Building an Effective IoT Ecosystem for Your Business

Transformative technologies have historically revolutionized the way industry works, not because a single new technology is created, but rather because consensus around a group of existing and emerging technologies emerges that provides a cohesive way to think about business innovation. We are now at the cusp of such a transformation with technologies around the Internet of Things (IoT) [1–3].

A large number of devices now come with embedded memory, processing power and communicating capabilities helping them become smart. Connected things allow data to be exchanged between the things and its operating ecosystem including manufacturer, operators, users, and other products [1, 2]. The data collected from these products can be then analyzed to make a better decision-making, improve operational efficiencies and enhance performance of the product.

Due to the rapid advent of Internet of Things (IoT), there exists an ever-growing demand for ubiquitous connectivity to integrate multiple data communication protocols, such as Zigbee, Z-Wave, wireless LAN, cable network, etc. The IoT will include not just for modern IoT-enabled devices but also systems that are already in place to-day and operate outside of the cloud-based IoT solutions. For example, RS-485 [1] based Modbus or BACnet are still popular protocols in building automation industry, and those automation devices do not offer full Ethernet or Wi-Fi interfaces.

The messaging and communication technologies discussed in this chapter can be used to connect devices and people (e.g., sensors, mobile devices, single board computers, microcontrollers, desktop computers, local servers, and cloud services) in a distributed network (LAN or WAN) via a range of wired and wireless communication technologies including: Ethernet, Wi-Fi, RFID, NFC, Zigbee, Bluetooth, GSM, GPRS, GPS, 3G, 4G (Fuqaha 2015; Dujovne et al. 2014) [1].

As the number of internet-connected devices increases, the complexity of management also increases. Hence, we need an appropriate device management solution. Like other products in the market, a connected device has a lifecycle from design, manufacturing, installation, operation or in-use, repair and replacement. In each phase of product life-cycle, there are different requirements from a management solution that shall provide the following to perform a life cycle management of IoT devices.

As a consequence of the proliferation of connected products and people, we must change our data acquisition and processing strategies to accommodate new data-driven business models and connected product development. A connected product enables us to gather and analyze various data including product usages, updates, operating status, maintenance history and interactions with users while it is in-service. As an OEM, you must able to use collected information to improve and enhance product performances and features. Also, you can develop new offerings based on new insights from data analytics.

The data service platform is the backbone of any IoT ecosystem. This is where all the data converges, and we start making sense out of it. Building a data service platform is not about just standing up a BigData processing instance (e.g., Hadoop ecosystem) or migrating all your data into a cloud environment; it is more about building the comprehensive platform for all your data needs and actions. This also includes security, access control, analytics, data lifecycle management and contextualization. Data service platform sometimes contains appropriate traditional technologies as well. In the last three chapters, we have discussed how device information is registered and delivered to cloud service. In this chapter, we will describe tools and technologies for storage, indexing, processing, and retrieval of connected data including telemetry data, metadata about connected devices and others that may generate during a product lifecycle. We will discuss the following topics in this chapter:

In the previous chapters, we have introduced various ways to collect and store data from connected devices. Data Analytics is the final frontier for generating the insights from gathered data. The goals of analytics are discovering useful information, suggesting conclusions, and supporting decision-making through a series processes including inspecting, cleaning, transforming, and modeling data. There are many different analytical techniques which can help generate insights. From the traditional business intelligence to extensions like massively parallel processing, to more sophisticated approaches like time series analysis, machine learning and advanced data mining, you can use all of them today because the power of IoT and cloud computing allows you to do so. These are very contextual to the type of data you have, and what you are trying to achieve with the data. There is no silver bullet with a prescriptive list of bests in class analytical techniques by industry domain or use case. Your best approach is to have a multi-model approach to building your analytics allowing you greater flexibility and capabilities. Building a winning data science team is also important to derive success, too. To make them more impactful on a longer term, some of your analytical capabilities have to be self-learning and adaptive. In this chapter, we shall introduce you to the world of possibilities around analytics and how to best leverage them for your desired business outcome.

When we look across the two spectrums of IoT—consumer and industrial, we see different things evolving. On the consumer front we see more focus on distribution of new capabilities to mass markets; on the industrial IoT front, we see more focus on deeper analytics. In both cases, if we do not have enough focus on closing the loop between sensing, data, analytics, and customer outcomes, the promise of IoT will be underachieved. There are many examples of businesses that have failed to take off despite having cool devices, technology, and analytics because they failed to close the loop with the customer. For large existing businesses making a foray into IoT, closing the loop with internal business operations is a huge challenge.

