Automation for efficiency: Data-driven solutions for Costanoa Ventures
Prof. Markus Pelger
Costanoa Ventures is a Silicon Valley-based venture capital firm that specializes in artificial intelligence and machine learning, data, fintech, and security. Costanoa primarily invests in early-stage business-to-business software-as-a-service startups at the pre-Series A level.
When betting on horse races, some bettors choose the most promising horse, while others place more emphasis on the jockey. Venture capitalists face a similar dilemma when deciding which startups to fund—is it more advantageous to bet on the product or service being offered (the horse), or the track record and qualities of the founders (the jockeys)?
Currently, Costanoa employs a manual system of combing through the internet in search of potential investments and builds spreadsheets to outline the associated estimated returns. Costanoa is interested in introducing automation into their system to improve its efficiency. Through our project, we produced a research report that provides insight into the most valuable sources and signals for identifying investment leads. We also generated the supporting scripts and data.
Techniques and solutions
We utilized natural language processing (Stanza), web scraping, BeautifulSoup, and text classification to build an automated system that generates leads for Costanoa to follow up on.