A AND B OF CHOOSING THE RIGHT IOT VENDOR
How to choose the right IoT vendor? The company Indeema discloses in the article.
Assume a pretty common situation when a potential customer is trying to figure out the way of choosing an appropriate supplier of the IoT development service. Let’s call him Adam. He is looking for an affordable customized IoT solution for his middle-sized manufacturing company. Having quite a good overall picture of the IoT, he is discussing the issue with Bob who is an IoT adviser being a proponent of the digital giants. This is how their discussion could look like.
Mr. Adam
As I can see, the entire realm of the Internet of Things software companies can be divided into two big cohorts in general: the IoT superstars where top-100 famous brands shine brightly including such business behemoths as IBM, Siemens, Microsoft, and GE, and a dark sea of all the rest. The division creates a trade-off between what you need and what is offered by numerous ranking resources. When you are a big brand looking for a big deal, the choice can be quite simple: a “Top 10 (20, 50, 100) leaders in the IoT technology” inquiry will initiate an abundant feedback from Google. But in case you are looking for a customized feasible solution for a small or middle-sized business, the “Top IoT companies 2017” rating can hardly facilitate your search.
Mr. Bob
What prevents you from choosing one of the top IoT companies? Their IoT solutions are as good as the rest their world-famous software by default. You use Android or Windows OS without looking for any customized tailor-made alternatives, don’t you? Besides, the ranking resources you mentioned represent the best IoT companies categorized in accordance with their areas of proficiency where each of them is strong: hardware, software, clouds, integrators — whatever you need. In addition, you should take into account their unlimited financial capabilities. It means they can afford the best specialists, the most advanced technologies, and the widest in-field testing.
Mr. Adam
Who disputes that the opportunities the giants have help them to create the best products. However, what is the reason to appeal to Daimler Benz when you need just a shopping cart? I strongly doubt that Microsoft or IBM can take up a customized standalone solution for a small business. Even though they can, it would be too expensive by definition. The best they can do in such a case is to introduce a smaller sub-contractor. The hardware, clouds, and AI-based analytics constitute the area where the giants are really strong. I may guess that they can accept even a simple solution, but only in case it implies a large serial production — something like a “smart” control unit for dishwashers, for example. At the same time, a small narrow-specialized IoT software outsourcer can easily manage with a simple tailor-made IoT solution to be developed and integrated into some existing machinery. Yet far from all IoT projects require processing Big Data by means of artificial intelligence.
Mr. Bob
The latter is disputable, as for me. The IoT always implies Big Data be it at industrial sectors or in a household application. And Big Data makes sense only in case of the data processing and analytics capable of delivering some useful insights to those who use the IoT. The artificial intelligence along with machine learning technologies outperforms humans in analyzing the huge amount of raw data coming from the “smart” things. Just collecting signals from sensors is not enough. It means that any complex IoT solution should include a layer of the sophisticated analytics.
Mr. Adam
You are right in general, but how to create this layer depends on every particular use case. Not every IoT system requires artificial intelligence for analytics. Even the industrial preventive maintenance solutions can do with the predetermined algorithms when the trackable parameters fit into the frame of a particular data diversity. An electric motor shaft, for example, cannot rotate faster than it was designed and slower than zero rotations per minute. So, what is the reason to apply AI to analyze it?
Mr. Bob
Sounds reasonable. Nevertheless, the advanced engineering practice suggests that by means of using AI features for the development process, many IoT vendors significantly reduce the period needed to create the new sophisticated IoT products launching them in six months instead of the two-year cycle usually required for numerous iterations in order to combine hardware and software in the complicated infrastructures. It means that at the competition race the advanced IoT creators can move three times faster than the rest software developers.
Mr. Adam
When the digital giants such as IBM or GE offer some ready-to-use AI-powered solutions for the IoT analytics, it worth using in many cases. The IoT paradigm should evolve from reaction to prediction after all. The “smart” device can be considered really smart when it is able to warn about some failure before it happens. Only in such a case, any IoT solution is feasible, especially in the industrial segment.
Mr. Bob
Exactly. Just consider how much money can be saved on the in-field maintenance of the industrial equipment when the machinery shifts from being mechanically or manually driven to software driven. When a service team knows exactly what, where, and when should be repaired, the saved fuel for the maintenance crew’s trips alone can bring a large extra profit. The task comes to providing such a remotely controlled workflow.
Mr. Adam
This issue relates to the IoT development companies, not to the customers. However, both are trying to follow the 4th Industrial Revolution where the AI and IoT technologies dominate. Many software development companies are going to perform the transition to the IoT paradigm having no background in it. They start with experiments using the hardware mistakenly recognized by the wide audience as amateurish and geek-related. I mean Raspberry PI and Arduino starter kits. Although the intention is commendable, their try-and-error approach is time-consuming.
