The factors contributing to the survival of startups are an emerging area of research, requiring a deeper understanding of the variables that influence their success. This study employs data mining techniques to analyze the relationship between business segments, target audiences, income models, business stage, and the survival rate of startups in Brazil. Starting from a public, unique Brazilian startups research data- base with 12,207 listed startups, we create a dataset of 2249 technological-based startups in the most representative business segments, such as education (Edutechs), finance (Fintechs), biology (Biotechs), and others. We call the business segment the “X” variable, the term’s origin in X-Tech. Our initial hypothesis was that the busi- ness segment (“X” variable) was a determinant of business survival. This dataset was used to construct a random forest model using Rapidminer software to predict which independent variables are more relevant to the survival of startups. The find- ings reveal, with an accuracy of 70% and κ = 0.72, that the choice of target audience (primarily B2C and B2B) and income model (particularly the marketplace model) are more influential in determining the survival of Brazilian technological-based startups. The marketplace model, offering visibility, cost-effectiveness, and conveni- ence, emerges as a crucial factor, especially with a B2C or B2B target audience. Both primary and secondary variables suggest that positioning a startup on a mar- ketplace platform targeting a B2B or B2C audience is more likely to enhance its chances of survival in Brazil. The study also shows that the business segment, the “X” of the X-techs, was not relevant to the survival rate.