Onsite SEO for ecommerce

Posted by Rob Bradburn on 03 March 2014

Ecommerce sites typically have a lot of content with potentially 1000’s of product pages and dozens of category pages. You need search engines to efficiently read your content and serve the correct pages for your target search terms.

This article aims to provide a comprehensive list of SEO recommendations for your ecommerce site to help your customers find you.

Stay up to date with SEO ranking factors

With search engines now releasing literally 100’s of algorithm changes each year, it is important to stay up to date with best practices for onsite SEO. Certain techniques that may have worked well a couple of years ago could potentially be harming your visibility in the search engines today.

Every 2 year’s Moz surveys 120 of the world’s top search marketers to find out which web page characteristics they believe correlate with higher search engine rankings. The data returned provides SEO’s with a good steer on the perceived level of importance of each of the 80+ ranking factors. As the latest survey reveals, onsite SEO ranking factors are still considered very important - 2013 Search Engine Ranking Factors.

When was the last time you analysed your onsite SEO? If you haven’t performed an SEO audit in a while, it is about time you did! Before following any of the tips in this article, make sure your keyword research is up to date!

Ecommerce Platform intricacies

The majority of our clients use the Magento ecommerce platform, which out of the box we find is a little more ‘search-engine ready’ than other ecommerce solutions on the market - however it isn't without its quirks! There is a useful guide to tweaking the Magento configuration for SEO here.

As Magento specialists, GPMD wanted a way to provide our clients with more control of the SEO on their Magento installation, so we developed the MageSEO extension - this may be a shameless plug, but if you are running a Magento site I highly recommend installing it!

Search is evolving, and so are the ecommerce platforms, so I suggest performing a google search to investigate the intricacies and quirks of the particular platform and version number you are running.

The SEO principles and techniques outlined in this article are the same no matter what ecommerce platform you are running on your site. 

Site Architecture

An effective site structure will ensure the search engines can index your pages easily, but should also help users intuitively navigate the site.

Trond Lyngbø made a great point in his article ‘8 e-commerce SEO tips’, which is to make sure SEO is considered from the very beginning of a site’s development process. This will ensure the information architecture of the site is fully optimised from the ground up - it can require a lot of work to optimise a site’s structure after the design has been finalised, the product catalogue has been uploaded and the site is live!

Flat site architecture

Keep the architecture of your site as flat as possible, you never want visitors to your site more than 3 clicks away from the content or products they are looking for. A simple ecommerce site structure should look like this:

ecommerce site architecture

site architecture image from Kissmetrics blog 

Most of your external links will point at your home page. This ‘link equity’ (I find the term ‘link juice’ a bit passé) trickles down through the site to the rest of your pages. Typically, the further away a page is from your home page, the less authority it is given and the less likely it will be to feature in search results.

Internal linking structure

Your link equity flows through the site using your internal links. Like the site architecture, the internal linking structure should be kept as flat as possible:

Home page > linking to > Category pages > linking to > Sub category pages > linking to > Product pages

Menus

Whilst I really like the use of a well styled mega drop-down menu, I don't suggest filling menus with links to unimportant areas of the site - ideally you want to ensure your link equity only flows through to the important pages of the site.

Clean URL Structure

People like clean URLs. We recommend keeping URLs free from query strings and as simple as possible. Whilst the use of keywords in URLs is no longer considered a serious ranking factor, we find clean URLs containing keywords can encourage more clicks in the search results. Which would you prefer to click on if you were searching for brown shoes?

http://www.exampleshop.com/index.php/products?category=4&&id=42/
or
http://www.exampleshop.com/shoes/brown

Some simple rules we follow:

  • Only use keywords in URLs if it appears natural - no stuffing!
  • Keep the URLs easy to remember
  • For consistency, we prefer to use the www. prefix (like Google does!)
  • Be careful of URL duplication issues (I will go in that later in this article)

On-page SEO

There are a number of factors to consider when optimising your on-page elements. To get an idea of which areas of your page need improving there are a number tools you can use, we like On-Page Grader (you need to be a Moz subscriber) and Seoptimer. They will analyse your page and make suggestions on your keyword usage in places such as meta data, headers, image alt attributes and internal links.

Meta Content

Title tags

Keyword usage in page titles are still an important factor in search rankings. Search engines use them to associate the page with a topic and/or set of terms.

