Before I get into the nitty-gritty details of designing, planning and executing a study, let's cover some misconceptions people have about ethnography.
It is qualitative: It's a common misconception that data from an ethnographic study is only qualitative. However, using methods like surveys or even conducting large-scale social data science projects, we generate quantitative data. I have not had much experience with quantitative ethnographic studies. However, I recognize that it holds help explain or explore the intensity of the impact of cultural phenomena across large groups of people.
Participant data is all we need: No matter how extensive an ethnographic study is, data always needs to be triangulated. This means using data points from multiple sources that support the claim or insights more rigorously. So, even after gathering a substantive amount of ethnographic data, always look to back it through secondary research, expert interviews, or market analyses.
There is an ideal sample size: Having designed and executed several research projects, I have learned that there is no simple way to determine the appropriate sample size for a study. It is a combination of what is possible and what is necessary for the data to be saturated, i.e., ideas to reoccur often. For one of the projects I was involved in for a large tech business with billions of users, we only talked to 16 people and yet felt confident in drawing novel insights from the data.
We should have all data about participants: If you are doing a qualitative ethnographic study, you are never going to feel like you have enough data. In fact, it's unethical to go in knowing you want to extract more than you need from respondents. It's normal to feel like you don't know everything about your participants. It is not a flaw in your research design or execution -- it's just the nature of the method. It is not just the data but how it is analyzed that also contributes to the research outcomes.