Data cleaning, or "wrangling" as some like to call it is an important aspect of the data analytics process. I've seen analysts spin big Hadoop clusters just to clean their data. Though at times that's necessary, usually, you can go pretty far with just SQL.
