Anomaly Root Cause Analysis 101
<p>We use metrics and KPIs to monitor the health of our products: to ensure that everything is stable or the product is growing as expected. But sometimes, metrics change suddenly. Conversions may rise by 10% on one day, or revenue may drop slightly for a few quarters. In such situations, it’s critical for businesses to understand not only what is happening but also why and what actions we should take. And this is where analysts come into play.</p>
<p>My first data analytics role was KPI analyst. Anomaly detection and root cause analysis has been my main focus for almost three years. I’ve found key drivers for dozens of KPI changes and developed a methodology for approaching such tasks.</p>
<p><a href="https://medium.com/towards-data-science/anomaly-root-cause-analysis-101-98f63dd12016"><strong>Read More</strong></a></p>