Data Science and Not-So-Big Data
Statisticians joke (yes, they do!) that half the job is wrestling the data into shape, half is performing useful analyses, and half is explaining the findings. The data wrestling too often expands to consume all my time, leaving me too ennervated to find any findings.
Big Data is what got everyone so excited about data science. But Bigger isn’t necessarily Better. I have interviewed many many people, and every single one is groaning about data quality. As my friend Jonas Dahl (a data scientist at Adobe) says, “Too much data can actually be crippling for an organization because it slows down our analyses. Just because you have a lot of data doesn’t mean that you have a lot of knowledge or insights. Organizations are too often obsessed with recording enormous amounts of data and not obsessed enough about turning this pile of data into actionable knowledge.
“Bottom line, you can’t get too much high-value data — but you can be buried in too much low-value data.”