Evolving Incident Metrics: Why More Incidents Can Signal Healthier SRE Culture

A recent InfoQ article, titled "More Incidents Don't Necessarily Mean Less Reliability," challenges the conventional wisdom that an increasing number of reported incidents inherently signifies a decline in system reliability. Instead, it posits that a rise in incident counts can be a positive indicator of a maturing incident management culture within an organization. The article suggests that as engineering teams adopt better processes, tooling, training, and operational discipline, they become more willing to formally declare incidents that might have previously been handled informally or even hidden altogether. For SRE practitioners and engineering leaders, this perspective is critically important. It directly addresses the often-stressful situation where improved visibility into operational issues leads to seemingly worse metrics. Understanding this nuance allows teams to move beyond simplistic incident counts and focus on more meaningful indicators of reliability, such as recovery effectiveness, customer impact, and organizational learning. This insight empowers SREs to advocate for transparency and blameless post-mortems without fear that increased reporting will be misconstrued as a failure. Ultimately, this shift in mindset can foster a healthier, more resilient engineering culture that prioritizes learning over blame. This insight aligns with a broader, well-established trend in reliability engineering that emphasizes learning from failures and moving beyond superficial metrics. Modern SRE practices have increasingly advocated for a shift from asking "how many incidents occurred?" to more pertinent questions like "how quickly were users affected?" and "how rapidly was service restored?" The focus has progressively moved towards understanding the underlying causes, improving recovery processes, and preventing recurrence, rather than solely minimizing the raw number of incidents. This evolution is also reflected in the growing adoption of blameless post-mortems and the recognition that incidents are invaluable opportunities for systemic improvement. In practice, SRE teams and leadership should critically evaluate their incident metrics. Instead of penalizing teams for higher incident counts, leadership should actively reward transparency and thorough incident reporting. SRE teams should invest in robust incident response frameworks that encourage early declaration, clear communication, and comprehensive post-incident reviews. The overarching goal should be to cultivate an environment where operational knowledge is visible and repeatable, leading to continuous improvement across the system. This also means placing greater emphasis on metrics like Mean Time To Restore (MTTR), Mean Time To Detect (MTTD), and the quality of post-mortems, as these better reflect an organization's true resilience and learning capacity in the face of inevitable system challenges.