IoT has dramatically enlarged the playing field with devices and data. While this brings many benefits, it also creates significant exposure to security and privacy vulnerabilities. IoT amplifies the access points for data and control, which in turn amplifies the intrusion points [1–3]. While we continue to build defenses in devices and networks, we also have to deal with a huge population of legacy devices, applications, and networks where inbuilt protection was limited [4, 5]. The threats are becoming more persistent and the impact more profound, sometimes debilitating to businesses. The threats are equally high for consumer IoT businesses as well as industrial IoT businesses. History is riddled with many examples of security breaches with significant impact. The discovery of Stuxnet [6] in 2010, a small 500 kb worm that infected the software for 14 Iranian nuclear power plants brought a lot of focus to this subject. However, there are examples from before. The Slammer worm disabled the Davis-Besse nuclear power plant in 2003. We continue to hear stories about credit card and another personal information breach all the time. A 2014 SANS survey reported 7% more respondents indicate a breach of their environments [1].

IoT is a very rapidly evolving space. This pace and nature of evolution is impacting businesses and companies in very fundamental ways. Large industrial manufacturers are becoming software companies, retail consumer companies are becoming analytics enterprises, cloud software solution companies are developing hardware capabilities that capture more value in IoT ecosystems, start-ups are challenging established players in almost every segment; the transformation across industries is quite unprecedented [1]. Skills and capabilities required to win in this new world order are vastly different to what companies have cultivated over past several decades. Moreover, IoT is impacting transformations much quicker than businesses are normally used to; which necessitates acquiring skills and capabilities swiftly. In such a dynamic environment, it is imperative for companies to partner with others to get faster access to solutions, quicker reach to markets, better scale and more comprehensive capabilities.

A holistic way to think about IoT is things, people, interactions, and outcomes. Often we get mired in the technology, but success requires us to look beyond the technology and applications [1]. Superlative success in building IoT-based businesses requires a strong focus on experiences and outcomes. IoT is significantly increasing the interaction users are having with devices and applications, often for much longer durations than ever before; this is driving user experience (UX) to gain new grounds. There are a few elements which are impacting UX more than ever—miniaturization of devices and displays, a deluge of data, shortening attention spans, interoperability between engagement platforms, increasing expectations of personalization, and the need for quick high impact. These factors create an interdependent tension in designing products as well as user experience. Everybody talks about how simplicity should drive design, but there is little guidance on how to bundle elegance, simplicity, and sophistication into designing user experiences, especially for IoT devices and applications. This chapter helps you with that.

Businesses and customers view the benefits of IoT differently. Everybody has a bit unique perspective on this. It is critical that you market your IoT efforts effectively to your internal stakeholders and your customers. This is important because you need to educate stakeholders about the inherent change management issues in how you transact business in the IoT context. You also want to demonstrate your thought leadership and technical stewardship in bringing new value to everybody internally and externally. Like any other marketing effort, you need to exploit multiple channels of marketing and also embed marketing efforts in your products and services. In this chapter, we will introduce you a framework of driving thought leadership through a product-patent-publicity framework and expose you to some of the best practices in the field.

Every technology hype-cycle [1] brings along with it the promise of substantial economic benefits and opportunities. Similarly, IoT is expected to create new opportunities worth billions of dollars and lead to benefits worth trillions. It does not matter which numbers we believe, one thing is for sure—the impact of IoT is big. When such projections are made, the premise is always based on assumptions of how IoT (or for that matter, any other technology) is changing how we live, work and engage with our broader environment. Projections are realized by appropriate monetization effort for each individual initiative. So it matters how you monetize your IoT efforts, as the industry does not change or move without your contribution. Moreover, any IoT initiative, like any other technology initiated or technology enabled action, requires investments. The relatively depressed economic climate over last several years has strained investment bandwidths. The window of expectations from any investment is shrinking, and appetite for large-scale investments is reducing. These factors demand more focus on monetization efforts for your IoT initiatives.

Like any new business venture, IoT domain also requires you to bring together an awesomely talented high performing team to be successful and win decisively. As we have discussed several times in this book, there are disruptive and transformative elements of IoT. So the criticality of having the right team is as critical as having the right offerings, platform, and business model. In this chapter, we shall discuss the human aspects of building a winning IoT business. While each IoT initiative is different, there are some common elements and best practices that you can benefit from. You will find some similar concepts and descriptions in Chap. 5 of our book “Making Big Data Work for Your Business”.