Industrial IoT software platforms do more than just connect things
I have written before about the ways industrial and manufacturing firms are reinventing themselves to become more efficient and better able to win, serve, and retain customers.
Over the past few months, I have been working with my US-based colleague, Michele Pelino, to analyze the industrial IoT software platforms that are so central to this drive. The result, “The Forrester Wave™: Industrial IoT Software Platforms, Q3 2018,” was published today, and we’ll also be talking about our findings in a webinar on October 1.
It’s a complex space, where traditional industrial conglomerates, established enterprise software giants, networking and telecom providers, hyperscale cloud providers, global systems integrators, professional services firms, and startups all munge (technical term!) together in interesting ways to offer their own perspectives on what “industrial,” “IoT,” and “platform” might mean for themselves and their customers. Few, if any, truly offer a comprehensive solution, and partnership typically lies at the heart of anything useful.
For the purposes of this research, we whittled a large and intriguing field of prospects down to 15: Amazon Web Services (AWS), Atos, Bosch, C3 IoT, Cisco, GE Digital, Hitachi, IBM, Microsoft, Oracle, PTC, SAP, Schneider Electric, Siemens, and Software AG. Clear — demonstrable — support for the industrial domains and their protocols was important, as was a strong international presence: Great companies that only really sell today in the United States (or Europe, or China) didn’t make the grade for this international analysis.
We talk about this a bit more in the report and will be digging into it much more deeply in upcoming research, but a few things stood out to me as we worked through this process:
- The public cloud is the place to be. Practical considerations around providing connectivity to remote locations, plus a general suspicion about the security, capability, and trustworthiness of startup-obsessed public cloud providers, led the early entrants in the industrial IoT space to invest in building their own networks of data centers. Those days are behind us. All of the evaluated vendors retain some ability to deploy in private data centers, but the direction of travel is clear: They, and their customers, are headed to the cloud. It’s a good time to be an analyst who covers both cloud and IoT!
- Analytics are a core component of the platform solution. In 2016, Michele described IoT analytics as an emerging category of functionality. Today, analytics, machine learning, and even some nascent use of artificial intelligence are more common and allow industrial firms to move on from simply monitoring the state of their connected machinery. Vendors must bake analytics, insight, and action deeply into their platform offerings to support predictive maintenance, machine-learning-powered workload optimization and scheduling, and more. I wrote about this last year, but there’s a lot more to come from Forrester.
- The digital twin has arrived, and augmented reality is on the way. Done right, the digital twin will lie at the heart of digitized industrial processes. On the less mature end, this might only be a graphical representation of the real world. But more mature solutions offer a data-driven bridge between the physical and the digital, reflecting real-world operating conditions and simulating possible future states. Augmented (and virtual) reality capabilities are further behind, with plenty of excitement but few tangible examples of delivering sustained value on the factory floor. My colleague Nate Fleming is writing about digital twins, and I have a doc coming on augmented reality in the industrial space later this year.
- Platform vendors are shifting from building blocks to finished results. Early IoT software platforms were collections of technical capabilities with the potential for assembly into custom applications. I often describe them as a box of Lego bricks. In 2018, industrial IoT software platforms are rushing to reposition their offerings, delivering broadly applicable solutions that address outcomes. Predictive maintenance is the most common use case, but there are plenty of others. The box of Lego bricks is no longer enough, and even the patterns, accelerators, and starter kits that some vendors offer may not cut it: Their customers want (need?) applications, powered by the underlying platform.