In an increasingly digitalized world, significant levels of potentially valuable data are available with more continuously being produced all the time. However, with this opportunity there is a significant challenge to manage it and to differentiate between smart relevant and non-relevant data. The research industry provides a good example where one of the goals is to identify relevant and valuable insights in consumer and product research using this smart data. Our goal was to conduct a study on a research project using 'big data' and to compare the outcome of the analyses with traditional survey research. Online shopper ratings and review data in social media is an exciting and 'on trend' data source and was compared to traditional survey data. The survey data included product tests, i.e. products were placed in-home and consumers evaluated the products using a standardized questionnaire. The objective was to derive substantive insights about the core drivers for a five-star-rating of consumer reviews in online shops or platform ratings for pet care products and compare these insights with drivers of liking from traditional research on pet care products already existing in the market. We validated the hierarchy of drivers for the overall product rating by conducting a meta analyses on previous product tests and assessed the drivers of overall liking.