Even with near universal adoption of CI/CD, CloudBolt Research finds K8s right-sizing still stays human-in-the-loop
CloudBolt Software, a recognized leader in Kubernetes optimization and hybrid cloud management, today announced the release of its latest CloudBolt Industry Insights (CII) report, “The Kubernetes Automation Trust Gap No One Talks About.” The research highlights a core constraint in modern cloud operations: enterprises increasingly trust automation to ship code, but still keep Kubernetes right-sizing changes human-controlled in production.
Based on a survey of 321 Kubernetes practitioners at organizations with 1,000+ employees, the report finds that automation is widely considered foundational—89% say it is mission-critical or very important. But when automation touches CPU and memory decisions in production, delegation to automation drops sharply. Only 17% report operating with continuous optimization.
Also Read: CIO Influence Interview with Gihan Munasinghe, CTO of One Identity
“Everyone says they trust automation right until it requires the authority to act,” said Mark Zembal, Chief Marketing Officer at CloudBolt. “Teams will auto-deploy code via CI/CD 50 times a day without blinking an eye. But the moment automation touches cost, performance, or reliability in production, hesitation creeps in. That hesitation is where delegation dies. Teams won’t hand over the keys unless the system is explainable, bounded by guardrails, and reversible on demand.”
Key Findings
- Automation is doctrine in delivery: 89% say automation is mission-critical or very important; 59% deploy to production automatically without manual approval.
- Delegation drops for right-sizing in production: 71% require human review before applying resource optimization; only 27% allow guardrailed auto-apply for CPU/memory changes.
- Manual control doesn’t scale: 54% of respondents run 100+ clusters, and 69% say manual optimization breaks down before ~250 changes per day.
The Cost of Caution
The findings point to a familiar pattern across enterprise Kubernetes environments. Teams have invested heavily in visibility, dashboards, and recommendation engines. But visibility without the ability to act safely at scale creates its own problem. Organizations know they’re overprovisioned. They’re choosing to absorb that cost because the alternative, letting automation touch production resources without sufficient guardrails, rollback, and explainability, feels riskier than the waste. That tradeoff is rational at the individual team level. But at the organizational level, it compounds. And it doesn’t get better with scale, it gets worse.
From Insight to Automated Delegation
The survey data suggests the next phase of Kubernetes optimization won’t be won by better visibility alone. Teams already see the problem. What they’re asking for is a credible path from recommendation to delegation: automation that earns trust incrementally, operates within SLO-aware boundaries, and can be reversed instantly when something goes wrong. 48% of respondents said visibility and transparency would most increase their trust; 25% pointed to proven guardrails; 23% said instant rollback. The demand isn’t for more automation in the abstract. It’s for automation that respects the operating reality of production.
“If you map all of this to a maturity model, it creates a clear continuum from Observe to Advise to Automate to Trust. Most companies are stuck in the early middle,” said Yasmin Rajabi, CloudBolt’s Chief Operating Officer. “They can see the problem. Some can even accept recommended fixes some of the time. But they stop short of letting the right-sizing system act autonomously. The final stage isn’t more insight, it’s trust. And until teams trust automation to optimize right-sizing in production, they will always be constrained with manual limitations that can never effectively scale.”
Catch more CIO Insights: CIO as Orchestrator of Cross-Functional Digital Strategy
[To share your insights with us, please write to psen@itechseries.com ]
