Work With Us

Why IoT Data Collection Is Essential To Get Right

(This post is based on the video, “Data Collection” in the Design Defined: Design Principles Explained series.)

One of the most important early decisions to be made when developing a smart device is to consider what kinds of data the product will collect. Decisions about IoT data collection are central to every intelligent product’s development, because of their powerful ripple effect.

IoT data collection impacts data deployment, storage, sensor selection, processing power, security, and more. Most importantly, it’s at the core of every smart product’s user experience — and consequently, its success.

As intelligent devices become the norm, we find ourselves helping clients plan for IoT data collection more and more. While there is plenty of nuance, we recommend everyone follow the four steps outlined below when deciding which data to collect.

1. Always Start with User Need

With so much compelling technology on the market, it can be tempting to let the latest tech drive your product. But product development should always be driven by user needs. To truly understand your users’ needs, you’ll need to conduct user research.

Upfront user research identifies two things: who your users are, and what unmet needs they have.

2. Define the Data to be Collected

Once you’ve determined user needs, you’ll translate those needs into data requirements. How could data help address your users’ needs? What are you trying to learn or capture? Allow these questions to define the data to be collected — and to inform what data you don’t need to collect.

For instance, Nest thermostats keep homes warm and cool, so they need to track temperature and humidity. Ring doorbells track who’s at your door and how often, so they need visual data and time-stamping. It wouldn’t make sense the other way around, and neither device collects more data than it needs.

3. Figure Out What Sensors Are Required

Once you’ve identified the data to collect, you can figure out what sensors are required. How will you capture the data you’ve defined?

Temperature and humidity sensors can gather home temperature and humidity data, and motion-activated cameras can collect visual data for smart doorbells. We worked with one client who launched a wearable device for tennis players that needs to detect the moment the ball connects with their racket — this device required impact sensors.

Keep in mind that there may be multiple ways to collect the same, or very similar, data.

4. Decide How Your Device Will Connect

Connectivity comes next. Will your IoT device connect to the Internet via Ethernet, WiFi, or Bluetooth? This decision will determine how a user is going to interact with your device, and drive cost and power consumption.

To determine the right option for your device, you’ll have to review the latest research, as well as your data collection goals.

Generally, Ethernet is better suited to a line-powered device with a high data-throughput requirement, such as a smart television. A Wifi or Bluetooth connection better serves a wireless, battery-operated device.

Answer Follow-Up Questions

These four steps will get you started and give you the foundational knowledge you need to begin developing your device.

Making these decisions first will also help you answer follow-up questions like:

  • Will your device store data locally or in the cloud?
  • Will users retrieve data through an embedded display or a companion app?
  • How will you secure the data you collect?
  • How much power will your device need?
  • And how rugged will it need to be?

Always connect these questions to the four steps above. Consider user needs, data requirements, sensor technology, and connectivity.

Consider Research and Engineering Early

Maybe the most important thing to remember is to make sure your team is interdisciplinary from phase zero. Include user researchers early in the process to determine user needs, and bring in electrical engineers to advise on technology and feasibility.

By including these disciplines in your upfront work, you’ll get the necessary input into project unknowns ahead of development phases of work. It’s always less costly and time intensive to get these insights before you begin designing and prototyping, and this upfront work will lead to a better, more successful product.

Use The Four Steps To Develop Your Next Smart Device

The next time you start an IoT development project, consider the four steps outlined in this post: What user needs are you trying to address, what data will that require, what sensors can collect that data, and how will you connect?

While these steps are high-level, by considering them early in your process, you’ll be able to drill down into the technical specifications and requirements, and you’ll have a more secure path forward.

In the meantime, you can learn more about user research, data collection, sensors, and reducing risk in connected device development on our blog, and we’re happy to answer any further questions.

Download our Design Defined ebooks to learn about more product design principles!