Use keyword research to determine the titles for your category and product pages. Optimise each title using the primary and in some cases secondary keyword.

We recommend the following format for page titles:

<primary keyword> | <sitename>

example: Magento Development | GPMD London

Meta Descriptions

Despite not directly impacting organic search engine rankings, meta descriptions play an important role in SEO as they sell the page and encourage users to click through from search results.

A good meta description can help to significantly improve click-through rate, especially for very specific terms. It is worth allocating some time to adding compelling meta descriptions to all of the pages on your site, including top-level category pages, subcategory pages and product pages.

Meta description lengths should be somewhere between 140-160 characters. Ensure the text is engaging and includes a call to action - adding selling points such as “fast shipping available” will help to improve the clickthrough.

Writing unique meta descriptions for your main pages won’t take you long, but where you have lots of product pages I would recommend using templated meta descriptions, for example:

The <Product Name> is a <short product description>. Available to order online now with fast international shipping.

To make your listings stand out even more we would recommend utilising rich snippets. You can read more about this later in the article.

Content Optimisations

Broad keyword usage

Your keyword research will have determined your content themes (read more about content themes in this post) and the target key terms. Your target keywords should be contained within the body of the page - Aim to mention the primary keyword at least once in the content, and include between 2-4 other targeted keywords.

This is not only for search engine benefit, users expect to find relevant keywords within the content, but it is important to ensure the content looks natural and keywords are not stuffed into the content.

Category landing pages

Organic visitors to your site should land on the page that is most relevant to their search query. Think of category pages as unique home pages. By optimising your category pages, you can drive relevant traffic directly on to them from the search engines.

Your keyword research will have identified ranking opportunities and themes for your category pages. Populate the metadata and page content with your target key terms - ensuring the content is useful to the visitor. Aim for over 200 words of unique content on each category page, but make sure that the products don’t get pushed too far down the page and can be seen above the fold.

Sub-category pages

Site’s with 100s of products within a particular category will need to use sub category pages to avoid user ambiguity. Sub-category pages provide additional opportunities to build landing pages to drive valuable organic traffic from search queries. Again, use your keyword themes to inform the content, helping the site to rank for very specific long-tail search queries and drive highly relevant users.

Rich snippets markup

Rich snippets can be added to the code on your site to tell Google what type of content you are using on your site so it can be displayed better within search results. The use of rich snippets can help to improve clickthrough rate.

Typically markup for ecommerce sites should include price and ratings tags. if you are publishing your own content through the site or blog, use rel author or publisher tags. See schema.org for more details.

Duplicate Issues

Search engines do not like duplicate content, and Ecommerce sites are notorious for having duplication issues. Part of the recent Google Panda updates were aimed aim at penalising duplicate content and pages without quality content. There are lots of ways that duplicate content can occur, both intentionally and unintentionally - you only want

search engines to read the pages that have quality, unique content.

Ecommerce expert Paul Rogers has written a comprehensive article about duplicate content issues with ecommerce sites that can be read on the econsultancy blog. I will whizz through them below:

Faceted navigation on category pages

Using layered navigation filters on an ecommerce site is a useful way for customers to drill down on the product they want to buy (e.g. by colour or size). Unfortunately the search engines may crawl the site and find every single filter variant of the same category page - which could result in them reading 100’s of pages with exactly the same content, e.g:

http://www.example.co.uk/shoes/?girls&size=36&colour=23
or
http://www.example.co.uk/shoes/?girls&size=37&colour=21

Ideally you only want the main page to be indexed: http://www.example.co.uk/shoes

The search engines will only give you a certain amount of crawl time, and by wasting time reading multiple instances of duplicate pages it may leave the site before indexing your more important content.

How can you prevent these filter pages being seen by the search engines?

There are numerous ways to prevent search engines accessing your unnecessary pages, and you can read about them in this recent article on the Google Webmaster Blog. These are our recommended options:

  • Tell Google the most important page in a cluster using the rel=”canonical” tag in the page header, and only include canonical URLs in the sitemap
  • Configure Google webmaster tools to specify how Google handles the various URL parameters
  • Use a “noindex,nofollow” meta robots tags on all unnecessary filter pages
  • Block URLs with unnecessary filter parameters using a /filtering/ directory within robots.txt

Duplicate product pages

This can happen when an identical product page appears within multiple categories. This occurs when you use hierarchical URLs, for example:

http://www.example.co.uk/boys/black-sunglasses.html
http://www.example.co.uk/accessories/black-sunglasses.html

You don’t want duplicate product pages competing against each other, so to avoid duplication issues like this you can use canonical tags to designate a single version of the page, but I would recommend moving all products to a top-level URL (use a re-write rule for existing product pages). You could also keep all of products within a /products/ directory so they can be easily tracked using analytics.

