The 4 elements of a successful IoT project

John McDonald, CEO of ClearObject, explains how ideation, edge, software, and cloud platforms must come together to successfully harness the Internet of Things.

The 4 elements of a successful IoT project

ClearObject CEO John McDonald sat down with TechRepublic's Alison DeNisco Rayome during LiveWorx 2018 to discuss the importance of product design in IoT projects. The following is an edited transcript of the interview.

John McDonald: Well, in our experience at ClearObject, we found that there are four elements of a successful IoT project. They are ideation, or good design, edge, as I like to call in our industry, which is sensors, network, security, pathways to get the data from the device, software or coding that goes in the device or around the device, and cloud platforms, which is where all projects in IoT start and end. Those four elements have to be in sync for you to successfully implement an IoT project. If they're not, you get a failure, and we've gotten a lot of practical experience in watching and observing that happen, and in helping companies sort of fix the problem after they've experienced those failures and try to get the project back on the rails.

In our experience, companies likely face the most challenges because they start with the edge. They make the device, and so they think they understand the device, the key word being think, and then they'll go immediately hire a bunch of developers and say, "Build me a data-driven service around this product." The result looks like a bunch of developers designed it, which is to say not necessarily the best from a user-centric point of view, and without a lot of thought to the cloud environment that it's going to run on. So oftentimes there's trouble with security or scalability.

SEE: Internet of Things policy (Tech Pro Research)

Worse case scenario is when the developers will take raw cloud space and build their own platform. They'll put databases and data storage and message handling, and all this and code that all in, rather than leveraging existing cloud platform, which would have been a lot speedier because they like to build a platform underneath it, but then the company gets all this baggage that it has to carry around, and maintaining not only the application, but the cloud platform itself.

So the easy way to avoid that is to start with the design, the ideation. That's where all good IoT projects begin, and let that evolve into the things that you need from the edge, and the software you need to write on top of a cloud platform and what it already gives you. That way you're only putting sensors on things you have to put sensors on. You're being more selective about the data that you actually need to get based on the objectives of the project. And you only have to build and maintain the software that it completes the distance between that project design, and whatever the cloud platform already does. It's the fastest, lowest cost way and most reliable way to get it done.

Also see