Netlify insights

A sudden increase in bandwidth consumption driven by a small number of URLs or content types indicates oversized assets, missing compression, or bot traffic targeting specific resources.

A concentrated spike in 404 errors on specific URLs following a deploy indicates broken links, incorrect routing configuration, missing files, or incorrect publish directory settings introduced by code changes.

An increase in cache miss rate for previously well-cached endpoints indicates cache invalidation issues, configuration changes, or dynamic content serving where static content is expected, leading to increased origin load and latency.

When a single client type (bot, crawler, or specific user agent) accounts for a disproportionate share of requests, it can indicate bot/crawler activity, potential abuse, or misconfigured automation consuming resources and bandwidth.

When Real User Monitoring metrics (LCP, FID, INP, CLS, FCP) show significant degradation on mobile devices compared to desktop, it indicates mobile-specific performance issues like oversized assets, render-blocking resources, or layout instability on smaller viewports.

A sudden increase in function invocations or errors immediately following a production deploy indicates a code change may have introduced performance issues, infinite loops, or unexpected retry behavior.

When p95 and p99 function duration metrics are significantly higher than average duration, it indicates a subset of function invocations are experiencing extreme latency, often due to cold starts, resource contention, or specific input patterns.

When form submissions are being received but not appearing in the dashboard, they may be silently filtered as spam by Netlify's Akismet integration, especially during testing with fake data or repeated submissions from the same IP.