Keep It Simple Stupid! Thermostat Concept Model Redesign

For the Rapid Ideation and Creative Problem Solving class, taught by Matt Franks, our class will be looking at re-designing the user interface for a Honeywell Thermostat.

To begin with, we created a concept model of the current thermostat interface to understand how the features are laid out and relate to one another. Below is my mockup of the main spaces of the interface below – make sure to click on the image because the screenshot below is just the first level of this complicated interface!

Part of this exercise is to understand where the interface system stands so that we may understand the problems and complexities which would arrive for a user when interacting with it. This allows us to design our own initial concept model for the interface design we’ll be iterating on in class this quarter.

A few problems I noticed with the current Honeywell Thermostat:

1. It’s Ugly and Confusing: There are so many options and ways to have complete control, that it’s visually and technically overwhelming. The majority of users – whether contractors or consumers – probably won’t use many of these features.

2. Heirarchy: The thermostat has four main tabs at the top of the screen, visible at the upper levels. They are: Home, Menu, Fan, System.

The Fan and System levels have very options to control, and their hierarchy of features only goes one level deep. The Menu screen is a mess – it has tons of options which allow fine grained control. Some of the options on the menu screen only go one more level down, except for the preferences option which then has a ton more options. They are so hidden it’s easy to miss them (which might not matter anyway).

3. Unexpected responses: When first touching the up or down arrows on the home screen to change the temperature, it will just wake and brighten the screen from sleep mode. After hitting up or down again, the large temperature reading doesn’t change – that’s because its the current temp. The desired temperature reading is located between the buttons, is smaller than the current temp. reading, and is often blocked by a user’s hand.

Secondly, buttons throughout the various screens labeled “Done”, “Okay”, or “Previous Menu” might behave differently or place in different locations depending on the current screen the user is interacting with. The language and location of buttons of these vital navigational functions are (literally) all over the map. Users aren’t sure what to expect, and may not be able to get to home.

Future Concept

The concept model for my future thermostat interface is all about keeping it simple. Most consumers will just want to know the current temperature, and click and up or down button to easily move it. I also want to add optional icons which would indicate current settings such as when the system is off, in an energy conservation mode or following a pre-set schedule. These icons would add additional information to the user, without visually crowding up the screen. Users who don’t know or care what the icons mean, can still easily use the system.

By hitting a “settings” button, they are given simplified options to control their thermostat. Most use binary functions like “on” or “off”, and at most give the user three options.

Third level screens are only used for more complicated functions like scheduling. An “auto” scheduling feature would learn from the user’s behavior and automatically adjust the temperature based on history of preference.

I’ve purposely removed or simplified functions found in the original Honeywell concept model. My intuition is that 95% or more of users don’t care about these options. By keeping things simple and easy to understand, I’m betting that users will be more likely to use the additional thermostat functions like scheduling in the first place – rather than give up.