Michael De Nil is the CEO and cofounder at Morse Micro—reinventing Wi-Fi for IoT.gettyThe Internet of Things (IoT) has been defined by a simple idea: Connect devices, collect data and act on it later. That model worked well for early use cases, smart meters, environmental sensors and basic asset tracking, but it’s increasingly misaligned with how modern systems operate.A new phase is emerging—often referred to as IoT 2.0—where connected systems are expected not just to report data, but to interpret it, act on it and do so in real time. This shift is being driven by the convergence of edge computing, artificial intelligence (AI) and the need for faster operational decision-making across industries. The implication is straightforward but significant: Connectivity alone is no longer enough. The underlying network must now support richer data, lower latency, higher device density and seamless interoperability.From ‘Pings’ To Processing In Real TimeEarly IoT deployments were built around low-bandwidth, intermittent communication and what many engineers refer to as “sensor pings.” A temperature reading every few minutes or a location update every few hours was sufficient. Today’s systems look very different. Modern deployments increasingly rely on video and audio streams for monitoring and analytics, edge AI models for anomaly detection and automation, and closed-loop control systems requiring near-real-time feedback. These applications require consistent network responsiveness and reliability, with network speeds of megabits per second rather than kilobits. In practical terms, this means organizations must rethink not just their devices and software, but the entire connectivity layer underpinning their systems.The Limits Of Legacy Connectivity ModelsMany of the connectivity technologies that enabled early IoT were never designed for this new reality. Low-power, wide-area networks (LPWAN), such as LoRaWAN, excel at long-range, low-energy communication. They remain highly effective for simple, infrequent data transmission. However, their architectural trade-offs, including limited throughput, higher latency and constrained real-time capability, make them less suited for data-rich or interactive applications.Cellular technologies, including LTE and 5G, address some of these limitations by offering higher bandwidth and broad coverage. But they introduce their own challenges such as recurring costs, higher power consumption and dependency on operator infrastructure. For many localized deployments, including factories, campuses or agricultural sites, this can be excessive both economically and operationally.The result is a fragmented landscape where organizations often combine multiple technologies, including short-range wireless, LPWAN, cellular and proprietary systems, to meet their needs. While functional, this approach increases complexity, cost and long-term maintenance burden.A Shift Toward Localized, High-Performance NetworksOne of the most important, but often overlooked, shifts in IoT 2.0 is architectural. Many deployments don’t actually need wide-area networks. Instead, they require high-performance, localized networks capable of covering large physical areas such as factories, warehouses, farms or smart city zones, while maintaining reliability and low power consumption. This is leading to a renewed focus on LAN-based architectures, extended beyond the confines of traditional buildings.In this context, newer wireless approaches are emerging that aim to bridge the gap between short-range Wi-Fi and wide-area cellular. For example, technologies like Wi-Fi HaLow (based on IEEE 802.11ah) extend Wi-Fi into sub-GHz spectrum, enabling longer range and better propagation while maintaining IP-native networking and standard security models. However, like any emerging standard, adoption, ecosystem maturity and regional spectrum differences remain considerations.The broader point is not about any single technology, but about a design principle: Connectivity should align with the physical and operational boundaries of the system. Not every deployment needs a WAN; many benefit more from a scalable, high-performance LAN.Best Practices For Designing IoT 2.0 SystemsAs organizations move toward more intelligent, data-driven deployments, several best practices are emerging:1. Design For Data, Not Just DevicesStart with the type of data your system needs to handle, be it video, sensor fusion or AI outputs, and work backward. Connectivity decisions should reflect bandwidth, latency and reliability requirements, not just coverage.2. Minimize Architectural ComplexityAvoid stitching together multiple networking layers unless absolutely necessary. Each additional protocol, gateway or translation layer increases cost and operational risk.3. Prioritize InteroperabilityStandards-based technologies, particularly those built on IP, simplify integration with existing IT systems, cloud platforms and future devices. Proprietary ecosystems may offer short-term benefits, but can limit long-term flexibility.4. Evaluate Total Cost Of Ownership (TCO)Connectivity decisions are often evaluated based on upfront cost, but ongoing expenses—maintenance, power consumption, infrastructure density and subscription fees—can be more significant over time.5. Align Connectivity With Deployment ScaleA solution that works for 50 devices may not scale to thousands. Consider device density, spectrum efficiency and network management from the outset.The Bigger Picture: From Connectivity To CapabilityThe transition to IoT 2.0 is less about any one protocol and more about a fundamental shift in expectations.Connected systems are no longer passive. They are becoming:• Context-aware, interpreting multiple data inputs• Autonomous, making decisions at the edge• Scalable, supporting thousands of devices in a single environmentThis evolution requires a corresponding shift in how we think about connectivity, from a basic utility to a strategic enabler of system capability. Technologies will continue to evolve, and no single approach will fit every use case. LPWAN, cellular and next-generation WLAN solutions each have roles to play. The key is selecting the right tool for the job, based on application requirements rather than legacy assumptions.What’s clear is that the future of IoT will be defined by what devices can actually do once they are connected.Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?