Don't trust the numbers!
summary
Argues that web analytics help reveal user flows but often mislead: visit counts depend on cookies and timeout settings, time-on-page systematically undercounts last-page dwell time (example shows reported 2.5 minutes vs actual 7.5), bounce rates mean different things for landing versus content pages, and chunking inflates page-view totals. Lays out a reproducible process — define the business problem, extract and scrub data, visualize trends — and recommends validate data with qualitative signals and avoid cross-product comparisons before acting on metrics.
AI-written summary · about
chapters (8)
- 0:00Introduction to analytics
- 1:12Benefits and pitfalls
- 3:28Data collection metrics
- 7:19The analytics process
- 11:39Interpretation guidelines
- 18:32Analyzing specific metrics
- 28:01Setting targets and wrap up
- 31:16Q&A session
related talks
52mAB-tests, lies, damned lies, and statistics.
12mHow Usage Analytics Can Improve End User Experience?
48mHow to Use Analytics to Optimize Your Product
29mThe Power of Analytics in Game Development for Meta Horizon Worlds
40mPanel Discussion: What Users Really Want
06mPerformance is about more than time series charts
51mFairytale vs. Narrative: Telling stories and making decisions with data
48mStanding Up for the Unmeasurable in The Design Process