Learn the vocabulary of keeping a paging system's alerts meaningful rather than desensitizing.
0 / 5 completed
1 / 5
At standup, a dev mentions on-call engineers gradually becoming desensitized to paging alerts because they receive so many low-value, non-actionable pages that a genuinely critical one starts blending in. What is this phenomenon called?
Alert fatigue describes on-call engineers gradually becoming desensitized to paging alerts after receiving so many low-value, non-actionable pages that a genuinely critical one starts blending in with the noise. Responding with equal urgency to every alert regardless of volume ignores how real human attention naturally degrades under a heavy, mostly-noisy paging load. This fatigue is what makes an unfiltered, overly broad alerting policy actively dangerous rather than merely inconvenient.
2 / 5
During a design review, the team wants to raise alert thresholds and add proper deduplication so only a genuinely actionable condition triggers a page, rather than paging on every minor anomaly. Which capability supports this?
Alert tuning through threshold adjustment and deduplication ensures only a genuinely actionable condition actually triggers a page, rather than paging on every minor anomaly a monitoring system happens to detect. Paging on every anomaly regardless of severity floods an on-call engineer with noise that erodes their ability to notice a truly urgent signal. This tuning is what keeps a paging system's signal-to-noise ratio high enough to remain trustworthy.
3 / 5
In a code review, a dev notices alerts are tiered by severity, so only the most urgent condition actually pages an on-call engineer while a lower-priority one becomes a ticket or a log entry instead. What does this represent?
Alert severity tiering routes only the most urgent condition to an on-call page, while a lower-priority condition becomes a ticket or a log entry that doesn't interrupt anyone immediately. Routing every condition directly to a page regardless of severity treats a minor issue with the same urgency as a genuine emergency. This tiering is what keeps a page reserved for something that genuinely needs an engineer's immediate attention.
4 / 5
An incident report shows a genuinely critical alert went unnoticed for over an hour because it was buried among dozens of noisy, non-actionable pages the on-call engineer had learned to mentally filter out. What practice would prevent this?
Tuning alert thresholds and severity tiering ensures only an actionable, genuinely urgent condition pages the on-call engineer, keeping a real emergency from being buried among noise. Paging for every anomaly regardless of severity is exactly what causes an engineer to mentally filter out most pages, including one that actually matters. This tuning and tiering is essential to keeping a paging system's most urgent signal reliably noticed.
5 / 5
During a PR review, a teammate asks why the team invests effort in reducing alert volume and tiering severity instead of just paging on every anomaly a monitoring system can detect. What is the reasoning?
Paging on every anomaly erodes an on-call engineer's trust in and response speed to alerts overall, since a mostly-noisy paging history teaches them to treat any single page with less urgency over time. A curated, tiered set of alerts keeps a page reserved for something genuinely actionable, preserving that trust. The tradeoff is the ongoing effort of tuning thresholds and reviewing tiering as a system's normal behavior and failure modes evolve.