Forecast: The Internet of Things, Worldwide, 2013

18 November 2013 ID:G00259115
Analyst(s): Peter Middleton, Peter Kjeldsen, Jim Tully

VIEW SUMMARY

The Internet of Things will include 26 billion units installed by 2020. IoT product and service suppliers will generate incremental revenue exceeding $300 billion, mostly in services, in 2020. It will result in $1.9 trillion in global economic value-add through sales into diverse end markets.

Overview

Key Findings

  • The installed base of "things," excluding PCs, tablets and smartphones, will grow to 26 billion units in 2020, which is almost a 30-fold increase from 0.9 billion units in 2009. The component cost of IoT-enabling consumer things will approach $1, and "ghost" devices with unused connectivity will be common.
  • There will be a $309 billion incremental revenue opportunity in 2020 for IoT suppliers from delivering products and services.
  • The total economic value-add from IoT across industries will reach $1.9 trillion worldwide in 2020.
  • By 2020, more than 80% of the IoT supplier revenue will be derived from services. Services that are associated one-to-one with individual products will face competition from higher-tier services, in which trusted IoT providers oversee multiple product categories with branded "umbrella" services.
  • The industries likely to see the greatest value added from the IoT will be manufacturing, healthcare providers, insurance, and banking and securities.

Table of Contents

Forecast Data

Table 1 shows total economic value-add from Internet of Things (IoT) technology. Conceptually, the total economic value-add across all sectors is a measure of GDP (actual GDP includes the impact of taxes, subsidies and other factors), so the values in this table represent incremental global GDP. Value-add can be derived within an industry sector by either increasing sales, or decreasing input costs, or both. The value-add represents the aggregate benefits that businesses derive through the sale and usage of IoT technology.

Table 1. Total Economic Value-Add From IoT, Worldwide, 2013-2020

2013

2014

2015

2016

2017

2018

2019

2020

Value-Add ($T)

0.3

0.4

0.5

0.6

0.8

1.0

1.4

1.9

Growth (%)

-

25

26

27

29

31

34

39

Source: Gartner (November 2013)

Table 2 shows a breakout of Gartner estimates of the percentage split of value-add across industry sectors.

Table 2. Worldwide Economic Value-Add by Industry Sector in 2020

Industry Sector

Contribution to Total (%)

Manufacturing

15

Healthcare Providers

15

Insurance

11

Banking and Securities

11

Retail and Wholesale

8

Computing Services

8

Government

7

Transportation

6

Utilities

5

Real Estate and Business Services

4

Agriculture

4

Communications

3

Other

4

Note: Numbers may not add to 100% due to rounding.

Source: Gartner (November 2013)

Analysis

Figure 1 compares the growth of IoT "things" with the growth in connected personal computing devices.

Figure 1. Growth of Things Will Be Rapid
Figure 1.Growth of Things Will Be Rapid

Source: Gartner (November 2013)

The growth in IoT will far exceed that of other connected devices. By 2020, the emergence of mass-market smartphones and tablets, combined with the mature PC market, will result in an installed base of about 7.3 billion units, which compares with the expected human population of 7.7 billion in that year (based on information from the United Nations Population Division). In contrast, the IoT will have expanded at a much faster rate, resulting in an installed base of about 26 billion units at that time. Installed base is important because it drives the value of service revenue, aggregate communications bandwidth and data center activity.

Due to the low cost of adding IoT capability to consumer products, we expect that ghost devices with unused connectivity will be common. This will be a combination of products that have the capability built in but require software to "activate" it and products with IoT hardware/software that customers do not actively leverage. Depending on the usage model, much effort will be required to educate consumers about the utility they will derive from using IoT products. Companies may offer different product tiers, with the higher-end products offering IoT-enabled services, and midtier products sold with the inherent hardware capability and needing only a remote software upgrade to unlock the IoT capabilities.

In addition, enterprises will make extensive use of IoT technology, and there will be a wide range of products sold into various markets, such as:

  • Advanced medical devices, including surgical tools, instrumentation and wearable medical sensors/monitors, and ingestible devices, such as smart pills
  • Factory automation sensors and applications in industrial robotics
  • Sensor motes for increased agricultural yield
  • Automotive sensors
  • Infrastructure integrity monitoring systems for diverse areas, such as road and railway transportation, water distribution, and electrical transmission

Figure 2 shows a distribution of shipments of things by category in 2020, ranked by volume.

