1) Build a network of experts you can rely on for evaluation 2) A lot of what makes a startup good or bad translates across different domains, so even if you c

1) Build a network of experts you can rely on for evaluation 2) A lot of what makes a startup good or bad translates across different domains, so even if you can't evaluate the technology yourself, you can still evaluate other things (e.g. how well the founders communicate, how quickly they get things done, etc.)