For the past several years, cloud computing has been an attractive option for small and midsize businesses. With limited IT staff and tight budgets, many organizations embraced cloud-first strategies to avoid managing physical infrastructure while gaining access to scalable, on-demand resources.
But the rapid rise of generative artificial intelligence and agentic AI is prompting a rethink. As small businesses move AI initiatives beyond experimentation and into day-to-day operations, they’re discovering that cloud isn’t always the best long-term fit for every workload.
Instead, many IT leaders are re-evaluating where their data lives and how it’s managed. That’s leading to growing interest in hybrid strategies and, in some cases, bringing select workloads back on-premises.
Cloud repatriation may sound like a drastic shift, but according to Eryn Brodsky, principal enterprise strategist for CDW, it’s more nuanced than that. “It isn’t necessarily about bringing everything back on-prem,” Brodsky says. “What we’re really talking about is rebalancing — evaluating your overall data center and cloud footprint and determining which workloads belong where.”
For small businesses, that kind of targeted approach is especially important. Rather than following blanket mandates, IT leaders are focusing on practical decisions that balance performance, cost and manageability.
SentinelOne Control
Prevention and detection of attacks across all major vectors with fully automated, policy-driven response capabilities and real-time forensics.