Figure 2. Internet of Very Different Things
Figure 2.Internet of Very Different Things

Source: Gartner (November 2013)

The graph ranks about 60 different categories, with a handful called out as examples. The highest-volume shipments will come from connected light-emitting diode (LED) interior light bulbs. Other areas with high shipment volumes include connected set-top boxes and TVs, wireless peripherals, smart meters, and smart pills. As we move to the right of the graph, the number of things in these categories falls off into a long tail composed of a great variety of different applications.

By 2020, component costs will have come down to the point that connectivity will become a standard feature, even for processors costing less than $1. This opens up the possibility of connecting just about anything, from the very simple to the very complex, to offer remote control, monitoring and sensing. Many categories of connected things in 2020 don't yet exist. As product designers dream up ways to exploit the inherent connectivity that will be offered in intelligent products, we expect the variety of devices offered to explode. The combination of many new applications, many of which will be specialized and ship in relatively low volumes, will form a long tail of devices, which looks very different from the picture in 2009, when PCs and smartphones dominated. The market dynamics in this emerging environment will be quite different in view of the high fragmentation of submarkets. IoT suppliers will need to carefully consider which markets to operate in, and different techniques will be required to reach customers.

Figure 3 shows different categories of things plotted on demand versus systemic friction axes.

Figure 3. IoT Adoption Will Occur at Different Rates
Figure 3.IoT Adoption Will Occur at Different Rates

KB = keyboard

Source: Gartner (November 2013)

On the graph, demand represents the combination of end-user interest and/or economic incentive to adopt a technology, where end-user interest typically applies to consumer applications, and businesses will consider the ROI when choosing whether to invest.

Systemic friction is effectively resistance to change. It represents the combination of technological, infrastructure prerequisite, business model/market structure, regulatory and legal, and standards-related "barriers" to adoption of the technology in a particular area. Thus, this graph gives a sense of the potential adoption rate of the various IoT product areas. The rate of adoption of a new technology is a function of both user demand and any factors that will resist change. The size of the bubbles on the chart represents the number of units shipped in 2020 and corresponds to the values shown in Figure 1.

Differences in demand and systemic friction will result in adoption of different technologies occurring at different speeds. Some things will move very quickly in the market because all the consumer needs is an app to use them — the infrastructure is already there. Other things will require a substantial change to institutional systems and will take a longer period before adoption is widespread.

One of the highest-demand sectors for IoT is street lighting. Embedded intelligence and connectivity allows LED lamps to be controlled and monitored. The overall benefit is reduced costs through reduced energy consumption. This is happening fast because there is little to slow the adoption process.

But compare this adoption scenario with the medical sector. Smart pills, for example, will take longer than lighting to move into the mainstream because of regulation, such as Food and Drug Administration (FDA) approval. Nevertheless, the new IoT technologies will add $285 billion of healthcare provider value to the global economy by 2020 — but the main benefits will be felt after this date.

For example, printers are shown with moderate demand because they are a low-growth (flat) market. In contrast, toys are potentially a more exciting end-use technology for consumers.

IoT Will Bring New Supplier Opportunities and Add Global Economic Value

Supplier Revenue Opportunity

The IoT encompasses hardware (the "things" themselves), embedded software (software running on, and enabling the connected capabilities of, the things), connectivity/communications services and information services associated with the things (including services based on analysis of usage patterns and sensor data). We refer to the companies that provide this hardware and these services as IoT suppliers. The incremental IoT supplier revenue contribution from IoT in 2020 is estimated at $309 billion. Incremental revenue refers to the difference in sales that results from adding IoT capability to a product/service offering. For example, LED light bulbs are in the early stages of coming to market, though early models are not networked. By adding IoT capability, LED lighting suppliers can charge higher prices for their products, and associated services will arise. It is this increase in opportunity associated with IoT that is referred to as incremental supplier revenue contribution. Another example would be adding IoT capabilities to home appliances. We refer only to this addition, rather than to the full value of the original appliance market. So, in this example, an appliance company could be a supplier of both traditional, unconnected appliances and a supplier of IoT-enabled appliances. Similarly, a company selling jet aircraft engines will see an incremental product and service revenue opportunity associated with the addition of connected sensors to its designs. We count the incremental revenue as opposed to the total sales of jet engine products and services, which is much larger.

