Today, a vast number of smart home technology providers are scrambling for a market that is currently worth 20 Billion. Here in Germany, at least every seventh person supposedly uses smart home applications. And the market is growing by 17% annually. But every fourth person finds that using smart home applications is too complicated, in fact, even people who do not use smart home applications feel this way. This is a good reason to take a closer look at the topic of the user experience in the smart home.
It becomes clear that this topic cannot be restricted to just smart home apps, but rather, smart home technology has to be seen as a whole, since providers of these technologies advertise with use cases like switching on and off the light using a smart phone. To switch a light on, the user has to unlock the screen, look for the app, open it, look for the lamp to switch on, find it, and switch it on. Clearly, more complicated than using a light switch.
In order to shed some light into the darkness and reliably optimize the user experience in smart homes, we opted to use the Rapid Contextual Design (RCD) process.
Where do inhabitants of smart, and not so smart, homes actually feel the pain? Is it really the light switch?
2015: The Simplest Configuration is the Smartest!
We performed qualitative context interviews with users at home – including owners of fully-fledged smart homes, users who solved specific problems with isolated applications and others who have no smart home technology at all. After structuring the statements in an affinity diagram, we saw some general assumptions confirmed:
- The people interviewed continuously attach importance to the atmosphere that they feel comfortable in. Light, temperature, air quality and music are, after security, the most important factors.
- Among the classic topics: comfort, safety and saving energy; the topic of comfort turned out to be the most important one. The topic of safety is valued differently by different users, usually based on the living situation. Energy saving is a topic for inhabitants of existing properties who don’t see a way for optimization without performing major structural alterations, and similarly, the surveillance of consumption data was also seen as redundant.
- It was surprising that automation was viewed skeptically: Reasons being that current solutions don‘t work. People are afraid to lose control. And there are activities that they don’t want to be taken out of their hands because they like to do them.
- It also became clear that the control of lights, blinds, ventilation and such should work as comfortably as possible: This means that blinds should not be controlled by the in-house display somewhere in the hallway, but rather, with a switch next to the window and thus have a clear proximity. And the switch has to work without a delay.
- Inhabitants who already have smart home applications that allow for flexible connections between switches and sensors described the configuration as extremely cumbersome. That’s why the settings have been modified only once and then never again — despite high dissatisfaction with the current configuration. They preferred to stick with the status quo.
- Further, it is a great challenge for users to think about automation rules that prove valuable in everyday life. Life is only partially routine and so it doesn’t always fit into exception-less scenarios. What a pain when you are sitting on your patio in the evening at the beginning of dusk and your shutters go down and you cannot get back into your house <lol>.
Therefore, we see the greatest potential for optimization – based on the current state of the art – in the configuration of switches and automation rules.
Example of our Tablet User Interface that simplifies the generation of automation rules.
What is Possible in the Future?
But even if the configuration is simplified, as described above, there is still the question of whether this is a significant improvement in comfort for the user. That’s why we went a step further and discovered, through so called ‘visioning’, further possibilities that promise even more comfort in the future. As a yardstick for the technical feasibility we shifted to the year 2025.
The solutions found in this way are not completely new ideas. But they were developed purely based on real user requirements, which guarantees that there is an actual need for them. And we can draft them in great detail by leveraging the statements from the affinity diagram.
2025: Only Self-learning is Really Smart!
The idea with the greatest potential – in our view – is a self-learning system that generates automation rules by itself. It has to be able to predict recurring user actions (e.g., blinds up) in connection with a specific situation and to perform the activity automatically.
This would be a real plus in terms of comfort and it solves a big part of the operational problems for the user. Scenarios can combine parameters such as brightness, time of day, day of the week, presence, movement, temperature (if necessary), season, human vegetative signals and more. The rules could be exponentially more complex and thus be able to describe the full diversity of life. Of course, this requires a specific willingness of the user to trust the system and to support it during the learning process.
Also, in this scenario, we think the light switch is still a necessity, even if it changes its purpose. On one hand, it could be that an inhabitant suddenly has a different need than the one predicted by the system. And they need to have the ability to intervene in order to not feel disempowered. On the other hand, the switch becomes a communication medium with a self-learning system. The user uses it to convey their needs.
Example of storyboards for a bedroom scenario with the values that are detected by sensors on different days and at different times. The basis for self-learning systems.
Bettina & Josef
More to this topic:
- From Rapid Contextual Design to Structured Requirements