Technisches SEO für Onlineshops means structuring categories, filters and product pages so that search engines can crawl, understand and index them sensibly. What matters are clear signals for status codes, canonicals, internal linking, titles, meta data and loading behaviour, so that technical findings can be fixed in order of priority.
In online shops, problems often do not arise from individual pages, but from variants, filter combinations and similar content. That is exactly where a clear technical structure helps.
- Status codes show whether a URL is reachable or redirects.
- Canonical tags define which version should be preferred.
- Internal links steer which pages appear more important.
- Titles and meta descriptions should be unique.
- Loading behaviour belongs in the check when pages are to remain technically sound and usable.
Faceted navigation
Faceted navigation means filters and facets that allow users to narrow products down by properties such as size, colour or brand. Technically, this often creates many URL variants.
For SEO practice, the key question is not whether filters exist, but which filter pages have their own search intent and which merely support navigation. A technical term for this is the canonical tag: it is a hint about which URL should be treated as the preferred version.
If many combinations generate their own URLs, you should define the rules clearly for each URL type:
- indexable
- not indexable
- referring to a canonical target page
It is also important to check the signals in the system. A site audit can make status codes, canonicals, robots directives, internal links and loading behaviour visible; on the audit view, such findings can be brought together.
If you want to see which filter or category URLs actually have search performance, a look at rank tracking will help. For classifying the findings, the dashboard view can bring the existing areas together in one place.
Whether a filter page should be indexable depends on whether it has its own, clearly described content. Pure sorting or combination pages are usually technical variants and not good landing pages.
Duplicate content
Very similar or identical content across multiple URLs in an online shop is often not created deliberately, but through categories, filters, sorting or product variants. For search engines, what matters then is which URL should be treated as the authoritative one and which is merely a technical variant.
This requires clear signals in the system. Typical examples are:
- a suitable canonical tag
- consistent internal links
- redirects when a URL is no longer to be used
A canonical tag is an HTML hint about which URL should be treated as the preferred version. It does not replace a clean structure, but it helps assign similar pages correctly.
Also check whether duplicate titles and meta descriptions are being created. Such patterns often show that templates or data sources do not distinguish enough. For assessing your own pages, Search Console data can be helpful, because it shows clicks, impressions, CTR and average position for the verified property. If you are planning new landing pages, keyword research can help separate similar search intents at an early stage.
For classifying the competitive situation, competitor analysis can also be added. This is about looking at the search landscape, not about technically correcting your own URLs.
The important point is this: the cause usually lies in the URL logic, in templates or in how content is distributed. That is where the clean-up should begin, not on the individual page itself.
Product page basics
Product pages should have a clear, unique URL. It is important that they are not displaced by several almost identical variants just because individual features differ.
The page title should describe the product precisely. The meta description should summarise the page content in a meaningful way. Neither should be copied blindly across similar products, otherwise important differences are lost.
Headings help structure the page content. A main heading should clearly name the product or product family. Subheadings can, for example, organise these areas:
- features
- contents of delivery
- care instructions
Product information must be embedded in such a way that search engines can understand the core content without relying solely on images or scripts. This is especially important when key details are otherwise only available visually or are loaded later.
Internal links from categories to product pages and back help search engines discover and classify important pages. Consistency matters more than having as many links as possible.
If product pages are empty, thin or highly redundant, the causes should first be fixed in the system:
- templates
- data feed
- variant logic
- rendering
A technical audit view is suitable for a systematic check. There, titles, headings, canonicals and status codes can be checked together.
Crawl budget in simple terms
Crawl budget, put simply, is the amount of attention search engines give a website for crawling. It is not a fixed guaranteed value, but results from size, freshness, internal linking, server responses and URL structure.
In online shops, crawl efficiency becomes especially relevant when there are very many URLs, including filter combinations, archived products or parameter variants. In that case, it can help to reduce unnecessary paths and make important pages easier to reach.
The aim is not to hide as much as possible. A sensible approach is rather:
- clean redirects
- as few faulty URLs as possible
- meaningful internal links
- a clear separation between indexable and non-indexable pages
Status codes are important here. 404 means that a page was not found; redirects send users and crawlers to another URL. Both should be assessed deliberately in the technical review.