Because of the diversity of IoT applications, it should be noted that the suppliers involved will not necessarily be traditional technology and service provider companies (i.e., those that we currently associate with information technology). In the case of the lighting example, the incremental revenue contribution will go to those that provide services and those that build IoT lights. Light bulb manufacturing has not traditionally been thought of as a high-tech industry, but connected LED lighting combines advanced optical semiconductor expertise with a software-driven, networked operational model (so it is very much a high-tech product). The utilities or home management companies that seek to derive service revenue from such products are likewise not traditional IT services providers. There are many other examples, so IoT will open up revenue opportunities in a wide range of areas as it augments traditional product and service categories.

As hardware costs fall, and solutions proliferate, the bulk of this opportunity falls to service providers. Although the industry will benefit tremendously from IoT technology, consumer end users will drive almost two-thirds of the IoT supplier revenue.

Economic Value-Add

Economic value-add for a sector is essentially the sales in that sector minus inputs and supplies from other industries (for example, manufacturing sales minus the cost of raw materials and energy). Conceptually, the total economic value-add across all sectors is a measure of GDP (actual GDP includes the impact of taxes, subsidies and other factors). Value-add can be derived within a sector by either increasing sales, or decreasing input costs, or both. The value-add represents the aggregate benefits that businesses derive through the sale and usage of IoT technology. For example, an IoT technology supplier derives value-add through its profit from sales of products or services. A business that invests in IoT technology (bought from a supplier) derives value-add from the use of the technology, which can be leveraged to increase the business's sales or reduce its costs. Economic value-add for a sector represents the total across all businesses in a sector. Figure 4 shows the growth in total economic value-add globally from the present through 2020.

Products and services purchased purely for consumer entertainment/utility do not generate economic value-add above the value derived by IoT suppliers. For instance, an IoT-enabled toy with an associated service (e.g., to enable multiple children to play simultaneously via a subscription service) generates value-add for the toy manufacturer and for the service provider (which may be the same company). However, consumer purchases of IoT products/services can contribute to value-add in diverse industry sectors. For example, consumer purchases of connected smoke detectors benefit the insurance industry by improving public safety, as do purchases of automobiles with IoT-enabled safety systems. Other examples of consumer influence include consumer use of healthcare or transportation services, whereby the level of consumer usage will impact the magnitude of the value that can be derived from IoT technology.

Figure 4. Total Economic Value-Add, Worldwide, 2013-2020
Figure 4.Total Economic Value-Add, Worldwide, 2013-2020

Source: Gartner (November 2013)

Figure 5 provides a breakout of the distribution in 2020 economic value-add by industry sector.

Figure 5. Economic Value-Add by Industry Sector in 2020
Figure 5.Economic Value-Add by Industry Sector in 2020

Source: Gartner (November 2013)

Economic value-add across sectors in 2020 is forecast to be $1.9 trillion. The $1.9 trillion in economic value-add partially overlaps with the $309 billion incremental IoT supplier revenue figure because suppliers derive value from the profits they earn in selling their IoT products and services. IoT value-add is composed of the combination of mature IoT, which is already yielding benefits, and a high-growth emerging IoT opportunity. It is derived from a combination of sector-specific technology (such as connected, automated manufacturing systems in the manufacturing sector), and more generic, widely used technology, such as the suite of "smart building" technologies, including LED lighting, smart HVAC systems, etc.

Some industries (part of mature IoT) are already highly leveraging connectivity, such as communications and banking, which has widely employed connected automatic teller machines (ATMs) and point-of-sale (POS) technology to lead to operational savings and enhanced revenue opportunities. While the unit numbers associated with early IoT are not large, the technology is already having a noticeable effect. That said, the growth in IoT will have a relatively small short-term impact but a potentially transformative long-term impact (with our 2020 date being possibly on the cusp of the transformative change, but not yet in the midst of it).

