Avoid the Proof of Concept “Black Hole”

Many companies think they need to design a proof of concept for their new products for a variety of reasons. When it comes to the internet of things (IoT), however, this type of undertaking is often unnecessary.

The only reason you need to prove you can do something is if it’s never been done before. And if it hasn’t been done before, it probably shouldn’t be part of your product development. The technology behind IoT has been around for 15-20 years, making it far from bleeding edge. In this blog, we’ll discuss why proof of concepts are typically not needed for IoT applications, and what your company should do instead.

Why do companies think they need a proof of concept?

A lot of companies start IoT projects by trying to build the device, rather than solving the problem. They latch on to a single idea or concept, ignoring any other possibilities. For example, if they are creating a sensor for garbage receptacles, they may focus exclusively on its ability to track when it gets full. They don’t think about what other data could be gathered that may be useful to other potential clients, such as usage patterns (number of uses until it is full, dumping times, etc.) or temperature, humidity, and precipitation data that could affect those usage patterns.

Another misstep is when companies start by building the IT infrastructure and cloud back-end setup. By the time they get to designing the device, they’ve already invested considerable time and money without truly understanding the device’s requirements.

With both scenarios, these companies moved forward without first completing the necessary work to determine the business value of their efforts. Essentially, they paint themselves into a corner where they feel some sort of proof is needed to justify the sunk cost. They think they need their proof of concept to show that the device is indeed something of value. It’s supposed to be the other way around. First you define the value, then you define requirements to meet the mission needs to create that value.

Why are proof of concepts problematic?

Proof of concept projects are typically not necessary, especially for IoT projects, for several reasons:

• There is nothing to prove – You don’t need to prove that cellular works or that batteries work. Instead, you need to create a scalable prototype.
• They aren’t scalable – You spend time, money, and effort to develop a proof of concept as quickly and cheaply as possible. But when it is done, if you did it right, it won’t scale for production, and you’ll still have to create a real scalable product.
• There are cheaper ways to prove you need something – You don’t manufacture a car to see if you need it. Your device should be no different. Instead of a proof of concept, do a thorough cost/benefit analysis.

 
What should you do instead?

When you have an idea for a device, don’t spend any money yet. Start with the business case and solid written plans. Then complete the system requirements and a system plan, followed by thorough requirements documentation for each solution. Only when your written plans are complete, and you’ve completed a thorough cost-benefit analysis, should you proceed with the design of each component.

By thinking things through, you know for sure what technology is involved for each component of your solution. You will be relying on real science vs “Star Trek science” that needs to be proven. If this is not the case, you know you need to do more research. You can also potentially discover alternative uses that have value to you – or to others that could be sold for additional profit.

If you’ve got an idea, questions or want to better understand how to apply these tools and technologies, contact the experts at F3 to help develop your device and bring it to market. If you’re not sure how to begin drafting device requirements, or even determining if you are ready to begin, we can help get you off to a good start.