Privacy-First Attraction Recommendation
The Challenge
Tourism and destination platforms need to personalise attraction listings but often cannot rely on persistent user profiles or behavioural tracking due to privacy constraints and the short, anonymous nature of most visits.
Our Approach
We design ranking systems that leverage structured attraction attributes, contextual session signals, and lightweight interactions, allowing relevant results without relying on personal user data.
Outcome
Personalisation that improves discovery without increasing regulatory risk or requiring intrusive tracking.
Cold-Start Event Discovery
The Challenge
Event discovery platforms often have low repeat behaviour and little historical data, making traditional recommendation methods unreliable.
Our Approach
We combine interactive UX signals, embedding-based similarity models, and adaptive machine learning that adjusts to available feedback. High-signal interactions such as swipe-based preference capture enable meaningful personalisation within minutes.
Outcome
A guided discovery experience that helps users identify compelling events in 1–5 minutes, accelerating booking decisions without requiring extensive user history.
Explore live prototype →Built on aggregated Fáilte Ireland event data
Intelligent Catalogue Discovery
The Challenge
Popularity-based ranking creates a feedback loop where the most visible content gets more interaction while new or niche items remain unseen. Over time, this reduces diversity and makes discovery less effective.
Our Approach
We implement adaptive ranking logic that integrates business rules, contextual signals, and structured content intelligence to distribute exposure more intelligently, without compromising relevance.
Outcome
Balanced exposure across the catalogue, better discovery, and stronger visibility for new or niche content.
See your use case here?
Start with a structured engagement, either a 30-day pilot or a signal assessment.