Emerging areas will witness rapid growth of connected things. For example, this will lead to improved safety, security and loss prevention in the insurance industry. IoT will also facilitate new business models, such as usage-based insurance calculated based on real-time driving data. The banking and securities industry will continue to innovate around mobile and micro-payment technology using convenient POS terminals and will invest in improved physical security systems. IoT will also support a large range of health and fitness devices and services, combined with medical advances, leading to significant benefit to the healthcare sector. Emerging connected sensor technology will lead to value creation in diverse areas such as utilities, transportation and agriculture. Most industries will also benefit from the generic technologies, in that their facilities will operate more efficiently through the use of smart building technology, such as connected LED lighting systems, HVAC systems and physical security (via connected door and window locks and monitoring systems).

From an IoT supplier standpoint, the primary value-add will arise in computing services, which will enable the mining of big data generated by the installed base of "things" — converting huge amounts of data into a manageable amount of actionable information for end users. The value-add from application software associated with the usage of IoT also falls under computing services. The value-add from hardware sales will arise in the manufacturing sector, while connectivity services will benefit the communications sector.

Assumptions

Market Penetration

By 2020, several IoT categories will have reached the steep part of the adoption s-curve, but the IoT market as a whole will be on only the cusp of the transformative change induced by IoT.

Some of the segments with long-term transformative potential will, due to various types of systemic friction, not yet have taken off by 2020. In our forecast for 2013 through 2020, we have assumed that services associated one-to-one with individual products dominate the picture. The likely emergence of higher tiers of umbrella IoT services across traditional product categories could potentially create an "IoT tipping point" but, in our analysis, we have assumed that this will not dominate the picture during the forecast period.

About 15% of LED lamps will contain connectivity technology by 2020 and will therefore be members of the IoT. This amounts to more than a billion connected lamps.

The use of LEDs results in considerable energy savings, combined with electronic controllability. About 90% of the energy savings will be in consumer markets, with commercial, street and area lighting accounting for the rest. The energy savings are so great that governments will encourage the use of LED lamps through subsidies and forced use. This adds a degree of certainty to the forecast for these devices. By 2020, LEDs will reduce worldwide electrical consumption by 1,400 terawatt-hours annually.

Cost Structure

By 2020, the cost of adding basic IoT capability to a consumer-grade end-user device will approach $1.

Any product that contains a microcontroller or system-on-chip device will have inherent IoT capability, once the incremental cost of basic connectivity and sensors comes down. This will enable the addition of latent IoT capabilities and will create a vast number of ghost devices that are IoT-enabled but not actively connected or monitored. Moreover, due to the low incremental cost, even products that contain no electronics or minimal electronics today could be offered in connected versions in the future (such as hand tools, door and window locks, toys, smoke alarms, thermostats, parking meters, and similar items).

Price levels for the combination of a basic microcontroller, Bluetooth wireless connectivity and a simple sensor, such as a temperature, vibration or acceleration sensor, will approach $1 by 2020. In other words, the cost of the component technology is not a barrier to IoT growth. More complex devices with additional onboard compute capabilities, multiple and/or specialized sensor clusters, autonomous operation, or advanced mobile/remote power capabilities will cost considerably more.

Business Models

By 2017, 50% of IoT solutions (typically a product combined with a service) will originate in startups less than 3 years old.

Through 2017, the IoT market will be in its very early stages. Similar to other technology advances historically, the growth promise associated with these early stages will lead to the creation and funding of a large number of startup organizations. These companies will maneuver to capture what they perceive to be early opportunities or overlooked product niches. This will lead to creative solutions and a wide range of products, many of which will fail in the market. Nevertheless, this process will lead to growth as the successful solutions are often consolidated by larger suppliers, and the overall market expands. These developments will diversify the IoT ecosystem and lead to the long-tail effect shown in Figure 2. It will also lead to unpredictable things and services. We don't know what many of these will be, so we have considered an incremental "unknown" category in our analysis.

By 2020, more than 80% of the IoT supplier revenue will be derived from services.

The incremental cost of hardware and embedded software is relatively small, whereas the service and analytics opportunity is much larger. Initially, much of the supplier focus in the IoT market(s) will be on hardware and software, as companies try different approaches and feature sets in an effort to build awareness of their products. As business models mature, however, the market will increasingly be driven by services (including data analytics services). There will be service opportunities associated with coordinating and managing the multiple things that typical consumers will interact with in an average day, both in their home/possession and in their daily travels. Indeed, the value chain for IoT devices and services will be multilayered, with usage data leveraged using analytics software that is designed to pull out trends useful for further product and service marketing initiatives. IoT creates a big data problem that analytics must solve — transform huge volumes of data into a small (readily synthesized by a human mind) quantity of usable/actionable information.

