How KHL optimised its content archive with SUBJCT

Content tagging at scale & streamlining editorial workflows

Like many B2B publishers, KHL faced significant challenges:

1. Accurate tagging of content — required for content categorisation and used for contextual targeting in marketing campaigns.
2. Manual internal linking — a time-intensive process impacting editorial efficiency, user experience and brand visibility in search.
3. Content engagement — increasing user session duration and optimising content recommendations, also a manual process.
4. SEO complexity — improving topical authority in the AI search era.
5. Scalability — a need for automation to handle high-volume content production across multiple platforms.

SUBJCT’s AI-powered content optimisation

KHL implemented SUBJCT’s AI-powered platform to streamline tagging, internal linking, and content recommendations. The goal was to:

• Accurately tag the KHL content archive — using entities (topics).
• Reduce manual SEO tasks for the editorial team — internal linking.
• Improve content discoverability and engagement — automating content recommendation.

Workstream — Automated tagging

The SUBJCT platform powers entity analysis at scale, which was used to analyse and tag articles from the KHL website archives. The solution was developed to understand the entities (topics) and categories within the archives and automate the tagging of articles moving forward. It was delivered via API to the KHL team, and can be used for additional use cases such as contextual targeting for marketing campaigns.

In the Semantic Web, an entity is the “thing” described in a document. An entity helps computers understand everything you know about a person, event, an organisation, or a location mentioned in a document or article. Categorising and optimising content around entities is crucial in the era of AI search.

The SUBJCT entity analysis also supports data-led content strategies, helping KHL to identify new topic opportunities to drive topical authority for their brands.

Key deliverables:

• Entity analysis to understand the entities (topics) and categories within KHL’s 30,000 archive articles.
• Entities broken down into organisations, people, locations, events, products and ‘other’.
• Categories defined using IAB 3.1 standards.
• 8 domains analysed within the KHL organisation.
• Delivery of the data via API to KHL, and via the SUBJCT Web Application.
• Delivery of the article tagging solution to KHL via API.
• Tags prioritised to support the user experience when presented to users.

Workstream — Automated internal linking

The SUBJCT platform also powers automated internal linking to reduce manual content optimisation tasks for editorial and SEO teams. Internal linking is a critical solution to deliver topical authority for publishers in both ‘traditional’ and AI search.

SUBJCT automates linking in two ways:

• Topic linking automates the process of linking an entity within an article to the relevant topic page when it first appears in the article.
• Article linking automates the process of semantically linking relevant articles to create topic clusters. SUBJCT chunks articles into passages and uses vector embeddings — a numerical representation of text that machines can understand — to find the most relevant article within the archive for that passage, then automatically places a relevant anchor text into the passage and links it.

Key deliverables:

• Automated internal linking solution for KHL — topic hub linking and article linking.
• Link prioritisation features.
• Delivery of the linking data via API to KHL and via the SUBJCT Web App.
• Removal of a manual process within the editorial workflow.

Workstream — Automated content recommendation

Adding related content to article pages was a manual, time-consuming process for the KHL team, while maximising content engagement and user experience is key for KHL.

The SUBJCT platform uses the data from its initial content analysis to power an automated content recommendations solution, delivered to KHL via API. SUBJCT used vector embedding technology to match related articles to the article being viewed by a user, automating the process and delivering relevant content recommendations.

Key deliverables:

• Article analysis on 30,000 articles, creating vector embeddings for each.
• Delivery of the content recommendation data via API to KHL.
• Removal of a manual process within the editorial workflow for KHL.

Key staff cohorts involved in testing & feedback

The ongoing work between SUBJCT and the KHL team is truly collaborative, including team members from operations, editorial, SEO and engineering.

Article tagging and internal linking have been tested to ensure that accuracy thresholds have been met, and any refinements required to match editorial standards will continue to be made. Regular team feedback sessions to support implementation and alignment across all teams were introduced.