Website Categorization API

Our Website Categorization API scans the text and metadata of URLs and then applies machine learning (ML) to classify them.

Our solution supports both the Internet Advertising Bureau (IAB) categories as well as Google Product Taxonomy categories.

Real-time classifications with API

Our API allows real-time classifications of both homepage domains or full-path URLs.

Detailed taxonomies

We support website classification into 1000+ categories for ecommerce websites and into 400+ categories for general websites.

Custom classifications

In addition to standard taxonomies, we can train and set up highly accurate ML models for custom taxonomies, specific to your case.



Use Cases

Use our categorization API for marketing leads, online filtering, brand safety, segmentation of markets or use it in your own services/apps. For latter, we also provide you with offline categorization databases.

Ad Exchanges, Demand Side Platforms (DSPs), Supply Side Platforms (SSPs), Ad Networks

Our API allows highly accurate and detailed categorization of websites, which can be used by many participants in the AdTech ecosystem. SSP (Supply Side Platform) companies can e.g. immediately identify the advertiser’s category to checks its eligibility for real-time bidding. At the same time, categorization allows exclusion of advertisers that may be deemed a threat (due to malware, phishing, counterfeiting, etc.).

Try it out

Ecommerce

Proper categorization can help Ecommerce companies, such as online stores, provide better user experience to their customers, which will stay on their website longer and will more likely find desired product and purchase it. Easier search, better filtering, higher conversion rates all lead from implementation of efficient categorization of products on online stores. We provide you with one of the most accurate Ecommerce classifications, with over 1100+ different categories.

Find Out More

Web Content Filtering / Cybersecurity

Web content filters are used for wide variety of purposes. They can be used by companies who want to implement internal policies on which content is acceptable to be viewed in workplace and thus want to filter out non-work related websites like social media platforms or sports websites. Another application is as part of cybersecurity framework, where we want to filter out spam, phishing and other undesired websites from being accessed by users. Our categorization services can help you in your web content filtering and cybersecurity initiatives.

Find Out More



Offline database for marketing leads

Our offline database provides you with categorization as well as the following additional information (for 1+ million online stores):

  • Number of products sold
  • Traffic ranks and popularity
  • Social media accounts and emails
  • Technologies used in online store (tech stack)
  • Similar stores for each given store in database (based on NLP analysis of products sold)
Try out free tool for domain categorization

1+ million URLs categorized each day

We categorize over 1 million URLs using our machine learning models daily.

Find Out More

Introduction to Website and Domain Categorization - application to Online Stores

When we go shopping to a local store to purchase grocery or other things, almost all stores have various parts labelled with categories, e.g. Fruits, Vegetables, Dresses, Jewelry, etc. This categorization system allows us to quickly find what we are looking for.

It is then not surprising that we expect the same kind of help from online stores as well. In a way, the categorization for online stores is even more important, because whereas in the physical store you could eventually visit all parts of the store by foot, the equivalent in online setting is much harder. Some online stores literally offer millions of different items to purchase.

Categorization improves the search, discoverability and filtering on the online shop websites and ultimately leads to better conversion.

Additional benefits of Website Categorization for Online Stores
There is also additional reason why online stores can benefit from web site categorization. A lot of them obtain their visitors via the search engines. For that to occurr, the online store benefits from having many webpages indexed in search engines, each one presents an additional opportunity to be found by a visitor searching using a certain keyword.

If an online store has many products then their categorization and addition of category pages (which show products in the same category) is an easy way for online store to increase their presence on rankings of search engines. And thereby potentially gain more visitors.

Product Taxonomies
When it comes to product categorization there are many possible options on how to categorize products into categories. In fact, many of the online stores have their own set of definitions, which are also called product categorization taxonomies. If one does not want to invest time in inventing own taxonomies, there are several taxonomies available to choose from. The most well known ones are those from Google and Facebook.

Also often used is the one from organization Interactive Advertising Bureau or IAB for short. The latter has several different taxonomies available, each suite for particular purpose. Though if one is interested in ecommerce website categorization then the taxonomies of Google and Facebook are the most detailed ones.

When categorizing products there are two main ways of doing it. One can assign a single category on certain Tier level. E.g. sunglasses can be assigned category "Apparel & Accessories" for so-called Tier 1 category, "Clothing Accessories" for so-called Tier 2 category and "Sunglasses" for Tier 3 category, within Google Product Taxonomy. Or one can predict the full path:
Apparel & Accessories > Clothing Accessories > Sunglasses
This is also known as the taxonomy path of the product.

The higher number of levels you use, the more easily can your customers find your products in your online shop.



Caption: Website categorization taxonomies, use cases and benefits



Automated website categorization with machine learning models
In practice, website categorization is done in an automated way, using machine learning model developed and trained specifically for this purpose. Although there are many possible machine learning models that can be used, from Support Vector Machines to Recurrent Neural Nets to name just a few, the key feature of any machine learning model is the accuracy it is able to achieve. An important role in regard to that is the size and quality of the training data set on which machine learning models are trained. As with all machine learning models, the more data we have available, the better is the accuracy.

We took great care in preparing large and high quality training data sets for our website categorization models resulting in very high accuracy scores.

URL classification for non-english languages
Our website categorization solution also works with texts that you want to classify but that are in non-english languages. We support website categorization of texts written in german, french, italian, spanish, portuguese and many other languages. Contact us at [email protected] for more information.