Matchmaking is an age-old practice in the Indian context. What started out as a tradition of seeking partners in close-knit communities, has today expanded to newspapers and online websites. These advertisements however continue to be extremely explicit, content-heavy, and rich sources of cultural and socio-economic individual preferences. In this study, I attempt to use this text, along with computational methods, to understand the most important preferences of individuals seeking partners through newspaper advertisements, and how their preferences vary with time. Moreover, I attempt to qualitatively understand how people employ print versus digital media to facilitate the matchmaking process, and how this impacts the relevance of newspaper matrimonial advertisements. I use robust methods of natural language processing and computational linguistics such as part-of-speech tagging and word embeddings to conduct a longitudinal study across years. I argue that individual preferences of brides and grooms are rooted in gender and cultural biases, and remain constant with time. Using newspaper matrimonial advertisements of a north Indian newspaper, The Tribune, from 2001-2014, I find that there exists a bias for “tall”, “slim”, “fair”, and “beautiful” women, and “handsome”, “tall”, and “intelligent” working men, from the Sikh or Jat community. This underscores the fact that these advertisements are embedded in stereotypical gendered roles and a strong emphasis on intra-faith alliances. The results of this research also highlight the increasing preference of individuals towards using online platforms for matrimonial advertisements. In doing so, this study discovers nuanced biases in the social fabric of Indian matchmaking traditions and raises questions about the relevance of newspaper matrimonial advertisements and their implications on the matchmaking ecosystem.