Generating MoM, YoY and CMGR from SQL
We’re not really big enough to claim that what we needed was a Data Scientist, but that is effectively what we were looking for. And as is our wont, we created a simple test for applicants, designed to show some ability to extract meaning from data (by creating a histogram), and a working knowledge of SQL. We do most of our running reporting through Periscope, so being able to generate flat datasets from our relational database is an essential part of the job.
Testing candidates is always revealing, but in this case especially so – out of perhaps 20 candidates, only two nailed it – the rest either got the wrong answer, or the right answer the wrong way (for example, joining the two tables in Excel, using a vlookup function.) This got me thinking – is raw SQL becoming a lost art?