In general, my preference is to reframe an increase in operational load as constructive forward looking system improvements (availability, utilization, documentation, incident responsiveness, latency) that can be prioritized against feature deliverables. Moving from resolving a "why was my request slow" ticket to "improve E2E p99 performance by NN%". (Insert me handwaving disclaimers and situational factors here...).
Mastodon Source 🐘
Reframing this way makes operations less a time tax than a peer of functional capabilities. Operations is part of design and daily work. There are usually systemic factors beneath isolated tickets. Increasing ops capacity can help drive those metrics down and the types of improvements required to affect the conditions producing those tickets often can't be achieved in rotation windows/dealing with the tickets.
Mastodon Source 🐘
Only operations produces observable data. Functional improvements (until they are delivered and verified) have exclusively speculative impact.