As an IoT platform and middleware analyst, I am asked constantly about the benefits of IoT platforms and \u201cwhat makes a great IoT platform.\u201d In response, I often ask these curious inquirers if they\u2019ve ever used IoT platforms themselves. Walking on the edge is exhilarating, but having hands-on insights, data and expertise on how to survive the journey is even better.\nWhat do users actually experience when they use IoT edge platforms?\nIoT edge computing is a technology architecture that brings certain computational and analytics capabilities near the point of data generation. IoT edge platforms provide the management capabilities required to deliver data from IoT devices to applications while ensuring that devices are properly managed over their lifetimes. Enterprises use edge platforms for factory automation, warehousing\/logistics, connected retail, connected mining and many other solutions. With IoT platform revenue slated to grow to USD63.4 billion by 2026, IoT edge is one of the most highly relied upon enterprise IoT platform approaches.\u00a0\nEnterprises spend a tremendous amount of time completing edge-related IoT platform activities. According to hands-on tests of IoT platforms in MachNation\u2019s IoT Test Environment (MIT-E), the majority of an enterprise user\u2019s edge-related time is spent creating visualizations to gain insight from IoT data. 35% of a user\u2019s time is spent creating dashboards with filtered alerts. And a combined 16% of a user\u2019s time is spent viewing sensor data for an individual device (8%) or a group of devices (8%). Data from an IoT platform are critically important, so the ability to assemble dashboard sensor data and alerts are key \u2013 expect to spend a lot of time doing it.\nSince the edge is critical for enterprises deploying IoT solutions, we\u2019ve identified the top five user requirements of IoT edge platforms, based on IoT platform users\u2019 experiences with these platforms.\n1. Pick a platform with extensive protocol support for data ingestion\nTo seamlessly bring data from devices into the edge platform, enterprises should choose leading IoT platforms that support an extensive mix of protocols for data ingestion. The list of protocols for industrial-minded edge platforms generally includes brownfield deployment staples such as OPC-UA, BACNET and MODBUS as well as more current ones such as ZeroMQ, Zigbee, BLE and Thread. Equally as important, the platform must be modular in its support for protocols, allowing customization of existing and development of new means of asset communications.\n2. Ensure the platform has robust capability for offline functionality\nTo ensure that the edge platform works when connectivity is down or limited, enterprises should choose leading IoT edge platforms that provide capabilities in four functional areas. First, edge systems need to offer data normalization to successfully clean noisy sensor data. Second, these systems must offer storage to support intermittent, unreliable or limited connectivity between the edge and the cloud. Third, an edge system needs a flexible event processing engine at the edge making it possible to generate insight from machine data when connectivity is constrained. Fourth, an IoT edge-enabled platform should integrate with systems including ERP, MES, inventory management and supply chain management to help ensure business continuity and access to real-time machine data.\n3. Make sure the platform provides cloud-based orchestration to support device lifecycle management\nTo make sure that the edge platform offers highly secure device management, enterprises should select IoT platforms that offer cloud-based orchestration for provisioning, monitoring and updating of connected assets. Leading IoT platforms provide factory provisioning capabilities for IoT devices. These API-based interactions allow a device to be preloaded with certificates, keys, edge applications and an initial configuration before it is shipped to the customer. In addition, platforms should monitor the device using a stream of machine and operational data that can be selectively synced with cloud instances. Finally, an IoT platform should push updates over-the-air to edge applications, the platform itself, gateway OSs, device drivers and devices connected to a gateway.\n4. The platform needs a hardware-agnostic scalable architecture\nSince there are tens of thousands of device types in the world, enterprises should select IoT platforms that are capable of running on a wide range of gateways and specialized devices. And these platforms should employ the same software stack at the edge and in the cloud allowing a seamless allocation of resources. Platforms should support IoT hardware powered by chips that use ARM-, x86-, and MIPS-based architectures. Using containerization technologies and native cross-compilation, the platforms offer a hardware-agnostic approach that makes it possible to deploy the same set of functionalities across a varied set of IoT hardware without modifications.\n5. Comprehensive analytics and visualization tools make a big difference\nAs we\u2019ve already discussed enterprises should choose IoT platforms that offer out-of-the-box capabilities to aggregate data, run common statistical analyses and visualize data. These platforms should make it easy to integrate leading analytics toolsets and use them to supplement or replace built-in functionality. Different IoT platform users will require different analyses and visualization capabilities. For example, a plant manager and a machine worker will want to access interactive dashboards that deliver useful information and relevant controls for each of their respective roles. Having flexibility in analytics and visualization capabilities will be essential for enterprises as they develop IoT solutions for their multiple business units and operations teams.\nEnterprises worldwide are using IoT to increase security, improve productivity, provide higher levels of service and reduce maintenance costs. As they seek to adopt IoT solutions to improve their critical business processes, they should conduct hands-on usability tests to understand edge platform capabilities. Keep watching as more and more enterprises start walking on the edge.