Personalization & Recommendations
Smart. Adaptable. Engaging.
Recommendations built on what your content is about, not only on what other people clicked.
Metadata makes the first recommendation as good as the hundredth.
4
Signals per article: cultural, geographic and temporal closeness plus editorial news value.
0
Cold-start articles. The knowledge graph places new material immediately.
Editors keep control: pin, exclude, boost a desk, or hold a slot for the story that matters today.
Interest profiles
Ask once, expand automatically, and let people correct it
Behaviour alone takes weeks to describe someone, and it never explains itself. A stated profile is instant and legible: three answers, expanded through the knowledge graph, editable by the person it belongs to.
01
Choose your subjects
Energy policy
Procurement
Infrastructure
Regulation
Labour market
Sustainability
Research funding
Public health
Offered from your own vocabulary, ranked by what the archive actually holds.
3 chosen · minimum 3
02
The graph adds the rest
Grid tariffsvia Energy policy
Wind powervia Energy policy
Tender rulesvia Procurement
Rail investmentvia Infrastructure
Three answers resolve to 27 connected subjects, so the profile is dense from the first minute.
27 subjects in profile
03
Edit it, any time
Energy policyChosen by youFollowing
Rail investmentAdded from your activityFollowing
Tender rulesToo much of thisMuted
Public healthSuggested for youFollow
Every subject in the profile is visible, labelled with where it came from, and reversible.
Profile stored per person
The same profile, whatever the archive holds
Subjects come from your vocabulary and the audience is whoever needs the content. Only the words change.
Publishing & broadcast
Readers and subscribers
A personal feed and newsletter from day one, with the archive treated as current material rather than backfill.
Enterprise & internal
Employees and support teams
Follow a supplier, a regulation or a product line and see every document that touches it as it lands.
Research & public records
Members and analysts
Standing interests across collections, so new records surface against a question someone asked months ago.
Subjects map to your own taxonomy, not a vendor’s
Stored per person, no cross-site tracking
Stated and learned interests are labelled separately
Where recommendations appear
One engine, orchestrated across the channels you already publish to.
Homepage
Modular slots that mix editorial curation with per-reader ranking.
Newsletters
A selection per subscriber, rendered at send time.
In-article
Related reading drawn from shared concepts, including archive material.
Personal feeds
A standing feed per reader, built from their own interest profile.
NearYou
Local journalism, local geometry
Every article gets coordinates from the places it mentions. Readers get what happened near them, ranked by distance as well as relevance.
Kalmar city centre2.4 km
Öland bridge9.1 km
Nybro31 km
No Cold Start
The graph fills the gap
A new article has no click history, but it does have concepts. The knowledge graph connects it to material readers already engage with, so it can be recommended from minute one.
Wind power→Energy policy→Grid tariffs
Curious what your archive is worth?
Bring a sample of your content and we will tag it live.