Internet of Things (IoT) Explained for Beginners

The Internet of Things, or IoT for short, means connecting objects to the internet. In this article, you’ll find everything you need to know about IoT, from a simple definition to real-world applications.

What does IoT, the Internet of Things, mean?

In short, IoT is about connecting all kinds of objects to the internet. That’s why it is called the Internet of Things. Objects can be devices like a smartwatch, a car, or even a coffee maker, but also various sensors that measure, for example, humidity or temperature.

Through the internet, devices can share information with each other and receive data. Sensors can also collect information from the physical environment. Based on this data, devices can operate independently or as part of a larger system. Some devices are two-way: they don’t just produce data, but also receive external data, process it, and adjust their behavior accordingly. For example, a fitness tracker can send a notification to the user once the daily step goal is achieved.

The scale is substantial today. According to IoT Analytics, there were around 21.1 billion connected IoT devices at the end of 2025, compared with 18.5 billion a year earlier.

For a business, however, the value of IoT does not come from the devices but from what is done with the data they produce. In Metsä Board's shipments, sensor data was combined with damage records, and it revealed where along the route the packaging was breaking (read the case). On Arla's pilot farm, machine vision monitors cow welfare alongside human supervision (read the case). How to get started with your own data is covered at the end of this article.

When was the Internet of Things born?

Connecting objects to the internet is not exactly a recent invention. The earliest experiments date back to the 1980s, the early days of the internet, when students at Carnegie Mellon University developed a Coca-Cola vending machine connected to ARPANET. The machine kept track of stock levels and drink temperatures.

However, the first true IoT device is considered to be a toaster created by John Romkey in 1990. The toaster detected when bread was placed inside and started toasting. It was connected to a large portable computer that controlled the power supply and lifted the bread once it was ready.

In 1993, the University of Cambridge developed a coffee machine called XCoffee. The coffee maker monitored the amount of coffee in the pot and took a picture of it three times a minute. Thanks to this smart machine, there was no need to walk to the break room only to find out that coffee had run out. The images made it easy to tell when fresh coffee was available.

The evolution of the internet enabled more advanced applications. In 1994, Steve Mann developed a wearable camera, one of the first cameras connected to the internet.

The term Internet of Things was first coined by Kevin Ashton in 1999. In 2000, LG released the first smart refrigerator, although it was too expensive to become a commercial success.

The number of internet-connected devices grew rapidly. According to a widely cited Cisco estimate, the number of connected devices passed the number of people somewhere between 2008 and 2009, and that moment is generally treated as the birth of IoT. It is an estimate rather than a measured milestone.

In 2009, Google began testing self-driving cars. The first driverless car was a Toyota Prius. By 2016, several other companies were also testing autonomous vehicles. In 2014, Amazon launched its voice-controlled speaker, the Echo, which became a commercial success.

Which networks do IoT devices run on?

Technically, an IoT system consists of either a wireless or wired network and the devices connected to it. In Finland there are two practical options: the mobile networks run by the operators, and a LoRaWAN network built specifically for IoT use. An IoT device can be almost any physical object, since an IoT device simply means a device that can connect to the internet to send or receive data.

Mobile networks have their own technologies for IoT devices: NB-IoT and LTE-M. NB-IoT (Narrow Band IoT) is best suited for stationary sensors requiring long battery life, while LTE-M is better for real-time applications. Both run on 4G and 5G networks, and in Finland they are maintained by Elisa, Telia and DNA. Note that 3G is no longer an option: DNA shut down its 3G network entirely between October 2023 and January 2024, and the freed spectrum moved to 4G and 5G. If your device fleet is old, check the network technology before planning anything else.

Alongside the mobile networks, IoT devices can connect to the wireless LoRaWAN network (Long Range Wide Area Network), which is based on LoRa technology. Thanks to its open standard, LoRa is a more affordable option than NB-IoT. LoRaWAN is also well-suited for mobile use, as the connection to the network remains automatically intact. In Finland, a nationwide LoRaWAN network is operated by Digita, which also has roaming agreements with international operator partners.

The network is only the beginning. Processing and managing the volume of data generated by IoT devices requires a scalable and secure data infrastructure, and the data has to end up in the same systems as the rest of your business information. Learn more about building scalable data infrastructure and about how integrations between devices and systems are built in practice.

What business benefits does IoT deliver?

The benefit does not come from the devices but from the data they produce. IoT reduces costs, increases efficiency, improves safety and reliability, and creates new revenue streams when the collected information reaches the place where decisions are made. Measuring on its own produces nothing if nobody looks at the readings.

With IoT devices, large amounts of data can be collected from the physical environment. Data is the core value of the Internet of Things. It can be leveraged in developing artificial intelligence or supporting decision-making.