Mr. Bob
You bet. By the way, when programmers decide to try their hands at developing IoT having no hardware available, they can use some computer game-like simulators, nevertheless. Microsoft, for example, offers an online browser-based simulator mimicking Raspberry PI microcomputer behavior. Of course, it means the developers should use it together with Azure IT — the cloud IoT solution from Microsoft.
Mr. Adam
Anyhow, it does not bring us closer to an understanding of the criteria we should apply to select an appropriate IoT vendor. Hundreds or maybe thousands of websites are available on the internet where the IoT subject is covered one way or another. If I am a potential customer having just a common understanding of the IoT, what features of the IoT development companies should be closely examined? How to figure out the relevancy of every particular IoT developer? What should I need to focus on?
Mr. Bob
You know I’m a propagator of the Big Brands. But in your case, it is more reasonable to consider someone from the outsourcing IoT developers. The field is huge — you can start with the spatial relevance in order to narrow the search.
Mr. Adam
Why? The software outsourcers are good just because they offer what they can with no regard to the places they locate. No matter an IoT development company has an office in Bangalore, Warsaw, or San Jose. The very geographical irrelevance allows the outsourcers to be outsourcers thanks to the contemporary means of communication. The attempt to recognize the top countries for software development is conventional.
Mr. Bob
You are right and wrong simultaneously. Indeed, both the end product of the Internet of Things software companies and the omnipresence of the internet make the geographical location of the software outsourcers irrelevant. However, you should take the “location” option into consideration while surfing the online resources in order to grasp the idea where just the IoT specialization prevails. Especially when it comes to the Industrial Internet of Things. The common logic suggests that the IIoT outsourcers should be concentrated in the areas where the industrial manufacturing is traditionally strong. The Europe-based software companies seem to take precedence over the rest locations in such an aspect. Besides, there should be quite a high level of the higher education in those areas providing an appropriate engineering background to the IoT specialists. Take clutch.co, for example, — a B2B guide allowing to find a software developer based on its specialization and location at the same time. Surfing through this resource, you can find that the Eastern Europe software outsourcing prevails.
Mr. Adam
So, you mean that the initial stage of my search should be based on two criteria: the IoT as a specialization and a Country where the industry and higher education are well developed. But how about the size of a company, annual turnover, experience and other indicators?
Mr. Bob
Let’s move step by step. I suggest you to select Ukraine as the area where both the industrial capacity and educational level meet our requirements. Just look at the results clutch.co represents: 240 IT outsourcing companies are available in Ukraine. Enough to have a wide choice but too many to check them one by one. Narrowing down the search to the IoT development it is easy to select a company where the focus on the IoT is significant. Then, linking to the company’s website you can check the other criteria you are interested in.
Mr. Adam
Well, websites often look great from the design viewpoint. However, it does not always mean that a company is as good at the IoT development as at the web development. Don’t try to convince me that a content on a website or especially on a landing page reflects more than the IoT-around discourse the whole internet is full of. This is for the pure beginners — Google and Wiki help them.
Mr. Bob
In many cases, the situation takes place as you describe, but don’t jump into conclusions. Pay your attention more to what a potential IoT developer shows as its practical experience than to how pretty it is represented. The real cue is the actual practice. And the practice is impossible without mastering the advanced IoT technologies such as GE Predix, for example.
Mr. Adam
Perhaps, an actually implemented IoT solution is the only reliable distinctive attribute of the true IoT developer. Whatever information is presented on a website and however well an experience is described, the very availability of an in-field tested product is utterly persuasive. Now I start recognizing the importance of the appropriate spatial location of the IoT software outsourcers: once they implement their industrial IoT solutions for a commercial use, there should be some working enterprises around allowing the IoT developers to conduct in-field testing of their products. Ukraine seems very relevant to the IIoT development since its industrial capacity is really world-famous.
Mr. Bob
It’s hard to argue with that. Nevertheless, I suggest you check the websites of the promising IoT developers attentively. Once “the first instinct is usually the right one”, pay your special attention to the use cases where the actually existing IoT solutions should be described. Besides, surf through the IoT developer’s blog to check how often the IoT discourse appears there. For the majority of the contemporary software developers, just the blog is the most vibrant and illustrative section of their websites. Don’t neglect the content: the diversity of the available IoT topics reflects how well the IoT developers know what they are talking about. Since the real deep knowledge always wants to come out, look for some educational materials created by the IoT vendor for students and beginners — the desire of lecturing is so natural.
Mr. Adam
This is exactly what you are doing now, right?
Mr. Bob
)))))
Mr. Adam
In any case, I thank you very much for this consistently intellectual discussion highlighting the challenge of choosing the IoT software outsourcers correctly.
P.S. Of course, we invented those hypothetical Adam and Bob just to show what criteria might be applied to figuring out the method of selection of the IoT developers. We realize that our arguments may seem subjective smelling like a self-promotion. Nevertheless, if you find this topic important, please share your opinion in comments adding your own thoughts and ideas regarding the issue.
