In order for data to be useful for analysis and visualization, it often needs to be organized and formatted in particular ways. Organization can include both procedural cleaning up of errors or mistakes and processing or transforming the data through calculations and logic statements to create new or summative measures.

The design of a data investigation is as important as the data collection process. Framing a data-based investigation requires identifying a problem or question to be explored. Additionally, the methods must be carefully chosen and the values and tradeoffs considered.

The methods and decisions made during data collection affect the usefulness of the data and its ability to answer different questions. It is important to consider the potential effects of methodological decisions when collecting data and to determine the methodological decisions made by others when using secondary data. It is also important to consider ethical practices of using other's data.

Data collected from real world scenarios is often complex across many dimensions including messiness, size, and structure. In order to be able to work with authentic real-world datasets of high complexity, these dimensions must be scaffolded such that increasingly higher levels of complexity are encountered as one approaches mastery.