This is true that most projects which later scammed once had good whitepaper, potential use case etc which make me most of the times fancy IEOs on top exchanges as the rate of exit scam is little to none. Nevertheless, what can we do than to increase our research skills, learning from how others were able to scam even when seen as good projects and using it as a basis to learn more about others. Also, another way most people tend to follow is waiting till the project is listed on good exchanges. However, in all sincerity, it is hard knowing how to spot scam projects because humans can't be predicted so at all times be watchful.