We need a bigger goal than collecting data
summary
Cloud storage and cheap collection let teams turn on the firehose and hoard metrics but scalability has a cost: Netflix-level two million time series, Spotify 170 million metrics and Google’s billions of measurements, yet engineers report 95% of metrics never looked at and some orgs spend hundreds of thousands on AWS just to keep monitoring data. Argues for measuring ROI, tracking time to actionable insight, pruning useless signals, and shipping lightweight alerts and dashboards that actually speed problem resolution.
AI-written summary · about
chapters (5)
related talks
38mThe Ticking Time Bomb of Observability Expectations
53mYour Attention, Please: Optimizing Metrics for Humans
24mMetrics-driven development
27mHow to be data-driven when you aren’t Netflix (or even if you are)
24mA Practical Approach To Unlock Value From Data and Analytics
30mHow to collect application metrics if it needs to be done yesterday
41mRethinking Measurement for High-Performing Teams
52mScaling Data Teams: Three Pitfalls and How To Avoid Them