Deutsche Telekom

The initial challenge addressed automating temperature regulation while maintaining comfort - "the user should never get back to a cold home" - while achieving energy efficiency. To comply with my non-disclosure agreement, I have omitted and obfuscated confidential information in this case study.

Heating Behavior Analysis

Understanding German heating systems was crucial for designing an effective smart thermostat solution. Unlike U.S. electricity-based A/C systems, German homes use different heating sources:

• Natural gas: 48% of German homes
• Oil: 25% (primarily older buildings)
• Both systems heat water distributed to desired areas

German Heating Sources Research

Inspired by the IFTTT app model (conditional device combinations), the design challenge was creating powerful functionality without excessive complexity for broader audiences.

Design Thinking Workshops

Multiple iterative sessions addressed holistic user interaction approaches, app navigation structure and layering, and optimal information architecture.

Workshop Focus Areas

• Holistic user interaction approaches
• App navigation structure and layering
• Optimal information architecture
• Device pairing and configuration flows

Design Thinking Workshop

We organized multiple design thinking workshops, we set up an iterative process to consider all steps needed to create an innovative product. In the workshops, we had small task groups that focused on a specific part of the product. My team and I worked on the holistic interaction approach for the users, how many layers should the app have to be understandable. How will users navigate through the different sections and so on.

Design Thinking Workshop

How might we... Exercise

Digital Thermostat for European Market

We designed a digital thermostat interface specifically tailored for the European heating market. Unlike US-based AC systems, European homes primarily use water-based heating distributed through radiators. The interface needed to accommodate different heating sources while providing intuitive temperature control and scheduling capabilities.

Making Thermostats Smart

Now that we had the individual control of a single thermostat we wanted to include a heating plan. Users wanted different temperatures in different rooms at different times of the day. So we had to design a completely adjustable program that the user can create depending on his individual needs. The Heating panel opens when the user pairs a new thermostat or he navigates to the device recommendation screen and clicks on heating.

In the Heating section, the user sees all rooms that have at least 1 thermostat. 2 default programs are already available so the user can experiment with them first.

When he clicks in the upper right corner he can add a new plan besides having some settings for the heating system itself.

Users can add a new plan with customized heating times. They can easily add or remove heating periods such as copying a complete day or week and re-use it in different rooms.

Once the user has finished editing he's a customized profile, this will appear as a new program in the Heating system screen, He can easily switch between heating programs.

Adding 'Scenes' Feature to combine devices

If we think about the most common devices we use in a group we think about devices from the same species, e.g. Lightbulbs, shutters, or thermostats. But what if we wanted to mix lightbulbs with shutters? This is exactly what we wanted to give our customers, extreme customization. When activating a Scene there is something that goes into a certain status. So the user needs to configure the status that will be activated with the scene and having the option to create a second trigger in the same scene that then will activate a second status. In this form, you can use a scene to turn on a lamp to 88% brightness and put the shutter to 20%, and optionally set an off trigger. This way you can now trigger your 2 defined configurations with only 1 device card.

Adding Automations to overlord of smart home customization

The term Human-centered is what drove us to create this concept for the maximum customization feature "Automations" To reduce the complexity and decreasing the learning curve significantly we decided to create a screen where the user always returns to see the configuration of his automation to see what he already has programmed. By having the dialog "If this - Then that - but only when" it made it easier to understand.

So by adding something in "If this" the user can select what will start his automation, this can be a device, a certain time, or even a specific location the user reaches. In the section "then that" the user sets what will happen when it activates, and optional is the section "only when" that represents a status that needed to be fulfilled to activate the automation.

In this example the user selects 2 devices as trigger, if the Netatmo rain sensor "detects rain for over 10 minutes" and the thermostat 1 "temperature is over 25°" then the living room shutter "is 80% closed".

Final Implementation

After all the iterations we had in the Ideation and creation phase we started to generate some content for the recommendation automation. The idea was to inspire users to set up some automation even if the necessary devices are not paired yet, this way users see what's possible and consider buying devices based on generated needs through the recommendations. We also had to remove the "Only When" feature from automation because from the technology side it was not feasible at the moment, but in the long term, it is still desired to implement.