Seems like you're just lacking robust statistical analysis. Survey 10,000 digital nomads on personality traits, where they have/haven't enjoyed, and then do fac

Seems like you're just lacking robust statistical analysis. Survey 10,000 digital nomads on personality traits, where they have/haven't enjoyed, and then do factor analysis to identify buckets of digital nomad "types". It's extraordinarily unlikely that "every person needs to find out for themselves by trying a bunch of places." Use similar questionnnaire to identify which nomad "type" the user is. Recommend them cities using a bayesian adjusting system. Figuring out how many digital nomads "types" there are will take a while and be tricky. But I don't think this is an intractable problem.