What stands out most is how fast and flexible querying is, even against high cardinality data we can slice and dice by any field (user ID, request ID, custom attributes) without needing to predefine dashboards or metrics ahead of time. This has made debugging production issues dramatically faster, since we can trace a single problematic request end-to-end instead of guessing from aggregated metrics. The BubbleUp feature is especially useful for quickly surfacing what's different about a slow or failing subset of requests compared to the rest of traffic. It's also made our on-call experience much less stressful, since we can investigate incidents in real time with actual context instead of jumping between multiple tools. Review collected by and hosted on G2.com.

The learning curve can be steep for teams new to observability-driven debugging, since it requires a shift in mindset from traditional dashboards and predefined metrics toward ad-hoc, high-cardinality querying. Instrumenting our codebase properly to get rich, structured events took real upfront effort, and getting the most value out of the platform depends heavily on how well events are tagged and enriched at the source. Pricing can also become a consideration as event volume grows, particularly for larger or high-traffic services. Additionally, some newer team members find the query interface less intuitive at first compared to more traditional metrics dashboards, so onboarding takes a bit of ramp-up time. Review collected by and hosted on G2.com.