Thin or nearly identical content

Try to avoid product, category or sub-category pages with either very little content, or content that is nearly identical to another page on your site. Wherever possible create unique descriptions for each page, and avoid creating category pages with very few products listed.

There may be instances where thin or near identical content is unavoidable, if this is the case use the “noindex, follow” meta robots tag.

What about out of stock products?

Leave product pages for out of stock items live - There is no point losing the beneficial search equity, plus you can use the page to offer visitors alternative products or options to come back and buy later

If you no longer stock a product, and there is no informational value in keeping the page live, then you should 301 redirect to the newer product or relevant category page

Pagination

When browsing an ecommerce site, I personally prefer to view all of the products on a long page and scroll down or filter through them. However, this does impact page load time and sometimes paginated content is necessary.

To avoid duplication issues caused by pagination. we recommend using the canonical tag and the rel=”prev” and rel=”next” pagination markup. Read more about this best practice solution on the Google Webmaster Blog.

Valueless Pages

As well as dynamic filter pages, Ecommerce sites can generate a lot of other irrelevant dynamic pages that may get picked up by the search engines causing poor quality content issues and wasting crawl time. These could include secure https pages, internal search results, customer session logs or shopping basket pages.

Use the robots.txt to disallow the relevant directories, paths and files from being crawled by the search engines. Each ecommerce platform will have a number of recommended exclusions - this article by Toni Anicic contains useful examples for Magento sites.

Trailing slashes

These two URLs may look nearly identical, but search engines may see them as different web addresses due to the trailing slash at the end:

http://www.example.co.uk/shoes
http://www.example.co.uk/shoes/

We recommend using a 301 redirect to the preferred version, personally I prefer no trailing slash.

www vs non-www URLs

Google will see both the www and non-www versions of the domain as different web addresses unless you tell it otherwise. As I mentioned earlier, here at GPMD we prefer to use www. URLs just like Google does. You can tell Google your domain preference within webmaster tools. You also want to 301 redirect the version you are not using to your prefered version.

Check the staging server is not being indexed

If you have recently launched a site, it is worth double checking that the staging version of the site is not being indexed. Ensure the robots.txt is using the "Disallow: /" protocol.

Technical Considerations

Google wants reading your site to be as efficient as possible. It is necessary to regularly get ‘under the bonnet’ and check the site indexing is going smoothly behind the scenes.

Crawl Errors

Keep an eye on messages from Google Webmaster Tools with relation to server response codes. You want to minimise the number of crawl errors that are logged by the Googlebot. If you have a large number of 404 or 504 errors appearing, you will need to get these resolved.

Crawl the site using a tool such as Screaming Frog, identify the pages that are returning status codes that are not 200 or 301 and remove or redirect to an appropriate URL.

Make sure that your site can handle 404 errors in the correct way and serves a 404 response code when an inactive page is visited.

XML Sitemap

Most ecommerce sites are capable of generating an automated XML sitemap. Ensure only canonical URLs are included, and submit it to Google using Webmaster Tools, periodically checking that it is being processed regularly.

Use of Javascript and Ajax

Navigate your site with javascript disabled on your browser and look for any issues with content not appearing. Whilst search engines can read most Javascript and Ajax, sometimes the way the code has been implemented can prevent them from reading the content.

Performance

Ecommerce sites can often become bloated and slow. Different tools exist such as Pingdom to monitor and diagnose speed issues with the site. Whilst performance is an SEO ranking factor, there will only be a negative impact if the site is very slow and provides a poor user experience.

Redirects

Ensure there are no temporary 302 redirects or redirect loops/chains. Use a website crawling tool to identify any issues - Screaming Frog has a useful report that can be run to identify redirects, and the ‘hops’ taken along the way.

 

You can follow Rob Bradburn on Twitter or Google Plus.

Comments

No one has commented on this page yet.

Post your comment