Services will initially be associated one-to-one with individual products. Through 2020, this will open up to allow higher tiers of umbrella services and analytics, which will associate services with multiple products (one-to-many) to generate significant economic value-add.

Under this umbrella model, a service could arise to manage, control, aggregate data or analyze information from multiple things. As an example of value creation under this model, where patterns of user behavior can be tracked (from the use of multiple things), there will be an opportunity to analyze them to generate useful information or to perform marketing analytics. However, there will be security and privacy implications of sharing data, so society (often individual countries) can be expected to pass legislation regulating the use of shared data. This is one factor that could affect overall systemic friction, as depicted in Figure 3.

Dominant IoT providers of higher-tier umbrella services are likely to emerge. Building end-user trust and brand equity (in part due to concerns over privacy and data security) will be key success parameters for prospective higher-tier IoT service providers. There will be two main approaches to these umbrella services. One version will be based on a common technology platform developed to connect and manage multiple categories and brands of things. The second one is a portal/customer-facing umbrella where consumers (or enterprises) perceive that all their services are managed by a single entity, such as Apple, Samsung, Siemens, or a telecom or utility company.

Forecast Methodology

Definition of a "Thing"

IoT is something of a misnomer; things needn't connect directly to the public Internet, but they must be connectable via a network (which could be a LAN, PAN, body area network, etc.) and individually addressable. Thus, if 40 LED lamps are clustered together into an LED light bulb, which is connected to a LAN, then this counts as one thing. If HVAC sensors are colocated with the light bulb and share connectivity, then this cluster of lamps and sensors counts as one thing. Similarly, if an agricultural sensor mote consists of a group of four types of sensors, along with a power source, microcontroller and radio, this will count as one thing. In some cases, connections will be hierarchical. For example, a car can have multiple addressable subsystems, and the vehicle itself could represent a thing (such as a node in a traffic analysis system). In these cases, we considered only the "leaf nodes" in our analysis. It should be further noted that we exclude PCs, tablets and smartphones from the count of IoT things. This is because personal computing devices such as PCs, tablets and smartphones will often be used as a hub to access information from, configure/control and manage things. In many cases, there will be a many-to-one relationship between things and personal computing devices (but not all personal computing devices will interact with things, particularly in the early stages of the market).

Value-Add Forecast/Hindcast Approach

The forecast presented here assesses the total value-add across all of the vertical sectors covered by Gartner.

Traditionally, when forecasting a given market, the starting point is a sizing of the current market, which forms the basis for the forecast of how the market will develop in the future. However, the IoT market is still at a very early stage, where many aspects of the market are embryonic or even nonexistent.

Given the characteristics of the IoT market, we decided to start with sizing the market in 2020 — which is when we believe that the market will be sufficiently mature to allow for estimates on how many devices of a given type there may be per person, per household, etc. Since gauging a market size seven years out is not a trivial task, we decided on an approach by which we triangulated the outcome of three different types of analyses based on different methodologies, rather than relying on the methodology of a single analysis. This is described in more detail in a later section.

The preceding years in the forecast are therefore essentially a "hindcast" based on the 2020 estimate — but of course also with an eye on the parts of the IoT market that are already visible today.

2020 Market Size Based on Three-Way Comparison

As mentioned, the methodology to develop the 2020 IoT market size forecast consisted of three approaches, which yielded results that were then cross-referenced against one another. An iterative approach was used where assumptions were adjusted until a converged and consistent picture emerged from the three methodologies. These approaches are referred to as (1) long-tail product category analysis; (2) n-per-population study; and (3) economic envelope.

Long-Tail Product Category Analysis

  • The purpose of this analysis was to generate a diverse list of conceivable applications.
  • Some of the product categories were accessible from existing Gartner forecasts, such as for LEDs, set-top boxes, printers, etc.
  • Gartner secondary research was used to help compile a comprehensive list, and we added an "others" category to capture items that we have not thought of or do not yet exist.
  • Installed base and shipment figures were developed for 2020.
  • Product categories were classified into consumer, enterprise, consumer/enterprise, industrial and high-grade industrial.
  • Relative price points were assessed for IoT capability depending on category, with consumer the least expensive and high-grade industrial as the most expensive.

