## Metadata - Author: [[Commoncog|Commoncog]] - Full Title:: Two Types of Data Analysis - Category:: #🗞️Articles - Document Tags:: [[Data team vision and mission|Data Team Vision And Mission]], [[Data team vision and mission|Data Team Vision And Mission]], - URL:: https://commoncog.com/two-types-of-data-analysis/ - Finished date:: [[2024-02-05]] ## Highlights > understanding variation is the foundation to becoming data driven ([View Highlight](https://read.readwise.io/read/01hnvwr0etzrat4wnhtnp94trj)) > Good product teams tend to prioritise the acquisition of knowledge. Bad product teams prioritise throwing features over the wall ([View Highlight](https://read.readwise.io/read/01hnvxaskfyxkc1wfgnqy7py60)) > if the behaviour change is subtle, this may not show up in my metrics! If I had run an A/B test, on the other hand, I would likely be able to pick up on subtler changes, assuming my experimental groups are sufficiently large. ([View Highlight](https://read.readwise.io/read/01hnvxevvg4qmmgpmdh02bmf66)) > Tiny, barely detectible changes aren’t worth pursuing when you have such small numbers ([View Highlight](https://read.readwise.io/read/01hnvxf4rd24hek9aq3fac4p9v)) > sometimes tiny changes are worth doing (e.g. you are Google, or Instagram, and a 2% change is literally worth millions of dollars). That’s when analytical methods from Experimental Studies really shine ([View Highlight](https://read.readwise.io/read/01hnvxfzhsrj11cvm3a19yyyth))