Automated content categorisation and tagging
Language models tag your content against a fixed taxonomy, so search, filters and recommendations return the right results.
Background
A large library of articles, products or documents is only as useful as its metadata. When categories and tags are missing or inconsistent, search, filters and recommendations suffer. There is usually too much content to fix by hand.
How it works
Language models read each text and assign categories and tags from your taxonomy. The fixed vocabulary stops them inventing tags. They also detect audience, content type, related products and language. Embeddings surface similar content for cross-links.
Use cases
Tagging existing archives retrospectively
Classifying new posts when they are saved
Mapping products from supplier data to the right categories
Sorting intranet or portal documents by topic and responsibility
Quality control
Every tag carries a confidence score. Confident suggestions go straight in, uncertain ones go to your editors, whose corrections refine the rules. We build the taxonomy with you, and it is often the more important part of the work.
Integration
In Statamic, results land directly in taxonomies and fields. Other systems connect through APIs. Processing runs with EU-based providers or on your own servers.
Large libraries are processed once in batches, then only new and changed content. That keeps costs low.