Pros and Cons of Building Your Own IoT Platform

"Should I build my own IoT platform? Is it worth it to do it myself? The answer is, of course, it depends.

You’ve got an idea for an awesome IoT project, and you’re ready to get started making it into a reality. But there’s a lot of options to choose from. Should you build your project from the ground up? Should you hire an IoT company to build it for you? Or should you take a middle path, using others’ work as a foundation on which to build?

No matter what you decide, you’re going to need a solid platform. But what’s an IoT platform? Simply put, it’s all the IoT technology stack components that we need to begin working on a solution. It includes hardware, firmware, operating systems, IoT data storage, communication channels, cloud resources, edge computing, backend services, and many other technologies. We can only start working on our actual project after we have all these ducks in a row.

When we put all this together, it’s not hard to see that developing an IoT platform from scratch is a lot of work—and we haven’t even started talking about maintenance. That’s why so many companies in the IoT space decide between building or buying their IoT platform.

In this article, we’re going to focus on what goes into building a platform so that you can make an informed decision. There’s plenty of good reasons to build, but it’s crucial to understand the downsides fully as well before jumping into the deep end.

Pros of Building Your Own IoT Platform

The best reason to build an IoT platform is full customization. IoT is still a young field, and it’s growing fast. If you have a truly innovative IoT product idea and the tech you need to make it happen simply doesn’t exist yet, then the building can be a great option.

“If you decide to build an IoT platform on your own, you’ll have complete control over the solution,” writes Riya Savjani for IoTNews. “You can customize it to make it a good fit for your industry and business. Everything from the data collection to the end-user experience is yours to modify, merge, and maintain.”

Another reason to build is leveraging in-house tech expertise. If your team has the right skills and know-how—from electrical engineering to IoT development to cloud administration—then putting great minds to the task of building something truly unique can pay dividends.

Lastly, we believe that technology R&D is inherently worthwhile. We actively contribute to open source projects like Nerves because we understand that furthering our collective knowledge makes the world a better place. After all, if Linus Torvalds had decided in 1992 that UNIX was a good enough platform, we wouldn’t have anywhere close to the nearly 600 Linux distributions that we enjoy today.

Cons of Building Your Own IoT Platform

Building a DIY IoT platform has some downsides as well. The most obvious one is that it costs a lot more. As Daniel Elizalde puts it, “You can't build an IoT platform for less than what it would cost to adopt a commercial one.”

Let’s take a moment to consider what goes into an IoT platform. You need to buy tons of hardware, write lots of code, and payroll all your people. Depending on how much you want to build from scratch, this can easily reach into the millions before you even have a minimum viable product (MVP). Plus, don’t forget about the ongoing maintenance costs for the product’s entire lifespan.

Then there’s the corollary, time to market. Building something from scratch takes a lot longer than buying something ready to go (or at least that’s ready to tweak and adjust), and IoT is no different. Prototyping, developing, debugging, and troubleshooting always takes longer than you think. It’ll probably be a year or two before your product is in manufacturing.

Over that time, your company has to shoulder all the costs without seeing any return. On top of that, you’ll also bear the opportunity cost of devoting time and resources elsewhere.

Finally, building a platform brings a higher risk. This process is riddled with challenges, and the results don’t lie. Gartner reports that 75% of IoT projects take twice as long as planned. Cisco reports only 26% of surveyed companies said their IoT initiatives were a “complete success,” while 60% report failed IoT projects during the early proof of concept stages.

It’s no secret that getting an IoT system to work isn’t easy. Even seasoned IoT developers encounter problems all the time. Building an IoT platform entails risks that span improper security practices to failed GDPR compliance to misconfigured firmware that leads to bricked IoT devices. Realizing one of these incidents can spell instant disaster for a project.

Conclusion: To Build or Not to Build?

Ultimately, deciding to build an IoT platform often comes down to finances. To this end, Rich Mironov warns us against the DIY illusion, “the idea that broadly available commercial products don’t give us exactly what we want, but our internal team can whip up just the right thing.” While the benefits are there, they usually don’t justify the costs.

That said, we’re not here to dissuade anyone. Innovators are crucial in the IoT ecosystem, and this includes engineers who work on the most foundational infrastructure all the way to those who develop the high-level IoT applications with which our users interact. What is clear, however, is that specialization and collaboration are the keys to success, so we want to pick our battles wisely.