Frustrated that your free time feels the same, even though you keep trying new things?

Who built this

Compatibility over popularity.

XPMatch grew out of MileMatch, a travel review and comparison concept built on one observation: two honest reviews can be completely different because the reviewers have different reference points. Someone who normally books economy and someone who normally books business class do not mean the same thing by “worth it.”

So instead of ranking you, XPMatch starts with a private baseline, prioritizes reviews from comparable people, and explains where an option fits, where it may fall short and whether people like you felt it was worth the cost.

The research behind the questions

Taste, Discovery and Follow-through are XPMatch's own framework, not a validated psychological instrument. Each area rests on well-replicated findings, and we will refine the questions with beta data.

• Follow-through. People who decide in advance when and where they will act are far more likely to follow through. A meta-analysis of 94 studies on these "implementation intentions" found a medium-to-large effect. That is why the quiz asks whether your weekend is planned before Friday. Gollwitzer & Sheeran, 2006, Advances in Experimental Social Psychology
• Discovery. More options do not help. In a classic field study, shoppers offered 24 jams were far less likely to buy than shoppers offered six. A short list you trust beats an endless feed, so the quiz asks whether you keep one. Iyengar & Lepper, 2000, Journal of Personality and Social Psychology
• Taste. People are surprisingly poor at predicting what they will enjoy and better at remembering what they actually did. Noting what you loved and what you would skip is the cheapest way to learn your own taste, so the quiz asks whether you do. Wilson & Gilbert, 2005, Current Directions in Psychological Science
• Asking directly. A study of 1,035 travelers found personality relates to travel preferences, but its authors say recommendation models built on personality still need testing. Asking people directly what they like remains the most reliable signal. Alves et al., 2023, User Modeling and User-Adapted Interaction
• Your exact spending is never shown to anyone. The profile uses broad ranges and expectations, not receipts.