Whether you’re a weathered SEO professional moving onto a new challenge; an in-house product manager interested in where SEO intersects with your ecommerce strategy; or a smaller brand looking to compete in the big scheme of things, this article is for you. We have put together a must-know, technical SEO guide to online ecommerce, with a slight focus on the fashion industry.  

We'll start with some questions that need to be asked for us to gain a clear picture of your site and technology.

What is ecommerce SEO?

Search Engine Optimisation (SEO) is the practice of gaining visibility for your site on search engines. For an ecommerce website, this differs from traditional SEO because it deals with inventory management practices and more technical concepts such as JavaScript frameworks that don’t often appear in SEO for content-focused websites.

Ecommerce sites can be very big because of all of the product filtering functionality available. They can also be too small as they are usually single-page applications, meaning much of the on-page content is updated without reloading the page.  Technically these websites can function on a single URL; unfortunately, SEO requires that every keyword has its own maintained URL.  

Is SEO important for ecommerce websites?

A 2018 SEO survey found that 35% of product searches happen on Google, making search engines one of the single largest channels to boost your on-site traffic and revenue. Your business will not have to commit any ad spend towards this traffic either and done right rankings gained can last years.

After months of testing, this fashion ecommerce client reached page 1

Here are 5 tips for your ecommerce SEO strategy:

Tip 1: Discover JavaScript Limitations 

Tip 2: Manage product inventory for SEO

Tip 3: Avoid Index Bloat

Tip 4: Realise SEO isn’t only on-site

Tip 5: Monitor your keyword split

Tip 1: Discover if JavaScript is limiting the site

The first step on your journey to ranking domination is to ensure that search engine crawlers can access your content.  As websites and content management systems have expanded features, search engine crawlers have not always been able to keep up.  

Googlebot can struggle with ecommerce websites because of the rich functionality they attempt to offer. Either the search engine won’t be able to render the content because it is too complex, or it will choose not to due to the processing power rendering at scale demands. Search Engine crawlers don't have the resource to render the JavaScript of every page on the web.

Switch JavaScript off and toggle CSS in your browser and explore your site, do you see any of the following:

Missing content 

Oliverbonas.com with JS Disabled

If navigation or images are not visible products and pages may not be discoverable. In this instance, the product offering is not clear and there is no left-hand navigation to further pages. With products and sub-category Product Listing Pages (PLPs) being ferried in by JavaScript it is very likely that many of these products are not indexed unless they are placed in the XML Sitemap. Even so, their importance in the site hierarchy may not be clear and so they will not rank successfully. A quick check of the View Source also reveals that no content is visible in there at all, reiterating that the site contains no content unless JavaScript is enabled, including links to navigate to other pages. 

Oliverbonas.com’s source code contains no content

Another common ecommerce issue is that banner content on PLPs is added to the page last with JavaScript. If this happens too late, or is not done with SEO in mind, then search engines may never see this content, negatively impacting rankings.  

CTAs are not visible due to their reliance on JS

Caveat:

If you find an example like this be sure to toggle CSS as well - if the content then appears the page should be accessible to SEO crawlers. Some website builds require both JS and CSS to show content on the page, by switching them off you see a version of the page that’s closer to what a crawler might see. 

Missing on-page links

Even if filters and side navigation are visible with JavaScript turned off, they may not be internally linked. Crawlers can’t click so JavaScript on-click events and other functionality that doesn’t use links may result in pages being orphaned and not appearing to search engines. If there are no a href tags, crawlers probably can’t find those pages. 

Boohoo.com’s side navigation is not accessible to crawlers

If crawlers cannot find a page it will not be indexed. If the page can be found but there are no links to it from other pages on-site (often referred to as an orphan page) then that page may be indexed but will not rank well. Important pages must be linked to prominently in a crawler friendly way. 

Caveat:
There are instances when you may want to hide internal links or content from crawlers, but doing this too frequently can be harmful. 

Don’t expect search engines to have to render your JavaScript content. They handle HTML and CSS easily, but JavaScript is resource-intensive for them. It is best to render your content on your servers and deliver already rendered HTML to search engines.  

Tip 2: Manage your product inventory for SEO

Fashion inventory is varied and usually changes rapidly depending on trends and seasonality. Failing to effectively manage this swift inventory rotation is the equivalent of leaving your brick-and-mortar store to overflow with last season’s offering. It is, therefore, crucial to manage that product offering properly, by asking these questions:

Are products in the correct categories?

Navigate to any page on your site - look only at the product grid. Can you figure out what page you are on without looking at the copy or meta-data?

If not, you will likely struggle to effectively rank that page. 

Having a huge number of products makes them tough to handle. Unfortunately, if this is done wrongly you may struggle to rank for your target terms. For example, having a red dress on the blue dresses page is not going to make the purpose of that page clear to bots.  

