Within marketing research, big data is often described as being “census” data for the population that it represents. The devil is in the details and when we take a closer look we can see that this isn’t the case. There are many situations that are not captured within the population that big data purports to be a census of. Big data isn’t even a census of itself since it’s not uncommon for records to be excluded either by accident during the collection process or by design in the cleaning processor. Unfortunately, our industry is so enamored with the size of big data that some users of data are willing to trade off precision for tonnage. Fortunately, if the shortcomings of big data are understood and corrected it can accurately represent the population that it measures in the correct proportion to the universe. We will discuss a method that Nielsen has developed called “Common Homes” that is designed to identify and correct the shortcomings of big data sets that represent media consumption.