The hype around the Internet of Things is now rapidly giving way to the reality of implemented products and services.

Analyst firm IDC predicts that the worldwide IoT market spend will grow from approximately USD 690 billion in 2015 to USD 1.46 trillion in 2020 with a compound annual growth rate of 16.1 percent. The installed base of IoT endpoints will grow from 12.1 billion in 2015, exceeding 30 billion in 2020.

Connectivity has moved from being an interesting feature to being a so-called “price of entry” requirement to achieve competitive product value and differentiation in many of today’s markets.

IoT products and services can range from the basic to the critical: cost-critical, availability-critical, brand-critical, even safety-critical. Therefore, the makers of products and services must understand and respond appropriately to the challenges of engineering for the IoT.

As connectivity increases the capabilities of IoT products and services, so it also increases their complexity. New capabilities bring new failure modes. Added complexity—unless managed appropriately—can increase the likelihood of failures occurring. Furthermore, the consequences of failure can themselves be hard to predict.²

Therefore, increasingly critical products and services require robust IoT engineering. The primary challenges include:

• Delivering compelling functionality (where the requirements might be continuously changing)
• Delivering appropriate dependability, in the form of safety (freedom from harm), reliability (availability of services) and security (freedom from intrusion, interference or theft)
• Delivering the solution in an open context—where some of the technologies and components that contribute to the solution are not under direct commercial or engineering control
• Delivering the solution with appropriate speed and at appropriate cost to respond to competitive threats and changing market demands

IoT-related products and applications will require a more systems-oriented approach to engineering. Systems thinking, especially the concept of emergent behavior (both wanted and unwanted) is crucial for high-quality IoT development and design. Systems engineering, especially system-of-systems engineering, can help reinforce the agility and quality of IoT development and design, especially if the product being designed needs to respond to other products and systems that are not under the designers’ control.

However, systems engineering approaches must be right-sized to apply to IoT, between the two extremes of, on the one hand, extremely agile ad hoc development projects and, on the other, meticulous and expensive aerospace-grade systems engineering. Special attention must be given to safety and security aspects of IoT systems, more so than for conventional apps and software products. Tools supporting such engineering approaches must be flexible and integrated so they can provide the right amount of control and rigor, but also meet the needs of fast development cycles and time-to-market pressures.

To make the most beneficial impact on IoT development, systems engineering approaches should be part of a comprehensive continuous engineering methodology. Continuous engineering makes use of the feedback available from connected products and systems to continuously inform product refinement and new design. It consists of proven principles and practices combining systems thinking and systems engineering, embedded software development and IoT application software development, together with appropriate automation to enact those practices efficiently in a real product development environment.

You need this new design thinking in your IoT projects. IBM is pioneering this Continuous engineering construct under WatsonIoT.

This happens all the time. A new technology is introduced and you think more value will be created because of the innovation. But the reality is that most times, despite the efficiency, technology destroys monetary value for the industry even though the consumers benefit significantly.

Consider WhatsApp, it is destroying monetary value for telecoms in Africa even though WhatsApp itself, directly we may say, is not creating further value by making money (being profitable). Sure, the valuation of Facebook which owns the product is indirectly associated with the number of users but the hard fact is that WhatsApp is not translating the value destroyed to itself in simple revenue numbers.

In the past, Skype reduced the revenue base of most telecom operators in Europe on international calls. But Skype as a company did not absorb that lost revenue into its balance sheet. What happened was the value was destroyed even though customers enjoyed largely free product.

McKinsey Study

McKinsey, on the same construct, has the  view that the digital dawn in insurance actually destroys value—transferring power from the carrier to the customer and eroding profits.  In the United States, McKinsey estimates auto insurance premiums could decline by as much as 25 percent by 2035 due to the proliferation of safety systems and semi- and fully-autonomous vehicles.

For a long time, the traditional insurance business model has proved to be remarkably resilient. But it too is beginning to feel the digital effect. It is changing how products and services are delivered, and increasingly it will change the nature of those products and services and even the business model itself.

Data and analytics are changing the basis of competition. Leading companies use both not only to improve their core operations but to launch entirely new business models. Insurers have valuable historical data. Yet in a few years’ time, will they be able to keep pace and still add underwriting value when competing with newcomers that have access to more insightful, often real-time new data culled from the Internet of Things (IoT), social media, credit card histories, and other digital records. Knowledge about how fast someone drives, how hard they brake, or even (more controversially) what they get up to as displayed on social media is arguably more revealing data on which to assess risk than simply age, zip code, and past accident record. (Facebook recently moved to prevent its users’ online activity being used by insurers in the United Kingdom—proof of the potential power of access to good data.)

And what if those with the necessary data and analytical skills and platforms that reach millions—a Google or an Amazon—not only offered well-targeted, tailored products, but also began to cherry-pick low-risk customers? If they did so in significant numbers, the insurers’ business model, whereby premiums collected from low-risk policyholders contribute to the claims of high-risk ones, could fall apart.

Auto manufacturers are arguably close to changing the game for insurers. The fitting of connected devices as standard in cars is not far off, potentially giving manufacturers unique access to data that could accurately ascertain the risk of their customers, as well as ready-made access to drivers in need of an insurance product.

The Future

In the near future, it is possible that insurance may not need to be offered by insurance companies. Google has a better chance of selling home appliance insurance with all the sensors coming from Nest. The same argument that car companies can simply use the collected data from their OBD sensors to sell insurance to car owners.

What happens to insurance companies? They will be cut-off of the loop.

Insurance companies may need to aggressively redesign their business models to ensure they can compete in this age because while we need insurance, we do not necessarily need insurance companies.