These are my takeaways from the McKinsey Global Institute report on the Internet of Things. The authors hypothesized that the hype may actually understate the full potential — but that capturing it will require an understanding of where real economic value can be created and a successful effort to address a set of systems issues, including interoperability.

Any errors, omissions, or misrepresentations are mine.


Significant technology changes have enabled the rise of IoT: ubiquitous, low cost wireless coverage; advances in big data and analytics capabilities; ecosystem emergence (specialized vendors in hardware, software, system integrators, and a community of users).

There are >9B connected devices around the world now. This is expected to grow to 25B-50B by 2025.

Value creation

Interoperability is necessary to create 40% of IoT potential value ($4T/year out of $11.1T by 2025). This is a complex systems design challenge that requires coordination on many levels — technology, capital investment cycles, organizational change, etc.

Organizations that use IoT technology will need better tools and methods to extract insights and actionable information from IoT data. Additionally, management innovations, organizational changes, and new business models must be developed and implemented. This could lead to a new “productivity paradox”—a lag between investment in technology and productivity gains that can be seen at a macroeconomic level.

More IoT value is likely to be created in advanced economies because of the higher value associated with each deployment, but the potential number of IoT uses is likely to be higher in developing economies.

Customers will see substantial value from IoT innovation, but will need to find ways to manage increasing information overload. On the other hand, there is more economic potential for IoT applications in B2B settings or in settings where consumer IoT systems can be linked to B2B systems.

IoT will enable (or force) new business models, and companies that leverage this will enjoy sustainable benefits. This could occur in three phases (like with PCs and the Internet): (1) “arms suppliers” succeed by providing the building blocks of the infrastructure; (2) companies build broadly scaled applications; (3) companies build adjacent businesses. End-to-end solutions today (enabled by the need for interoperability and customization) will lead to more "horizontal" platforms in the future. Businesses will differentiate on technology, data, and software platforms. Division of value will shift towards software and analytics suppliers over time.


Human: smart pills, nanobots, wearables and continuous health/treatment monitoring, augmented reality. Benefits in health, fitness, productivity, safety.

Home: chore automation, energy management, security. Customer usage data can drive future design. Interoperability is crucial here.

Retail: automated checkout, layout optimization, customer tracking, inventory optimization, smart CRM, real-time/in-store personalized promotions, inventory shrinkage prevention, energy management, condition-based maintenance, improved staff allocation. Omni-channel shopping erases distinction between online/offline shops. Can benefit manufacturers of goods too. $410B-$1.16T economic impact by 2025.

Offices: automated security, energy management (heating, cooling, lighting), human productivity improvements.

Factories: operations optimizations (manufacturing, farming), productivity improvements, better equipment maintenance, inventory optimization, worker health/safety. Example in manufacturing: auto-sensing equipment, condition-based maintenance, automated quality control, real-time dashboards, production/supply chain optimization. Augmented reality can provide heads-up display of useful data/information. IoT production/inventory systems provide greater visibility into supplier economics, which can lead to better pricing.

Worksites (e.g. oil and gas exploration, mining, construction): improvements in operating efficiency (planning, coordination, automation), equipment uptime/maintenance/useful life, human health/safety.

Vehicles: monitor and improve performance, condition-based maintenance, safety and security. Enables new service and business models, new customer interactions (traffic data, entertainment, productivity), dynamic insurance pricing.

Cities: transportation (traffic flow, autonomous vehicles), public safety (crime monitoring, emergency response, lights/building structural health), public health (air/water quality monitoring), resource and infrastructure management (electricity/water distribution losses), service delivery. Highly dependent on regulation and governmental technical depth. Interoperability is paramount. Could lead to new tax models.

Outside: logistics routing and precise navigation of ships, airplanes, and other vehicles (autonomous or not) between cities, tracking containers and packages in transit.

Enablers and Barriers

Technology: basic hardware costs/power consumption must drop — sensors, MEMS, RFID tags for tracking, battery power, data communication links, computing and storage. Ubiquitous connectivity is a must. Analytical/visualization software must also improve.

Interoperability: standardization in technology stack, integration between vendors, open systems, platforms, and standards. Standard protocols for sharing/accessing data.

Privacy/confidentiality: provide compelling value prop and transparency in collection/usage of data to establish trust with consumers.

Security: interconnectivity multiplies normal data communication risks. Consequences of potential failures/breaches/even bad data forecasts must be managed.

IP: who has ownership of data produced by connected devices? Need industry guidelines or standards.

Organization/talent: IoT bridges technology, physical assets, business metrics; companies need to link teams making operational decisions with those with expertise in data-driven decision making. Hardware-focused companies must develop core competencies in software. Organizations must develop new processes to enable algorithms to make decisions, while allowing decision makers to monitor metrics and set policy.

Public policy: regulation enabling operation of autonomous vehicles/machinery, creation of new market rules driving adoption, rethinking compensation models e.g. in healthcare, creating rules for collecting/sharing data.


Big data and advanced analytics on IoT data will allow companies to personalize services based on consumer behavior, usage, and context. The continuous stream of real-time data can be used for customer microsegmentation and dynamic pricing models.

Service-based business models are enabled for new kinds of products, e.g. power, transportation. IoT can allow products to not depreciate in value, but become better while in service — Hal Varian's "product kaizen".

IoT data can also be monetized, subject to privacy, confidentiality, and ownership rights.

Foundational technology suppliers (hardware, sensors, IoT device clouds) and installers of IoT systems will capture less value in 2025, giving way to makers of packaged software, services, analytics, and applications developers. This may occur in three phases:

  1. Connectivity, sensors, physical setup/infrastructure, devices/hardware
  2. Software, analytics, platforms, security
  3. Adjacent business models (e.g. analogues of ecommerce, Amazon, Airbnb)

For companies to maintain share over the long term, they need four distinct competition bases:

  1. Distinctive technology: low-cost, low-power sensors and connectivity, proprietary IP. There is a risk of commoditization for IoT components which may require development of more complete solutions.
  2. Distinctive data: access to real-time streaming data + large amounts of historical data.
  3. Platform providers: successful platforms exhibit network effects and tie the customer to the platform provider. Large-scale platforms tend to be developed later in the evolution of an industry (when a critical mass of successful use cases and solutions has developed).
  4. Ability to provide end-to-end solutions: companies providing hardware, software, installation, and service can build deep relationships with customers.

Competing vertically (specific niche) requires deep understanding of the industry and solution's setting + integration across the value chain. Competing horizontally (platforms) requires a critical mass of vertical solutions satisfying specific consumer needs. Other IT technologies have seen vertical solutions being developed first, which then lead to scaled platforms.

Companies seeking to cross industries need to specialize in very basic tech (hardware or connectivity), or develop expertise and customizability for specific verticals.

Using the same software and algorithms across different deployments could be challenging, though there are some algorithms that could be used in different implementations: pattern recognition of streaming data, resource allocation (keeping specific constraints + impact in mind), 2D/3D layout optimization, path routing, computer vision and hearing, emotion/mental state analysis.

Invest in capabilities, culture, and processes as well as technology. If you are small, find ways to obtain data on the scale required to compete with larger companies that already have the data in-house.