For IT and security leaders, the real question is how to architect that mix, deciding which systems stay on-premises, which move to the cloud, and which partners can build it well. Recent insights from IDC's Cloud Video Surveillance Survey, sponsored by VIVOTEK Inc., points to a fairly consistent pattern already forming.

Why getting the split right matters

The payoff shows up in the same data that explains why organizations move surveillance workloads to cloud or AI-based systems in the first place. Improved detection accuracy is the top-cited driver, at 50%, ahead of centralized and remote management across sites (42%), faster access to new features and upgrades (39%), and lower total cost of ownership (35%), as shown in Figure 2.

Each of those benefits depends on putting the right workload in the right place. Centralized management only pays off if the systems being centralized are integrated. Lower total cost of ownership works the same way. Say, an organization builds an on-premises system that follows data residency rules. Later, it moves part of that system to the cloud without checking whether the same rules still apply. That mistake creates a compliance problem, and fixing it costs extra time and money. The organization ends up paying for the same work twice: once to build it, and once to fix it. This is why architecture decisions are not a small technical detail. They are a central part of the business case for modernizing at all.

Core systems integrate first, everything else follows

Organizations are not unifying their entire security stack at once. Video surveillance and access control lead integration into a single platform, with 47% and 46% of organizations, respectively, describing these systems as fully integrated today. Alarms and intrusion detection follow at 41%. Newer or adjacent capabilities lag well behind: environmental sensors sit at 32% fully integrated, guest and visitor management at 37%, and robotic or drone-based inspection systems at 29%, the lowest of any category measured.

Organizations unify core physical security functions, the ones with the clearest safety and compliance stakes, with centralized management and cloud analytics first; peripheral capabilities follow once that core platform is proven.

Some processing stays on-premises for a reason

On-device or edge-level video processing, which limits how much raw data ever leaves a site, is a deliberate design choice for many organizations, not a stopgap while they wait to move to the cloud. It ranks as the second most common privacy-preserving priority today, at 43%, behind only automated data retention controls at 49%. Regional data residency requirements register at 37%, and data security, privacy, and compliance concerns are the top-ranked barrier to full cloud migration overall, at 45%. For organizations handling sensitive video under strict retention or residency rules, keeping processing local is often the correct long-term architecture, driven by compliance requirements rather than technical limits. That has a practical implication for budgeting and vendor selection, in that organizations should treat on-premises processing as a permanent line item, edge architecture weighted as heavily as cloud capability when comparing vendors and partners.

What organizations look for in a partner

Selecting a partner for a hybrid environment comes down to three requirements. Demonstrated experience and reference cases in similar environments rank highest, and buyers want proof a partner has built something like this before, with evidence of success. Roadmaps are important; however, the ability to deliver real capability within industries is more important than where the technology is headed next. Close behind is a track record of reliability and long-term financial stability, since a hybrid architecture is not a one-time project but an ongoing relationship that has to hold up over several years of change on both the on-premises and cloud side. Third is proven integration and interoperability with the IT and OT systems already in place.

All three point to the same underlying concern: organizations want a partner who can prove they will work within the environment that already exists and stay for the long haul, not one asking to start over or move faster than the buyer is ready for.

The practical takeaway

Architecting a hybrid surveillance environment well comes down to three decisions the survey data already supports: unify core security systems before extending integration to peripheral ones, keep privacy-sensitive processing at the edge by design, and choose partners based on proven interoperability with what you already run rather than the most ambitious roadmap. Getting there is less about resolving cloud versus on-premises as a single, final decision, and more about treating the split itself as the thing to get right.

Guest IDC Blogger

Stephanie Krishnan

Associate Vice President at IDC

Stephanie Krishnan is responsible for shaping research across manufacturing, supply chain, and retail, with particular focus on Asia/Pacific and worldwide supply chain strategies. She covers transportation, warehousing, logistics, global trade, warehouse and inventory management, and integrated supply chain execution, alongside emerging themes such as agentic AI, physical AI, cybersecurity, and sustainability. Through market intelligence, survey-based research, and vendor assessment, she helps organizations strengthen resilience, improve orchestration across multi-enterprise networks, and identify technology priorities that drive operational and business outcomes.