Data is the lifeblood of AI. Without high-quality data, there can be no high-quality decisions, whether made by a machine or a human. The same applies to decision-making: the quality of the data determines the quality of the decisions. That’s why it is crucial to ensure both sufficient quantity and quality of data. The practical steps are covered in how to tell whether your company is ready for data-driven decision making.

With smart devices and AI, work processes can be automated partially or completely. An IoT system can, for example, perform monitoring and automatically send alerts in specific situations. Data produced by devices, remote control, and automation lead to significant cost savings and improved efficiency.

When a device does not just measure but AI also interprets the observation and triggers an action, the term is physical AI. The article on what physical AI is and how it differs from IoT covers which part of it already works with existing sensor and image data.

Where is IoT used?

IoT use cases fall roughly into two groups: consumer devices, where the device serves its owner, and industrial applications, where sensors produce data used to run an operation. The five examples below illustrate both.

1. Self-driving Vehicles

One of the most interesting use cases for IoT is undoubtedly the self-driving car. For a driverless car to safely navigate public roads, a large number of IoT devices are required. Other road users and road conditions are detected using cameras, radars, sensors, and microphones installed in the car. These devices collect vast amounts of data, which is then used to control the car.

The goal of autonomous vehicles is to reduce traffic accidents and make transportation more convenient. However, development work is still needed before driverless cars become an everyday reality. The degree of automation is described by the six levels of the SAE J3016 standard, where levels 0 to 2 are driver support systems and at levels 3 to 5 responsibility for driving shifts to the system. The US National Highway Traffic Safety Administration describes automated vehicle safety and the levels on its own site.

2. Smartwatches and Other Wearables

Wearable technology is the most visible consumer application of IoT. The most popular are various smartwatches and fitness trackers that can measure, for example, heart rate or daily step count. Data from the watch is transferred to a smartphone via Bluetooth, and from there onward to a cloud service.

Other wearable technology includes VR glasses, or virtual reality headsets. These are mostly used for gaming, but also in simulations, such as training surgeons by simulating surgeries.

3. Drones

Drones have brought new dimensions to photography and videography. Shots that were previously impossible are now relatively easy and affordable to capture from the air. In addition to stunning videos and images, drones are also useful to farmers for detecting weeds or other crop-threatening issues in the fields.

Beyond commercial use, drones have been deployed in rescue operations. Aerial views provide a rapid overview of a situation, improving both the safety and effectiveness of rescue missions. Some of the best use cases have been in mapping large areas, such as during wildfires or major building fires.

4. Agritech

New technology is transforming agriculture toward greater sustainability without compromising efficiency. By using sensors and cameras, the amount of pesticides can be reduced through more precise targeting. Similarly, irrigation needs can be better assessed, resulting in significant water savings.

In livestock farming, technology enables even large numbers of animals to move more freely. Human monitoring can be complemented with automated surveillance. This improves animal welfare without increasing workload.

5. Transport and Condition Monitoring

In logistics, a sensor travels with the shipment and measures location and conditions throughout the journey. With outsourced transport, the shipper usually sees only the handover points, such as arrival at a terminal. Sensor data also shows intermediate storage, how long it lasted and how conditions changed along the way. The value only appears when this data is compared with the condition in which the shipment arrives. The next section gives an example.

What should you do with IoT data?

Collected data produces no benefit until it is combined with other information and delivered to the place where decisions are made. The most common mistake is not the wrong sensor but readings that stay inside their own system, which nobody looks at during the working day. The first question is therefore not what to measure, but which decision or task changes once the measurement is available.

In practice the approach is the same as in any automation work. Look for a task that repeats, takes time and consists of checking information: someone reviews readings, compares them against limits and reacts when something deviates. That is usually where automation pays for itself fastest, and the same scoring works for sensor data and system data alike. The article on where to start with business process automation walks through the scoring step by step.

A concrete example comes from the paperboard industry. Sensors were attached to Metsä Board's shipments to measure location and conditions throughout the journey. What mattered was not the measurement itself, but combining the sensor data with the damage records logged at storage points: only by joining the two sources did it become clear where along the route the packaging was breaking. Read how real-time transport monitoring was implemented.

In most companies the devices and sensors already exist, but their data stays inside its own system. In that case the work does not start with buying hardware, but with connecting the existing data to the systems and tasks where decisions are made.

If you want a ballpark estimate of what repetitive manual work costs you per year and how quickly automation would pay for itself, the automation potential quick estimate answers that in about two minutes. It is free and does not require contact details.

---

Empirica Finland is a Finnish provider of operational AI and automation that builds automations from the data produced by a company's systems as well as its devices and sensors, and is responsible for keeping those automations running.

← Back to home