Energy Conservation in the Home

Project Overview

For this project, my team and I wanted to focus on how to encourage energy savings. Our user group was young people who are living together in a home. The users were trying to conserve energy, save money, and be eco-friendly. They were also trying to accomplish these tasks while cooperating with other household members. A system was needed to address the numerous challenges users faced in trying to accomplish these goals.

Research Problem

Project Overview

For this project, my team and I wanted to focus on how to encourage energy savings. Our user group was young people who are living together in a home. The users were trying to conserve energy, save money, and be eco-friendly. They were also trying to accomplish these tasks while cooperating with other household members. A system was needed to address the numerous challenges users faced in trying to accomplish these goals.

Research Problem

My Team

Yannu Li, Xi Chen, Yizhou (Fred) Liu

My Role

In the research phase, I synthesized the findings from the competitive analysis and literature review that we conducted, contributed survey questions and aided in distribution, and paired with Fred to conduct three of our six semi-structured interviews.

In the analysis phase, I contributed to building our affinity diagram, creating our task analysis, and to identifying user challenges and expectations.

In the design phase, I participated in our multiple brainstorming sessions and contributed to our mind map. I made mock-ups and storyboards for one of our design alternatives. I expanded and added interaction for our final inVision prototype.

In the evaluation phase, I helped form our evaluation plan, led 2 out of 5 evaluations, and performed analysis on the data we collected.


Research Methods

We started approaching this problem space by conducting online research to find competing products on the market that involved cooperation, visualization, energy control, gamification, and smart homes. During this competitive analysis, we found that a more intelligent system than a manual controlled system could ease the burden on users and focusing on more than one energy source would help the products have a broader impact.


Additionally, having a simple way to visualize, but also including suggestions are important characteristics to consider in an energy saving system. Allowing multiple users to cooperate using a system is also a crucial functionality. Finally, making sure the system accounts for a user’s motivations will help the system have a more sustainable impact.



competitive analysis.png


We felt the next step was to design a survey to get a general impression of our users' current energy usage and roommate situations. The goal was to narrow down what the main challenges and motivations were. We created our survey using Qualtrics and received 45 responses from our user group.


Our findings include the following insights:

  1. Main challenges were lack of energy awareness, maintaining habits and forming new habit​s, cooperating over appliance usage, and controlling appliances away from home.

  2. Top motivations for saving energy included saving money, being eco-friendly, and acting on personal values.


Top Challenges


Semi-Structured Interviews

The survey gave us a sense of which topic areas we needed to understand on a deeper level. We used the survey results to form the questions we asked in our semi-structured interviews. Using contact information collected in the survey, we were able to recruit 6 participants who matched our target user group.


I want to be green and do my part to help the environment, but I feel like small changes won't have much affect.

- Interviewee 2

The top insights from the interviews and affinity diagram were:

  1.  Although users have a desire to save energy, they do not have enough motivation to form better habits or complete daily actions to achieve this goal.

  2. Our users knew they wanted to save energy in their home, but did not know actions they could take.

  3. Most users lacked regular awareness of how much energy they were using, besides the monthly utility bills they receive. 

To analyze the findings from our interviews, we decided to do an affinity mapping approach. Our goal was to find the similarities and common themes among our participants by grouping together related observations and notes. 

affinity diagramming.jpg


As we carried out our research methods, our main challenge was finding our participants. Since our user group encompassed a broad set of people, the problem space we were dealing with was too complex and we needed to narrow it down. Through the survey, we were able to find that energy behaviors were dependent upon age and motivation. Older participants (> 45 years old) had a more stable lifestyle and therefore had fewer appliance control issues with housemates and were more aware of energy saving practices. Therefore, we chose to eliminate them and focus on a younger generation (18 - 45) who faced more challenges in regards to energy awareness and energy savings. Additionally, we decided to eliminate people who had no desire at all to save energy since that is the task our users are trying to accomplish. 

By narrowing our user group in this way, we were able to collect a more accurate representation of our user group for our semi-structured interviews. Although, even with these new restrictions, our users still had a wide variety of differences: geolocation, home type, and relationship to housemates. We handled these differences within our design implications.


Synthesis of Findings


Using the demographic data and the motivations and challenges we uncovered during our research, we formed the following two personas to represent our user group:


Journey Map

We outlined the different steps users take when they are using energy in their home and put together a journey map that highlighted the different highs and lows in the process. This method allowed us to identify meaningful design opportunities.

Design Implications

Using our affinity diagram, personas, and journey map, we brainstormed the design implications of our research. We combined our ideas and then prioritized which design implications were most important to address. I will only list the top priority design implications here:

  • Provide continuous stimulation and feedback for the users to change their habits.

  • Involve external help (people, devices, community) to motivate the formation of good energy habits.

  • Suggestions or guidance provided by the product should consider the user's current context and capabilities.

