We’ve worked with many very large e-commerce and marketplace websites that have developed specific taxonomy structures specific to both their industry and geographic markets. What customers tell us is that it is often complicated to analyze site data by taxonomy categories (or filters) quickly and easily. Additionally, making meaningful changes to the taxonomy and using data insights to improve SEO can take a long time.

To help product-led SEO teams analyze site data using taxonomy categories we’ve added functionality that allows customers to upload a taxonomy specific to their business. Then you can understand demand and other SEO data points using the taxonomy as filters. For instance, you might want to understanding the most important facets in a group by demand, impressions, traffic or even domain diversity. This allows a multitude of query combinations, for example:

All of this can be done using the Similar.ai no-code user interface.

It allows the SEO and growth team to ask questions based on the - their site specific taxonomy.

One example question is: Which facets have the most demand? These data insights can be used to look at ways to clean up and change the taxonomy of your site to better suit demand and what users are actually searching for. With the detailed granular GSC data we get for our customers, you can understand seasonality too. These are just a couple of basic examples.

Apart from just insights, you can also use this data to update internal linking. For instance, you can grow internal linking to the pages with facets that are about to explode in seasonal demand. And, because the Similar.ai platform is always on, these links will stay up-to-date throughout the year, as seasonality changes.

To find out more about using site taxonomy for data insights or to keep your site up-to-date with demand, you can request a demo of the Similar.ai no-code SEO platform.

The Similar.ai platform enables new ways to interrogate your site's data using distributions. The data is displayed as a histogram, which you can then drill down into further and choose which combination of metrics to review.

This allows you to quickly and easily learn about your site's structure, giving the ability to see the distribution of different metrics across your site.

All of this can be accessed in a with no need to write SQL queries.

Below is example screen shots of the summary table and a histogram looking at page demand ranked by number of listings per page:

Using Similar.ai you can build up a library of recipes that can be easily reused for analysis and automating site update. Read more about 'what are' and how you can benefit from recipes HERE.

We've talked to tens of different product-led SEO teams about their struggles with internal linking. We heard that

If these are problems with which you struggle, our data driven internal linking with functionality could be a game changer for your SEO team.

Recipes are simple no-code logic to filter your pages. You can choose which pages you'd like to link from or to and which pages you'd like boost inlinks for.

Linking ideas / strategy can include from, to and boost. For example:

The Similar.ai platform also measures the impact any linking changes you implement. Allowing a test and learn approach

Through our simple server-side API integration your page can query our API and get a list of anchor texts and destination URLs, optionally grouped into different blocks or modules and then render them. You might choose to render them:

For inspiration, here are some example recipes that our customers have used:

You can read more about what makes recipes powerful HERE. And if you want to learn more, please request a DEMO.

We recently launched a feature that maps your site taxonomy & faceted navigation to demand.

It's simple to use. We pull in the keywords for which organic competitors rank and group those into topics, using the knowledge graph we build for your site. Similar.ai identifies the topics which

and then creates the correct URL for the new page using the logic of the site works today. Through our one-off simple API integration, you can choose the action 'Create', click 'Publish' and the pages will go live the same day. Similar.ai dashboard will immediately start tracking how quickly Google crawls, ranks and delivers traffic.

Not going fast enough for you? You can also boost inlinks to these new pages to give them a helping start with an internal linking recipe.

This means you can launch new pages within your taxonomy and faceted navigation that have lots of relevant products and relevant keywords for which they could rank & for which your competitors get traffic. These new pages naturally get a lot of organic internal links from the site, because they are part of the site taxonomy itself.

Optionally, you can tweak the recipe above. (recipes are no-code SEO programs: here are some examples). For instance, you could decide to:

Read more on how to do faceted navigation for SEO.

What is ? Domain diversity looks at how many different domains are available on a particular Search Engine Results Page (SERP) and provides a relative score of domain density as a ratio between 0 to 1.

A SERP with high domain diversity (highest =1), includes results from lots of different domains. For a transactional query, those would mostly be multi-brand fashion retailers. A low domain diversity SERP would normally reflect a navigational query. In between can also be interesting.

Domain diversity is a useful measure because we show it a . So, although it reflects how attractive the topic is for a multi-brand retailer to rank, you can use it to make decisions about your pages, such as whether you should clean up a page that's not ranking or getting enough traffic, or whether you should boosting inlinks to that page and enriching with additional content.