Below is a screenshot of a brand’s ranking URL for “Red Shoes” (from page 6 in the Google SERP):

None of these examples are red, except the flip flops, which are not shoes. 

Are products named with SEO in mind?

There are two parts to SEO for ecommerce products, both involve naming the product correctly:

Firstly, consider user intent: Google the name of your product. Are the results you see the same type of product? A great example is Googling “workwear”; fashion brands often expect this to mean smart casual clothing for the office. Unfortunately, these get mixed in with combat trousers and rubber boots. 

Similarly, consider your customer’s geographical location. A user in Australia may refer to flip flops as jandals, or thongs. You will not rank effectively for terms that search engines don’t associate with your product offering. 

Secondly, if the product’s name doesn’t contain its colour, or style, or what it is, then there is not much information to help the algorithm figure out which term it should rank for. 

Products must be accompanied by their descriptions.

Even Nike with their famously non-descriptive products have given a thought to poor crawlers:

What happens to old product pages?

The worst-case scenario is that they are removed from internal linking but left live on the site. You may have years of products cluttering up the search engine index and competing with the site’s current offering.

You also do not want to be serving a 404/410 status code on old product pages. At least, not if they have backlinks. Ideally, an organisation should be checking their product backlinks regularly to see what should be salvaged, and what can be removed from the site. 

What happens when a Product Listing Page runs out of stock?

If the answer is that it returns a 404 status code, then an immediate change in processes will be necessary.

If a page returns a 404 status code for an extended period it tells both browsers and search engines that the page doesn’t exist. Google may revisit this page a few times, but within a few days, it will be removed from search engines eradicating all rankings associated with it.  

Stock comes and goes but the URLs the stock sits on should not. Otherwise, when stock returns to the page it will no longer be ranking. 

Tip 3: Avoid Index Bloat, or having too many pages 

Most, if not all, ecommerce websites have index bloat. Search engine resources have already been discussed in this article when referring to rendering. The same principle goes for visiting the pages on your site.  

Search engines have to monitor the entire internet, which takes time, server space, power, and all the costs associated with those.  

If your site contains thousands (or millions) of additional pages that you are not aware of you can be certain the site as a whole might suffer when search engines don’t find the time for your important pages. 

Smaller sites do not get a free pass though. Fashion ecommerce brands that have a higher proportion of optimised pages (optimisation score, in the graph below) perform much better than smaller sites that are less optimised.  

Ball size depicts website size:

The proportion of optimised pages compared to the market share of top fashion brands.

So, even if resources are spent on optimising products and categories at scale, there may be no performance increase if optimised pages are outnumbered by non-optimised pages. 

The explanation for this can be seen in an example from a major Fashion brand who, despite making efforts to remove pages from the index, found that these were still being heavily crawled by Googlebot. 

92% of crawled pages were set to Noindex 

The result is that Google may not visit your important pages for months: the example below illustrates a 3-month absence for Googlebot on this Fashion ecommerce site. 

Cache inspected on 29th June 2020

Here are some questions to ask to identify this kind of problem:

Does your ecommerce website have Orphan Pages?

Orphan pages are unlinked to the site. They exist, perhaps in Google’s index, on links on other websites, even in your XML Sitemap, but they can’t be navigated to by clicking around on the website itself... 

These pages are out of sight and out of mind, so newer versions may be created which will never rank because of the duplicate content issues present. Moreover, they stay in the search engines index and are continuously visited by search engines.

How are you limiting the number of facet or filter selections?

Facets, or Filters, are usually found on ecommerce websites and provide users with a way of filtering the product selection on category pages.   

When these are internally linked and accessible to crawlers they can be helpful for ecommerce SEO. Facets are useful for SEO because they categorise the different types of product on a page, illustrating to crawlers that the site has a varied collection of, say, Jeans.

Problems can arise when the options are not controlled properly with SEO in mind. For example:

3. Do facets also have a category elsewhere on the site?

Polo Shirts happens to have its own category on this site, with its own selection of colours resulting in pages for:

• Polo Shirts (facet) + colour (facet)
• Polo Shirts (category) + colour (facet)

The result is more Polo Shirt pages than products.. 

4. When selecting a filter, how is the URL built?

If you visit Facet 1, then add Facet 2 to reach a page with both Facet 1 & 2’s product option, for example, Classic + Polo Shirts, some sites may change the URL depending on the order you clicked the facets. 

The result may be two URLs with the same offering competing with one another:

website.com/Facet-1/Facet-2

website.com/Facet-2/Facet-1

Given all of this complexity, some ecommerce companies choose to not have these filters be accessible to search engines at all. Instead, they are solely for the user’s experience of the site. If they are included some should be Noindex, Nofollow, - which is entirely up to the site.

