Abstract

Startups are important to a nation’s economy because they are sources of disruption and innovation, which in turn accelerates economic growth. Using crowdsourced data from Crunchbase, we aim to determine the factors that contribute to the success of a startup company. We established that the best machine learning model is gradient boosting method and using feature engineering we further tuned our models.

Results show that acquired startups are generally in the biotech and advertising categories and originate from SF Bay Area and China, predicted with 85% accuracy. Startups that have successfully gone public similarly are in the biotech, pharmaceutical, healthcare, and advertising categories from China, predicted with 72% accuracy. These models can be similarly adapted to Philippine startup companies and may be used to guide them towards success.