In the four years that I’ve been working in the SEO industry, the vast majority of the websites I’ve worked on have been ecommerce, so I wanted to share some of the things I’ve learnt over this time.
There are a host of common duplicate content issues with ecommerce websites, most of which are caused by types of navigation. Here are some of the issues that I’ve seen a lot with clients and websites I’ve worked on.
Faceted navigation / dynamic filter pages:
Faceted navigation is widely known for being a bit of a nightmare for SEOs, as it results in lots of duplicate variations of category pages that need to be kept out of the index.
My preferred method of getting around this issue is to apply meta robots rules to pages with set parameters appended to the end of their URLs, so for example website.com/clothing?brand=clothingbrand. I would always recommend using noindex,follow meta robots tags, as this tells search engines not to index the page but also to follow the links and continue crawling the website.
The meta robots tag should be set in the <head> area and most developers should be able to write the rules.
Lots of people tend to miss filter pages that set the order of results for category pages, however these pages still feature duplicate content. I would include rules to add the meta robots tags to these pages as well.
Some people tend to prefer using the canonical tag or parameter settings in GWT, but from previous experience I’ve had much better results using meta robots tags.
Alternatively, if you have a good development team, you could look to use AJAX to filter the results, this would allow you to use one static page rather than having filter queries appended to the given URL.
My favourite option (by far) is to turn dynamic filter pages into static pages – this gives you lots of new landing pages (that you can optimise for longer tail queries) to play with and eliminates the duplicate content issue. I would only recommend this to those with strong development resource behind them, as it’s really important to do it in a way that doesn’t compromise your site’s usability.
Trailing slash canonical issue:
Lots of big and small ecommerce platforms (and bespoke systems) have issues with trailing slash pages being, and SEO’s then have the issue of these pages being seen by Google (and other search engines) as duplicate content.
With Magento, the platform that we use most, it’s proved extremely difficult to eliminate this issue as rewrite rules have caused complications with the back-end of the website. So, if this is the case with the platform that you’re using, I would recommend using the rel canonical tag to point one version to the preferred one.
Last year Google announced the introduction of the rel next, prev and all tags, which are designed to prevent duplicate content issues with pagination. Find out more about implementing these tags here.
Catalog Search Pages:
Catalog search pages often get indexed by search engines due to them obtaining internal or external links. These pages don’t feature any unique content and shouldn’t be made accessible to search engines.
In order to prevent catalog pages from being indexed, simply disallow the directory (the majority of search pages are structured in this way) eg: /catalogsearch/ in the robots.txt file and then remove the directory in Google Webmaster Tools.
Removing duplicate pages:
Once you’ve used the robots.txt file or meta robots rules to tell search engines not to index the pages, you will have met the requirements to submit removal requests in Google Webmaster Tools. Submitting manual removal requests for dynamic pages can be very frustrating, however it’s by far the fastest way to get pages out of the Google (or other search engines) index.
Although I’m dubious about how much site speed affects your rankings it’s clearly a fundamental part of user experience. For ecommerce, having a fast website is vital as users who are forced to wait for results often leave prematurely.
Companies like Zappos and Amazon have spent huge amounts of money on optimising the performance of their websites and have seen dramatic improvements in their conversion rates as a result.
This article from Wordstream outlines the importance of speed for desktop and mobile website experiences.
There are hundreds of things that can be done to improve the performance of a website, but here are some that we’ve focused on recently for our clients.
There are a number of arguments for using top-level products URLs, these are:
I always have both html and xml sitemaps on ecommerce websites as standard, but I’d also recommend having video sitemaps and category sitemap pages.
Category-specific sitemap pages:
I always recommend that larger ecommerce websites have category-specific html sitemaps for all content related to the category (blog content, products and sub categories). In addition to indexation benefits, this is also great for using navigating around the website.
I work with a lot of fashion companies, so we tend to have individual html sitemaps for the different brands featured on the website – these pages will then feature all products, collection pages and blog content for the given brand.
There are two main benefits of using video sitemaps, getting featured in video results and getting video markup for listings. For ecommerce websites with product video content I would always recommend having a video sitemap.
Having the video snippet on your listings helps to increase click-through on the given page as the page stands out a lot more than the other listings for the query.
Rich snippets have been revolutionary for ecommerce ever since Google started reading and using structured data. By adding the schema markup to different pages on your website you will be able to display rich snippets that will help to install trust for users and improve click-through rate.
The video markup is a less obvious addition to an ecommerce listing, however it’s great for increasing click-through rate. I would recommend combining the video markup with the price and availability schema as this makes it clearer that it’s a product listing.
The rating schema is probably the most commonly used schema and it has been proven to help improve click-through rate for sites in lots of different verticals. If you don’t have the development resource to help you build this into your review system, I’d recommend using Feefo or TrustPilot to generate these rich snippets.
At SearchLove, Guy Levine also talked about a new schema that is soon to be recognised by Google, a green tick that will be displayed next to trusted websites. You can find out more about implementing schema /microdata on your website here.
For ecommerce websites, meta content is about much more than just keyword targeting, it’s about providing an enticing and accurate proposition for your products / offering.
It’s vital that your off-site proposition is compelling and enticing as otherwise searchers will click on a competitor’s listing, so utilising schema markup (as I mentioned earlier in the article)
Good meta description:
Bad meta description:
With product and brand-specific terms, you’re likely to be competing against other retailers who are selling the same product – so outlining USPs and having obvious call to actions is key.
If you have any questions about any of the points I’ve discussed, please feel free to ask them in the comments field below.