.svg){kind=icon}
About the Client
We worked with a specialist travel company selling villas and holidays across several countries and regions, with a small trading team setting prices.
The Challenge
Demand moves constantly by region and season. Some properties sold out fast while others sat unsold for weeks, and the team could not always tell in time which ones needed attention. Pricing reviews meant scrolling through spreadsheets, comparing numbers by eye and relying on memory of last season. Issues were often spotted after the best time to act had passed. The large revenue management systems used by airlines and big tour operators need far more data than a company of this size has.
How We Helped
We built a revenue management tool designed for smaller data sets. It groups similar inventory into clusters, for example by region and date band and custom tags, so the data is reliable enough to price from. Each cluster is measured against a pacing curve built from several years of booking history and price sensitivity analysis.
The main screen is a daily exception report by country and region. It shows which clusters are selling faster or slower than expected, and by how much. Where the right price move is unclear, AI suggests one and can save it as a rule for next time. A what-if view compares like-for-like inventory, for example three-bedroom villas with pools across regions. Once the analyst decides on a change, the tool logs it and pushes it to the reservation system.
What We Achieved Together
The trading team now sees every property that is off pace each day, instead of finding problems by chance. Trading meetings start from a short list of clusters that need a decision. Analysts spend their time on price moves rather than data preparation, and every change is recorded with its history.
What This Means Going Forward
Smaller travel companies can price with the discipline of large operators without needing their data volumes. Clustering inventory and tracking it against a pacing curve shows where to act early, and AI helps decide the move. The same approach works for 100 properties or 50,000.
