Cleaning and standardizing are easily the biggest "small" problems. With small-ish data sets (<= a thousand records), I often find it's easier to have a cheap c

Cleaning and standardizing are easily the biggest "small" problems. With small-ish data sets (<= a thousand records), I often find it's easier to have a cheap contractor do the work by hand than to spend time writing a programmatic solution.