A standard set of rules for managing facet and filter indexing for SEO are:

• Noindex, Nofollow after three selections from any filter
• Noindex, Nofollow two filters from the same type e.g. two Style filters. 
• Noindex, Nofollow sizes
• Noindex, Nofollow other brands (situational). 

What happens when a page is deleted on the website’s content management system?

What happens to a page if it is deleted from the CMS? Find and visit old product URLs and PLPs, they may well still exist and be indexed.  

Often to find index bloat or a crawl trap you will need access to server logs. This is the raw data from your website’s servers listing every visit from users and bots alike. You can use them to find out exactly where search engines robots are spending their time, though you will need to know each crawler’s User-agent.

• Even if you have inventory managed you could end up with index bloat:
  • Some causes are new categories, multiple URLs.
  • Having too many filters discoverable.
  • Having the same filtering options in different categories.

Takeaways:

• Any optimisation, if not discovered, will have no impact. 
• More effort is needed for SEO for larger sites.
• If you have a small team, keep the site small and manageable.
• Only show Search Engines purposefully optimised pages. 

Tip 4: Realise SEO isn’t only on-site

If internal teams don’t play ball you won’t see improvements. Different teams are likely responsible for adding merchandise onto the site, categorising those products, buying, producing copy, and deciding on on-page features. Without cooperation across all of these, with SEO-friendly processes a consideration in those teams, you will likely see rankings slipping over time. 

Communicate internally, find the limitations across the teams, and adjust your strategy accordingly. 

Product lifecycles may hinder site SEO

If products regularly change, it may not be worth ranking them. SEO uplifts can take time. By the time Crawlers discover a page, then begin to rank effectively for the target term, the product may be at the end of its lifecycle. Instead, the focus should be on ranking category pages and ensuring these fast-moving product pages are cleaned up. 

Unfortunately, SEO can be a slow process, though once rankings are gained they are unlikely to fall off dramatically. 

Fluctuations in consumer interest in seasonal fashion can damage performance

Stocking products that users do not search for is a common reason for businesses losing faith in the value of SEO. Page 1 rankings are earned, but traffic and revenue do not reflect those gains. 

This is where keyword research is hugely valuable for a business. SEO can only catch the search volumes that are out there - unlike advertising, it cannot prompt users to search for a product. 

Technical limitations may stop your SEO strategy in its tracks

Your site may have technical limitations on redirects, canonicals, and many other things. A good example is if your site is built with multiple CMS (as many are) there may be limitations in how the two interact with one another. 

This usually comes in the shape of not being able to redirect from one to the other, leaving the site with a serious hindrance when attempting to retire old pages. 

Tip 5:Monitor your keyword split

What are branded and generic keywords?

Branded keywords include the site’s brand name or branded product e.g. Levi’s Jeans.

Generic keywords (non-branded) do not contain any brand names e.g. mom jeans

Most ecommerce sites struggle with generic keywords. Usually because of the existence of the issues mentioned above. Having too few signals of relevance and too many pages can leave search engine crawlers befuddled, as a result, most ecommerce site traffic leans heavily towards branded queries.

Businesses that can crack this difference generally perform far better. Their traffic is not reliant on brand recognition. 

Brands can easily rank for their brand name, for example, Nike shoes, or Levi’s Jeans.  It’s very rare to see another site take the top spot for these, in the ‘Levi’s’ example below they have the first 4 spots. There will not be a way to displace them for their brand terms.

Generics are the keywords that you will be trying to rank for. A user searching a brand keyword has all but converted, they have already decided what they want. Generics though, give access to a world of consumers still in their research phase, waiting to be convinced to buy your product. To rank for generics is difficult and so is converting these searches, but they make up the majority of searches within any industry.

Longer generic phrases may still reward you with a conversion, though they have a far lower volume.

These are just the starting points before you begin to get into the details of your go-to-market strategy.  

Many of the tips are transferable beyond fashion to most online businesses. There are some caveats included because, in reality, tips can’t always cover the variety of situations that might impact a site.

Successful ecommerce strategies include all of the following:

• Well-thought-out site architecture
• The correct approach to Keyword research 
• Consistently monitored backlink profile
• Audience-focused content marketing
• Simple yet effective Digital Analytics Tracking
• Proper UX and Conversion rate optimisations

We hope you’ve found these tips useful and can take these technical strategies away and interpret for your own ecommerce businesses. If you want further reading on this subject, check out our ecommerce guide for optimising through site taxonomy.

If you’d like a more in-depth personal analysis and a bespoke strategy, our SEO experts are on hand. Get in touch with the team here, we'd love to hear from you