The Internet of Things is increasingly becoming a disruptive force for many industries by connecting the physical world to the Internet and consequently changing the fundamentals of business models. Although sensors and connectivity are the foundations of the IoT revolution in terms of hardware and infrastructure, without the ability to analyse and generate actionable intelligence and insights from the data, no real value can be derived from the IoT revolution.
A new joint report by Camrosh and Ideya Ltd., titled IoT Data Analytics Report 2016, to be updated and expanded on an annual basis, helps businesses understand the potential and value of IoT data across various IoT analytics applications, and guides them in the selection process.
The IoT Analytics vendor market is young, dynamic, and growing. It comprises start-ups entering the market with new IoT analytics tools, as well as established IT enterprise vendors and BI companies adding IoT Analytics capability to their offerings. Since 2012, the market for IoT Data Analytics tools started to pick up. This is clearly shown in the 47 sample tools, reviewed in the IoT Data Analytics Report 2016.
Potential of IoT Data Analytics
Business data is often stored in bulk, and if not forgotten, then it is analysed in batch mode “later”, either partially or in full. This is one simplified explanation of how a Business Intelligence (BI) approach to data usually works. Connected devices however, generate a continuous stream of data from multiple sensors, which demands real-time, or near real-time analysis, because the cost of storing such volumes of data for later analysis is huge, and the real value of the data is in immediate insights derived from it, rather than the hindsight that might be generated later.
However, when it comes to working with streaming data there is both value to immediate analysis, as well as in storing main events of streaming data for additional analysis in the future. There is also value in complementing streaming data with other types of data, for example geospatial or social network data to give additional context to the streaming data, and generate new, more sophisticated insights. Therefore, an IoT data analytics platform requires the ability to analyse high volume of live streaming data, store the data, fully or partially, and have the capability to perform more sophisticated analysis through meshing different data sources for predictive and prescriptive analytics further down the line.
Strategic Approach to Selecting and Employing IoT Analytics Products
Jumping on the IoT bandwagon is on the agenda of the majority of businesses for two main reasons, namely moving from selling products to selling services, and generating new revenue streams through new business models. There are also more straightforward advantages, such as increasing operational efficiency and profitability, better customer service, and higher inter-organisational collaboration and communication, i.e. transparency and sharing of data. Nevertheless, a large number of businesses are finding it difficult to make the move into establishing IoT based processes and practices, because the technology is fast moving, and the vendor market still emerging and dynamic. These uncertainties and a confusing picture of the market increase the risk of decision-making.
Although many businesses currently active in the IoT space are primarily focusing on the data collection aspect, managing and analysing IoT data is of utmost importance to generate value from that data. Therefore, selecting the right IoT Analytics service that fits the specific requirements and use cases of a business is a crucial strategic decision, because, once adopted, IoT analytics impact not only business processes and operations, but also the whole supply chain and the people involved, by changing the way information is used, and the overall impact it has on the organisation.
According to the IoT Data Analytics Report, when selecting IoT Analytics products, businesses need to consider the following product features and factors:
- Key IoT Data Analytics product features including: Data Sources Coverage, Data Preparation, Data Storage & Processing, Data Analysis, Data Presentation, Administration Management, Engagement/Action Management, Security and Reliability, Integration, Development Tools and Customisations, Customer Support,
- Additional factors influencing purchasing decisions including: Scalability, Flexibility, IoT Analytics Applications and Use Cases, Vendor’s Experience in Business, Vendor’s Industry Focus, Pricing and Key IoT Clients.
Depending on the maturity and complexity of the IoT Data Analytics tool and its key features, the tool can either provide full services, such as an integrated IoT platform that covers all needs from device connectivity to advanced analytics tasks combining various data sources, or offer a partial solution, which has different advantages such as speed, lower cost, lower latency, etc. Often platforms provide a full IoT stack covering everything from device connectivity to advanced data analytics and data visualisation/dashboarding.
Partnerships play a key role in the ecosystem and enable vendors to address specific technology requirements and other aspects of providing services through partnering with enablers in the ecosystem. With the emergence of new use cases and their increasing sophistication, industry domain knowledge will increase in importance. Other factors, such as compatibility with legacy systems, capacity for responsive storage and computation power, as well as multiple analytics techniques and advanced analytics functions are increasingly becoming the norm. Having a good map to find one’s way through the dynamic and fast moving IoT Analytics vendors ecosystem is a good starting point to make better decisions when it comes to joining the IoT revolution and reaping its benefits.
