When enterprise teams migrate off a log-heavy observability stack, the first instinct is almost always the same: keep everything as logs. It feels safe. But at scale, querying logs for everything gets expensive fast and for high-volume messaging infrastructure, that model does not always hold up.

This talk shares a real-world perspective from an observability migration for a Fortune 50 company, focusing not only on the technical implementation but also on the challenges that shaped it: the friction of moving teams away from a log-centric mindset, the practical limitations that influenced what was possible, and the open questions that remain around how teams change the way they reason about telemetry.

This session focuses on the practical decisions, tradeoffs, and lessons that surfaced during a live enterprise migration.

Takeaways:

-What was learned from applying a log-to-metric approach to high-volume enterprise infrastructure

-Real limitations and tradeoffs encountered during an AppDynamics and Splunk migration

-Honest reflections on the challenge of shifting a team’s log-centric mindset