  • Use strong visuals to indicate information and achievements.

  • Involve notifications or reminders to help people to remember to take actions.

Journey Map Style #2.png


Divergent Brainstorming

For brainstorming, we used sticky notes to put up as many ideas as we could think of within half an hour. Our ideas were strongly influenced by the design implications we had formed. To organize our ideas, we created a mindmap.


Convergent Brainstorming

Using the mind map, we focused on the ideas that best addressed our design implications and from there we sketched out 3 design alternatives.

ar game.png

Idea #1: Community AR Game

​Build a virtual iceberg as a community by completing tasks that conserve energy. Use an accompanying app to receive actionable suggestions and mark them complete to earn points for your household and add ice to the iceberg.

Idea #2: Energy Home Assistant

Incorporate energy saving suggestions and reminders into an existing voice technology such as Google Home or Alexa. It would be able to provide quick feedback on utility usage and give reminders to the user when they were using too much energy via a sound alert.  

Idea #3: Digital Painting

Use an ambient display (such as a digital painting) to represent real time changes in energy/utility usage. This display would be paired with an app that gave actionable suggestions on how to save energy. It would be displayed in a common space so that all house members could see it.

Feedback Sessions

Once we had these 3 concepts sketched out, we conducted two feedback sessions.


Round 1

We asked 4 users for their thoughts on our design alternatives. Found that most people preferred the digital painting idea the most.

Round 2

We asked 10 users for their visualization preferences and also for any suggestions they had for the digital painting and app. Users liked using metaphors or artistic representations the best. We found that users wanted the painting to have more interaction and automatic management. A few users also expressed concern over the cost of a digital painting.

Feedback Conclusions

We considered the feedback from these sessions very carefully, listing pros and cons of different paths forward on a whiteboard. While the cost of a digital painting may be reasonable, if we wanted to add interaction (as many of our users desired), then the cost of our product would increase quite a bit. To solve this dilemma, we decided that we should switch from a digital painting to an existing digital platform in the home so that we could add interaction and reduce costs. We chose the NEST thermostat, one of the most relevant technologies in our market research, as our new platform.

This product adjusts the temperature in one’s home efficiently by learning a person’s schedule. It also lets a user keep track of how much energy they are using, although only in relation to temperature. Making this switch kept the benefits of a visualization through metaphor (albeit a smaller surface space), roommates would interact with it frequently since many of our users change the temperature often, and it offered more interaction in terms of scrolling and selecting (allowing us to give actionable energy suggestions through the interface). Additionally, the app component of our idea could act as an add-on to the existing NEST mobile app, making the two components more cohesive. 

Final Design

The first part of our solution is an app that does several things:

  1. Allows users to explore different energy saving tasks and choose which ones they want to add to their daily routine.

  2. Gives reminders for the task lists that the user has chosen to complete.

  3. Allows users to view their utility usage and the amount of actions they have completed (along with actions completed by any roommates).

  4. Is integrated into NEST's current app. A user can choose their preferred visualization in the app and that is the visualization that would be displayed on their NEST thermostat.

You can explore the app through the inVision prototype displayed below.

The second part of our system is on the NEST thermostat itself. Its functionalities include the following:

  1. Show visualization that represents a household's current energy usage.

  2. Allow user to see tips on how to save energy.

  3. Give reminders to complete task lists.

The inVision prototype below gives a sense of the different visualizations that could be displayed (more grass or more polar bears means better energy saving, less means worse). The pictures show what a tip or reminder would look like on the NEST.

Nest_one tip Copy 3.png
Nest_one tip Copy 2.png


Methods & Results

We used benchmark tasking to assess the usability of our app with 5 users. We gave users 5 tasks:

  1. Go through the on boarding pages and explore.

  2. Read the phone notification and complete the task list.​​​​

  3. Find out what the tip is on NEST.

  4. Get the NEST reminder before leaving home.

  5. Interpret the NEST metaphors.

We asked the user to think aloud as they went through each task and gave them an After-Scenario Questionnaire (ASQ) to complete after each task was completed to assess the user's satisfactionOnce all the tasks were completed, we asked users to fill out a SUS form to measure the subjective usability of our overall system.


This chart summarizes the results from the ASQ forms. A lower score indicates higher satisfaction. Therefore, we can see that the tasks users struggled with were task 1 (going through the on boarding pages) and task 5 (interpreting our metaphors).

Our SUS score was 72.5, which surpasses the cutoff of 68, indicating that our system was usable. 

The think aloud method allowed us to take notes on why users were confused in certain places. Going through these, we realized we needed to add clearer descriptions (like what the metaphors were for) and support more interactions (such as being able to click into the tasks on the onboarding page). 

For the next steps, we would modify our prototype based on all of these results and go through further iterations of usability testings and design modification.