Below is an example screenshot of a high domain diversity:

Here's an example screenshot of low domain diversity:

Domain diversity is a data-driven way to automate identifying

A common issue for programmatic SEO is that it's easy to generate text which is grammatically incorrect. There are plenty of demanding languages with cases and genders, such as German, French, Dutch or Russian. Even in a language without gender, like English, it's easy enough to find places where some pages talk about something in the singular (a dress for example) and other pages talk about things in plural (shoes for example). In these cases, it's really useful for the surrounding sentence to be able to change dynamically to account for it.

You'd just like to wave a magic wand and say "Correct my grammar!". This is precisely what our new grammatical error correction feature is built for. It's like Grammarly but it will happily scale to the 100,000s of pieces of content you generate every month.

When might you use a check for correct grammar? For each place in a template where the text might need change to account for grammar, you list out the possible choices. Common places are:

You can even let the platform choose between showing an article or showing nothing, such as when you're choosing between an article for a singular or plural noun, like "Are you looking to buy a dress?" vs "Are you looking to buy shoes?"

How do you use our grammar correction in practice? Simply give the possible alternatives in curly brackets, separated by a semicolon. For instance, you could express the above sentence with "Are you looking to buy {a; } {filter_attrs}?" where {filter_attrs} is our special entity to express the intent of a page. The platform will take care of the rest.

Grammatical error correction is currently available in German & Dutch, but let us know if you'd like to use on content in other languages, or if there are use cases on which you'd like us to focus.

Similar.ai demand-based content helps you reduce duplicate content by increasing the uniqueness, freshness and depth of your category pages:

For instance, the content entity finds the three colours with the greatest total search volume from amongst all the keywords with the page needs.

These types of entities form the heart of our demand-based content. However, until recently, it was sometimes hard to know why the platform chose certain content. Not any more. Now you can quickly check which keywords and volume were used for each entity on any page. Just click on the inspect icon from the main content page to see zoom in to each entity.

For instance, how would the platform reach a decision to create the answer "The most popular colours for an Abarth are grey, yellow and black" for the page?

Just click on the "inspect" icon

Here we can see that the platform found a total demand of

It used this data to create the answer "The most popular colours for an Abarth are grey, yellow and black".

This new inspection functionality lets you into how the content is created for any of your entities on any of your category pages.

Our content entities let you power your very own content optimization system to target the untapped demand for your categories. But after writing a template with entities, it sometimes took 30-60 minutes to generate them for just a few thousand pages. That meant it sometimes took a few hours to edit and tweak templates. Not any more.

Now, with what-you-see-is-what-you-get content editing (aka WYSIWIG to its friends), you can edit a content template and test it out directly on a few pages from your current segment.

This makes iterating on your template 100x faster and, more importantly, it makes finding and correcting small errors in the language simpler and quicker, so that all of your automatically generated content is accurate, grammatically correct and engaging for the user. And of course, because it's dynamic it will stay up-to-date with changes in your demand and inventory over time.

Related Searches is a Similar.ai product feature which helps improve internal linking for enterprise SEO.

Links, whether they be internal or external, are one of the most important signals to Google and other search engines. Links indicate which page is the best answer for a search intent.

Many different keywords can express the same search intent, so we cluster keywords accordingly. Total demand is the search volume of an intent.

We cluster keywords into sub-intents and show the total demand for each. One way to think about search intent is all the keywords for which a page might rank.

We check for which sub-intents a page does not get much traffic and where the page does not target that sub-intent. These can add incremental unbranded organic traffic.

When we link through a page, we generally use the main intent as anchor text to link. Pages with more demand get more internal links. For a proportion of these, we also use opportunity intents.

We should take the intent of the main keyword of a page as the intent of the page. When we link, we link with anchor texts which are distributed across the main keyword and opportunity keywords.

The main intent is still the dominant way users think of a page, but this extra anchor text diversity increases the number of pages which rank for lots of higher demand intents.

Category pages for large sites are often split into browse and search pages (understand the difference between browse and search pages).

An intent makes explicit all the implicit assumptions a user typically has about a set of queries and all the implicit expectations about the results. We express an intent as entities in our knowledge graph. For any search, browse or other category page, the Similar.ai platform understands the full intent.

We've released new functionality to translate the intent into the site structure, attributes and use a free text keyword for the remainder. This expresses an intent as a page on a site in the most idiomatic way possible.

Expressing an intent as a page is powerful for three use cases:

This means a site can have fewer category pages which cover more of its relevant search intents.

For instance,