A technical note: in its public documentation on crawling, indexing and ranking, Google Search Central describes its own guidance on these topics; the respective page may change and should therefore always be checked for the latest version.
Frequently asked questions about technical SEO in online shops
Not every filter page should be in the index. The decisive factor is whether the page has its own search intent and shows content that is genuinely relevant to that search. A pure combination or sorting page often does not fulfil this.
So, for each URL type, check what role the page has in the shop:
- its own landing page with clear content
- technical variant for navigation
- canonical version of a similar page
What matters is that the rules in the system are consistent. If filter pages are indexable at times and not at others, contradictory signals appear quickly. This often affects internal linking, canonicals and control via robots directives.
A technical term for this is the index status: it describes whether a URL can be included in the search index or not. The decision is not based on the mere technical existence of a page, but on its purpose.
Is a canonical tag enough to deal with duplicate content?
No. A canonical tag is a hint about the preferred URL, but not a substitute for clean structures. It helps search engines assign similar pages, but it does not solve every cause of duplicate content.
If similar pages arise from categories, filters or variants, several signals should align:
- canonical tag
- internal links
- redirects
- templates and data logic
Only when these levels are coordinated does the technical picture become clearer. Otherwise, a page may have a canonical, but still appear contradictory through other signals.
The real cause is often not the canonical itself, but the structure that creates too many similar URLs. That is where it is worth making the correction first.
Why does Google Search Console show an average position rather than just a fixed rank?
Because a search position is not stable. It can vary depending on location, device, language and personalisation, and can change over the course of the day. That is why a single fixed rank is not a reliable value in everyday use.
For the verified own property, Google Search Console shows, among other things, clicks, impressions, CTR and average position. This average position is a mean across impressions, so it is not a single fixed placement.
This is useful for technical work because you can see trends and patterns. For point-in-time observation, however, separate rank tracking is needed, which measures a query for a chosen location and device at a specific point in time.
Which data is most important for the first technical check?
At the beginning, the findings that directly affect access to and classification of pages matter most. These include above all:
- status codes such as 404 and redirects
- missing, duplicate or overly long titles and meta descriptions
- canonical tags
- robots directives
- internal links
- loading behaviour
Work by priority here. Errors that block pages or send them to the wrong destination should usually be checked first. After that come recurring patterns that affect many URLs.
That turns a long list into a technical order for implementation.
How do I deal with a large number of findings?
Do not work page by page at random. It is better to group patterns and look for the cause. If a problem affects many URLs, it is often rooted in templates, data sources or rules.
A good process is:
- Check errors with a direct impact on accessibility
- Organise duplicate or missing page signals
- Review internal linking and variant logic
- Look at loading behaviour and rendering
This makes it easier to prioritise technical findings properly. Individual pages can still matter, but the more effective lever is usually the shared cause.
Can I use Search Console data to check what search engines see?
Only partly. The data shows what was measured for the verified own property, not the whole internal assessment of a search engine. It does, however, help to classify visibility and search queries.
For technical questions, you also need an audit view of status codes, canonicals, robots directives, internal links and loading behaviour. Only together do they produce a usable picture.
As an observation, if individual pages have little visibility in Search Console, this can point to technical or structural causes. But that is a working hypothesis, not a fixed rule.
If you want to bring such findings together in one place, SerpDash can be used; it is being developed by ballmüller solutions in Berlin and is first being used for the publisher's own websites, with some functions still under development.
Next step
If you sort technical findings for categories, filters and product pages, a shared view of the areas relevant to this article helps. That allows you to combine hints from different sources and organise them for implementation.
The following are particularly useful in practice:
- rank tracking
- Search Console data
- site audit
Rank tracking measures the position of a query for a chosen location and device at a specific point in time. Google Search Console shows, for the verified own property, among other things clicks, impressions, CTR and average position.
SerpDash by ballmüller solutions in Berlin is first being used for the publisher's own websites; some functions are still under development.