Internet of Things: Should Designers Focus on Devices First or Services First?

Dorothy Shamonsky

Dorothy Shamonsky, Ph.D., is a User Interface/User Experience (UI/UX) Designer for ICS, who holds broad practical experience and theoretical knowledge in the field and works extensively on new touchscreen product development at ICS.

By Dorothy Shamonsky | Tuesday, February 17, 2015

So often, the Internet of Things (IoT) is discussed in terms of the technology that enables it, particularly focusing on cool, smart gadgets that will propagate in our lives. There is inevitably mention of in-vehicle entertainment and navigation, smart home appliances, wearables and robots. At its foundation, the IoT represents a whole lot of technological innovation, albeit driven by certain human desires, but herein holds many opportunities for usability design!

User experience (UX) designers will need to meet the challenge of making all of these new devices and services a success with users. UX designers will still be asking users fundamental questions such as, “Is it convenient? Is it transparent to use? Will you get any benefit from it? Do you like it?” Designers will also require new knowledge that relates specifically to the IoT. In particular, designers need conceptual frameworks to approach this new context of use

At the outset, there are (at least) two different broad approaches a designer can take to an IoT project, the first, is to focus on designing devices for which logical services evolve, or second, design services, which then evolve to include appropriate devices. If you are a designer reading this, I can guess you will be most comfortable with the approach that most closely resembles the priorities on your current projects. If you are focused on devices, you will likely see that as a starting point and if you are focused on services, you will begin there. Is one approach better than the other? Read on and see.

Discrete Devices Now Connected

Here at ICS we provide a wide variety of design and development work and much of our business is embedded devices or systems. In recent years, we’ve moved a portion of our design practice from graphical user interfaces (GUI) to touch or natural user interfaces (NUIs) as requested by clients. In this past year, greater connectivity has become a theme with clients to extend their digital services. Considering embedded devices are often entry points to both monitor and control a machine or system of machines, ease of use for each individual device a paramount. It’s hard for us not to focus on these dedicated devices themselves. As designers, we tend to think about input and output modalities a lot, and prioritizing design for device-specific user experiences.

Clearly, our customers, more and more are thinking about building useful connected services around their devices. This feels like a very natural way for the IoT to evolve, growing from existing devices that users both need and are already comfortable using. Although one could argue that starting with a digital service and working toward a design for an appropriate physical device is a better way to proceed, there is an argument against it.  When imposing new devices that will require new behaviors and habits in the physical world, it is much harder to do than customizing a digital service around an existing product.

Value is in the Service

For those designers already working on delivering consistent content and experiences across multiple platforms, focusing on digital services is a natural entry point into IoT. The real power of the IoT may lie within digital services along with the ‘smart’ data that is collected for use. However, the convergence of both the devices and the vast amounts of data gathered by cameras, sensors and other networked tools make the IoT road to usability a challenging one.

User’s attention will be divided between a myriad of devices. The content of each device will need to be appropriate for the attention that it may (or may not) get from users. Adding to the attention issue, the volume of data will explode. Raw data is just information noise. Data needs to be crafted to provide real value to a person, essentially made into stories that the user can process.

An argument against the device first approach is that digital services require more attention than piecemeal evolution, in order to provide longevity and robustness.  Personal convenience devices such as wearables may be supported just fine on lightweight digital services, but supporting factories, power plants, transportation and the like requires a service first approach. People’s lives can depend on a smart infrastructure functioning correctly. One would hope, once in place, these devices would last for years to come.  

In conclusion, I would argue that users are more comfortable with incremental changes in their patterns and habits of using technology and from that perspective, a devices first approach works better. Digital services can be difficult to understand and often remain invisible to the user. Witness the frustration with cloud services and the often misapplication of ones own data. I recently was alerted by my desktop notifications that a person I only marginally knew was having a birthday “tomorrow.” Did I care? No. Meanwhile, a good friend who was also having a birthday “tomorrow” was not mentioned at all and I almost forgot it. 

It has become clear, in order to build robust digital services to support large, industrial systems, a service first is likely a better approach. As is often the case in design, there is no one answer to the question, should designers focus on devices first or services first? For now, one thing is for certain with the IoT, there are only more contexts to consider.

Have a question or add to the conversation: Log in Register