Number-per-Population Study

  • The purpose of this analysis was to judge the size of the market in relation to populations of people/objects/market sizes that influence the usage and purchase of connected things.
  • For example, we can relate the number of connected appliances to the number of households, or the number of automotive sensors to the number of cars in service, or the number of traffic signals to the length of a major roadway.
  • The saturation level per unit of population and the percentage of penetration were estimated for 2020. Saturation represents the number that would be used per population metric once the technology is fully mature at some point beyond 2020.
  • We assumed that quantitative product sales could be associated with these populations. For example, one could assume that 20 light bulbs are used in a typical home and measure the penetration of connected LED lighting against that saturation figure.
  • The various categories of things were grouped according to usage by economic sector and consumers.
  • Estimates were developed for IoT supplier sales, IoT supplier value-add and sector-as-customer economic value-add:
    • IoT supplier revenue was derived from the total shipments of things in 2020 and the cost categories.
    • IoT supplier revenue represents the 2020 incremental revenue opportunity from delivering IoT products and services.
    • It is also the total of the costs paid by each of these sectors to the IoT suppliers (so it is an input and not part of the value-add for the sector as a technology user).
    • However, there is a partial overlap with the $1.9 trillion economic value-add figure. We included the value-add to the IoT suppliers. So, an estimate of the value-add portion of the supplier sales is included in the value-add for manufacturing, communications and computing service industries (where the manufacturing sector provides the hardware, the communications sector provides the connectivity, and the computing service sector provides software/analytics/services).
    • The $1.9 trillion total economic value-add includes both the value-add derived by economic sectors as customers (i.e., organizations within the sector purchasing IoT technology/services and deriving incremental revenue and/or cost savings) and the value-add derived by IoT suppliers.

Economic Envelope

  • The purpose of this analysis was to ensure that the market size that we forecast fits within the economic constraints of the global economy.
  • Global economic value-add from IoT by sector was constrained by total value-add data from IHS Global Insight.
  • Value-add from revenue gain and value-add from cost savings by sector were estimated for select years through the forecast period. For the remaining years in the forecast period, value-add estimates for each sector were calculated using an exponential growth model approximation of adoption patterns that were assumed to follow the initial part of an s-curve.
  • Economic value-add for a sector is essentially sales in that sector minus inputs and supplies from other industries (so, for example, manufacturing sales minus the cost of raw materials and energy).
  • Conceptually, the total economic value-add across all sectors is a measure of GDP (actual GDP includes the impact of taxes, subsidies and other factors).
  • Value-add can be derived by either increasing sales, or decreasing input costs, or both.
  • When we looked at the IoT impact in a sector, we considered ways that IoT technology can increase sales (e.g., new product or service opportunity beyond the existing situation) or reduce costs (e.g., smart lighting leading to reduced electricity expense).
  • The use of IoT can involve the replacement (in full or in part) of existing legacy processes — e.g., stop sending a driver/truck to each high-traffic vending machine every two days and start IoT-enabled just-in-time delivery.
  • Value-add has been assessed as a first-order analysis:
    • In some cases, savings in one area (e.g., vending machine refill operations) will lead to cutbacks in others (drivers, transport vehicles, etc.).
    • These higher-order effects were not considered in the current analysis.
  • Products purchased purely for consumer entertainment/utility do not generate economic value-add above the value derived by IoT suppliers. For instance, an IoT-enabled toy with an associated service (e.g., to enable multiple children to play simultaneously via a subscription service) generates value-add for the toy manufacturer and for the service provider (which may be the same company).
  • In contrast, some products purchased by consumers lead to value being derived by specific verticals — for example, use of sensor-enabled cars and connected smoke detectors both improve public safety, which will benefit insurers even if they do not directly purchase the technology (insurers may, however, subsidize or lobby for government subsidies for such products).
  • Some applications, such as LED lighting, offer a generic benefit (energy savings and improved lifetimes/reduced maintenance). These generic benefits apply across industries.