampleURL.com" /> So, why does the index have urls that are only different in terms of tracking urls? I would think it would ignore everything, starting with the question mark. The index also shows paginated pages. I would think it should show the one canonical url and leave it at that. Is this a problem about which something should be done? Best... Darcy
Intermediate & Advanced SEO | | 945010 -
Proper Form for Title & Description Tags
Greetings MOZ Community: I operate a real estate web site in New York (www.nyc-officespace-leader.com) that I suspect has been hit by Panda 4.0. I believe a problem is thin content on product pages, which in my case are 350 listing pages. However I am also looking at how title and description tags are formatted for these 350 pages to ensure this is not a factor in the ranking drop. The title descriptions are written like this: <title></span><span class="webkit-html-tag">Flatiron loft for rent | West 21st Street | 1441SF $6604/month</span><span class="webkit-html-tag"></title> Is this sufficiently diverse? Will constantly repeating various street names, square footages and prices work against me? Will Google in a sense consider this thin or repetitive content? It does provide the visitor with key information. The descriptions meta tags are written along these lines: description" content="One of the most desirable full floor sublets in Midtown South. Recent build out, pristine condition, panoramic views, tech chic, spectacular. Top location." /><meta< span=""></meta<> From an SEO perspective are these critical tags written the way they should be? Thanks everyone!! Alan
Intermediate & Advanced SEO | | Kingalan10 -
Pagination, Canonical, Prev & Next
Hello All
Intermediate & Advanced SEO | | Vitalized
I have a question about my Magento setup. I have lots of categories which have many products so the categories paginate. I've seen info about making sure the Canonical tag doesn't simply send Search Engines back to the first page meaning the paginated pages won't get indexed. I've also seen info about using the rel=next & rel=prev to help Search Engines understand the category pages are paginated... Is it okay to use both? I've made sure that: category/?p=1 has a canonical of category/ to make sure there isn't duplicate content. Here's an example of category/?p=2 meta data:
http://website.com/category/?p=2" />
http://website.com/category/" />
http://website.com/category/?p=3" />0 -
How important is it to canonicalize mobile URLs to desktop URLs?
I know many SEO's prefer a stylesheet and single URL, but if you use m.domain.com, do you canonicalize to your desktop URLS?
Intermediate & Advanced SEO | | nicole.healthline0