As a data scientist (startup / Yahoo) turned product manager (4 startups), I've used a variety of stacks in the past - from plain SQL, to Mode, to Mixpanel, Loo

As a data scientist (startup / Yahoo) turned product manager (4 startups), I've used a variety of stacks in the past - from plain SQL, to Mode, to Mixpanel, Looker, Interana, and Hive. Recently we started using PopSQL (https://popsql.com https://popsql.com) and love it. When I don't have a dedicated data team, my philosophy is: 1) Make it difficult to get wrong answers - Don't use Google Analytics. It's too easy to generate incorrect charts, and too difficult to verify them. - Have a limited sandbox of reports for non-SQL writers - Keep the SQL close to the report, so it's easy to verify the underlying query. - Push people to learn even basic SQL 2) Make it quick and easy to ask iterative questions - PopSQL is way faster than Mode. Like 20x faster. 3) For metrics that matter (e.g. KPIs), instrument them directly and even build a custom analytics dashboard if it's important. (beware dashboard clutter! https://twitter.com/andrewchen/status/1193619877489192961 https://twitter.com/andrewchen/status/1193619877489192961 )