One of the biggest decisions facing many companies undertaking the development of connected products or services is whether to build IoT platform or to adopt an existing one.
There is a broad array of capabilities and expertise required to build a truly scalable and secure IoT platform, so companies must ask themselves if developing it all in-house is the best path. The cost and time associated with building and maintaining data infrastructure in-house, which isn’t adding any differentiation to products or services, are substantial.
Before deciding whether they would tackle building their own IoT platform, companies should seriously think about these critical questions:
What building an IoT platform entails
In the decision to build an IoT platform, one should be aware that the most mature IoT platforms on the market have accumulated over ten years of development. Even mirroring the basic functionalities of the existing IoT platforms may take years, since building an IoT platform is a multidisciplinary complex endeavor comprising of lots of moving parts. If companies aren’t experts already, there is a steep learning curve that will likely result in failures that can delay their time-to-market. IoT platforms, like WolkAbout IoT Platform, were built from the ground up to incorporate the necessary features and functionalities required for any successful IoT solution. But during the development process, there were many lessons learned from past mistakes. Benefit from mistakes paid for by IoT platform vendors is all the more reason not to follow the DIY track. Unless companies have already done a few large-scale IoT implementations, they may not figure out all the possible pitfalls in time to avoid them, they will maybe run into problems as they scale their solutions, or as they try to make them interoperate with their future products and those from other manufacturers, or they will face issues as IoT technologies continue to morph and evolve.
Are we capable of building proper availability, security, and scalability?
If the backend isn’t constantly working, smart connected products or services at best are no better than the products/services companies are trying to enhance, and totally nonfunctional at worst. Therefore, they must ask themselves if they are ready (both technically and organizationally) for 24/7/365 monitoring and consistent uptime. If not, the consequences can be steep.
There are multiple vulnerability points in an IoT system that most companies with the lack of internal expertise, do not know how to manage. Companies must make sure that their team is fully capable of using the latest strategies and technologies to protect data from attack or theft.
The speed and volume of device traffic will more than likely grow as a company grows, and the company needs to make sure that its platform has the ability to handle the increased workload. It’s one thing to build a demo and run a limited trial, and quite another to manage billions of transactions in real-time and at the global scale. If a company uses platform that isn’t built for scale, it might encounter one of many issues, including having to rebuild its system in a mid-life cycle.
Do we have the in-house expertise in all the areas of technology that are needed for building an IoT platform?
Since IoT projects are complex (even with an outsourced IoT platform), it takes both money and time for the employees to understand advanced cloud computing, protocols such as MQTT and building agents onto hardware devices. Furthermore, IoT platform, like any IT investment, must be maintained throughout the entire product lifecycle. So, the team must research, architect, design, develop, and maintain the platform. Hiring a developer team that is able to not only build, but maintain company’s platform long into the future is a serious challenge. Having this in mind, it is clear that the total cost of ownership (TCO) can be lower in a buy scenario as you rely on 3rd party developers to handle new features, support, and maintenance.
Finally, does it make more business sense to build or buy an IoT platform?
The research conducted by IoT Analytics (2016) stated that by outsourcing the already developed IoT platform, companies can benefit from a ~50% shorter implementation cycle and the assurance of working with IoT experts. Solutions built in-house usually take at least 6-12 months to get off the ground, since there is likely to be a lot of research and potential setbacks to get through before reaching stability. That means that time spent developing an IoT platform leads to delayed time to revenue and a full year of advantage for competitors. By eliminating the need for custom development, companies can focus on what they do best, building great products or services and adding value where it matters. Only after they’ve done that, can they begin to focus on IoT use cases that actually benefit the business and presumably, generate a return on investment (ROI). Building an IoT platform is not what generates an ROI, but rather solving business challenges which put the company ahead of its competition.
Considering time-to-market, developer resources, scalability, support, and security, it might be better and less risky if a company focuses on integrating with existing IoT platform and on developing the rest of the IoT project (hardware integration, external interfaces, and additional modules development) instead of starting from scratch. When a company decides not to reinvent the wheel, but to rapidly deploy the existing IoT platform, then the question remains - how to choose the right one among many that are on the market?
References: IoT Analytics (2016, January 26). 5 Things To Know About The IoT Platform Ecosystem. Retrieved from www.iot-analytics.com