Consumer's decision has two stages. The first stage is whether or not to drink/eat the product. Then, the second stage is how much to dink/eat the product if they decide to drink/eat it. What factors affect the first stage decision as to whether or not to drink/eat the product? What factors affect the second stage decision as to how much to drink/eat the product? These will differ by stage and by product. Differences of affecting factors to consumer's two stage decision by product could give us useful information to examine the tendency of fluid milk saturation and a possibility that decreases in fluid milk consumption could be replaced by increases in manufactured dairy products. With consumer survey data on consumption of fluid milk, yogurt drinks, and cheese in Japan, the Heckman's two-step estimators of the type Ⅱ tobit model for each of the four products indicated that higher evaluation on milk utility and higher health concerns are more important determinations on how much to eat yogurt and cheese than on how much to drink fluid milk (including yogurt drinks). This would be an evidence of the fact that peaple with good images for milk are now increasing yogurt and cheese consumption instead of fluid milk, or that fluid milk consumption is replaced by yogurt and cheese. Because non-drinks of fluid milk is already only about 10%, persuading current drinkers to drink more fluid milk is important. However, our estimation results indicate that there are few factors we could use to motive them to increase fluid milk consumption because most factors are insignificant for deciding how much to drink fluid milk. This would be an evidence of saturation of fluid milk cnosumption in Japan.