CSE 135 Lecture 13: L13 2/8/19
8 Feb 2019
School
Department
Course
Professor
Document Summary
Question: collect - what, how, store - db, file system, retrieve - clean, sample , etc, analyze - visualize, report. Make a change, do the whole process again data-driven thinking. Reality: we mostly guess wag (wild-ass guessing) Careful of the testing effect : that whoever you"re testing is not aware of what"s going on. Analytics can be a site designer"s best friend if we have the right attitude. Analytic results you can actually analyze and do something with. In general, task-drive sites or apps tend to be easier to show value than informational driven projects. Falsehood - informational sites do not have conversions and obvious analytics. Analytic approaches (where to collect?: log file analysis. Pros: gets everything bots, single images, etc. Cons: it focus, no client insight! (not really a lot of information: script tag. Cons: script off, bad data, misses some objects: network sniff. Pros: very transparent, gets everything and good for large farms, insight to network.
