Grundlegende Erfolgsfaktoren bei der Implementierung einer IoT-Lösung

Even Internet of Things projects are not free from failure. They are often highly complex, more than others they concern several departments and very specific knowledge, and the topic of IoT is still uncharted territory for many companies. But what are the success factors when implementing IoT projects? In addition to the usual precautions for IT projects? What else can be done to ensure project success? I would like to address four points.

Implementing IoT is a process optimizer

Two weeks ago I wrote that IoT is not a technology, but implementing IoT must be accepted as a way to optimize processes. This thinking is also the first step towards a successful IoT project. Do not think about technology (however fascinating it may be), but about the processes you will optimize by integrating additional data. This perspective is important when it comes to shaping the project, to present it, as well as to receive proper support from senior management and all relevant departments. A clear presentation of a relevant use case and a clear presentation of the ROI are extremely important – not just in companies that operate under to the principle of shareholder value ( Roi-internet-things). Otherwise, your project will run the risk of being stuck in pilot phase, because budget and support are not provided as needed.

Implementing IoT and IT are technically very different

Our ISG colleagues Ron Edler and Sridhar Manickam wrote a very interesting article on “A Practical Approach to Advanced Analytics in the IIoT “. It discusses many aspects of a promising Advanced IoT Analytics project approach, which is widely used implementing IoT in business processes. The difference between IT and IoT projects is one very important factor: the monolithic Big Data approach does not work in an IoT environment. Too many data from different sources, which are also subject to their own rules. To be able to master this extremely powerful sensor data stream, you require a different architecture that includes several layers with different processing and control steps. IoT is therefore clearly a project that follows its own rules. This has to be considered already in the project planning, otherwise the project runs the risk of failing at the technical differences between IoT and IT.

Implementing IoT Analytics Struktur

Three Layer-Approach – Source: ISG Insights 2016

Data quality is cruicial 

Another important aspect is the quality of the data. Within the boundaries of a company, we can ensure high standards for consistency, quality, consistency and error tolerances in a given data flow. In the large sensor data stream of an IoT solution, collecting and aggregating data from a wide range of sources, there is much more “dirty” data to be cleaned and processed first. New standards must be developed, according to which the data quality is evaluated. Much more experimentation has to be done before the IoT solution will work reliably. It is also important to consider where additional data can be obtained from outside and integrated into the process to improve overall data quality. Partnerships with external service and data providers, which provide data and also analytical methods, can therefore be vital for implementing IoT.

Partnerships are essential 

This is the fourth success factor for me: do not try to solve everything on your own. IoT projects are complex and – as already said – very knowledge-intensive in their details. An answer to a question within the scope of an IoT project is only partly similar to the answer to get within a complex IT project. IoT projects often require entirely new architectures and methods to work really effectively. Service providers and specialists in the IoT environment help you not only solve very specific problems effectively, but also show you new perspectives on how to reach your overall goal of process optimization faster and more effectively. This also applies to the question of the right platform.
Do not automatically set your strategy to build your own IoT platform in oder to integrate others. Restrict the first projects and prefer to work within existing third-party platforms This gives you greater flexibility and the ability to change platforms in the course of further development, or to add additional platforms.
With my customers I find it extremely important to point out that Internet of Things projects are only partly comparable to classical IT projects. They follow their own rules. Surely there are other success factors in the IoT project which I did not discuss here. Which are especially important to you? Please comment here.