View the Presentation Slideshow
A Toolkit for the Modern User's Ecosystem
Get people interested in their impact on the environment through visuals and comparisons, then spur action to reduce that impact by creating an energy information hub.
At the time of the performance of contextual inquiry, we had a preliminary name for the application and a completely different concept than we ultimately pursued.
View Raw Data View Script and Consent Form View Contextual Inquiry View Task DescriptionsWho is going to use the system?
The primary users of the app will be people who are paying their own bills and managing their own utility services. While anyone living in any setting could potentially be interested in where their resources come from and how much their household uses in a particular period, the people most likely to meet these criteria are heads of household.
What tasks do they now perform?
There is currently no comprehensive system that allows people to do what we seek to provide with this app, but the component tasks can be accomplished separately. People interact with their service provider company on a monthly (or, in the case of trash collection, weekly) basis and pay bills. People learn where their resources come from or how much other people use, by looking online or contacting their utility provider. To track their own usage, people can pay close attention to their bills, water meter, or how many bags of garbage they produce each week, but no specialized tools exist for environment-related versions of the task fo budgeting.
What tasks are desired?
We found that our interview subjects liked the idea of aggregating data about their utilities and wished there were a single place to check all of it. One interviewee also mentioned curiosity about how his family compared to the average and what others were doing differently.
How are the tasks learned?
Our app would make use of both icons and words, and reasonably intuitive placement of menus and coloration of elements of the view. In our design, we assume people know how apps in general work, but make an effort to make things pretty clear in our particular design by using consistent setup of different screens and consistent icons representing the same actions. We also plan on a healthy blend of map- and list-based information to prevent textual overload.
Where are the tasks performed?
The tasks are performed digitally, which means that the user determines where they will be performed. We intend for our product to be simple enough to use while a user is on the move and away from a desktop computer.
What's the relationship between customer & data?
The data in our app, aside from the user's own information, would be mostly public knowledge, like where the power plants and landfills in the neighborhood are. Users can view their own personal usage data and average data, which we hope will inspire a "cohort effect" that motivates users to improve based on the mean behavior of others similar to them. In this sense, since everything is either individual or averaged, a user will have either complete or partial ownership over all of the data. The product is structured to function with any level of information added by the user: so even if they choose not to add any information, it will still allow them to see where their power comes from. Even if they already know all the information, having it all in one place should make them more likely to pay attention to their resource options and usage.
What other tools does the customer have?
Other than this proposed app, people have their own records, monthly bills from their utility provider, and very little else. People have their utility provider as a resource for any possible information, and the standard array of budgeting tools.
How do users communicate with each other?
How much energy or water different households use are not currently commonly discussed topics; we hope this app will raise general awareness of the subject and become a conversation point. For instance, the only time one is likely to notice a neighbor's water usage patterns is during a drought when a single house on the street still has a green lawn. Our goal is to increase awareness, so that people are conscious of usage even when it is not at an extreme level. Our application is not currently designed with the goal of users communicating with other users.
How often are the tasks performed?
The timescale is on the order of several times a week to once a month. Tasks like monthly bill paying forces resource usage back to the front of people's minds, and thus bring users back to update and check this app. The app can also be set to provide notifications based on these billing cycles. It also has the ability to notify users when articles about resource usage, such as drought-related reports, are front-page news or trending on social media. These are triggers which could remind users to check the app again to remind themselves of the particulars of their providers. Other than these periodical returns, however, the app would not be the sort of activity people return to day after day.
What are the time constraints on the tasks?
In general, completing any of the tasks in this app should be easy and fast. Referring to the information presented on the screen or filling out additional information on the profile creates tasks which were previously nonexistent, constituting an added time demand in a user’s day. For this reason, if nothing else, onboarding will be fast and optional. The app will include the option to view some of the provider and resource information without logging in. To further ease the process, it will offer sign up through existing accounts, such as Facebook or Google log-in, and then later prompt the user to fill out the additional usage and location data required for personal data analysis.
What happens when things go wrong?
If things go wrong in the "real world" of resource purchase and use -- for example, if a consumer doesn't pay their bill -- then access to the resource is lost. In terms of our app's usage, if things go wrong the app would not be nearly as useful as it would be when things go smoothly. Since it is designed to be a tool and a quick lookup feature, inaccurate data or frequent slowness would vastly reduce the usefulness. If the app doesn't prompt a user to maintain updated their usage data, the app will in turn give them wrong information. If the app loads slowly, users will easily get bored or frustrated and leave. Rigorous user testing will allow us to identify the most likely software issues and resolve them before the app reaches production.
After synthesizing our contextual inquiries, performing task analysis, and brainstorming possible directions for our work, we devised three viable application ideas.
The first was an individual-based resource and energy usage application, similar in function to OroEco, the website discussed earlier. Users could understand their energy consumption levels on an hourly or daily timescale by including data such as the electricity consumed by each appliance or the water used on each trip to the bathroom. This would drive behavior change through a more Quantified Self and data-tracking approach.
Our second idea was to analyze impact based instead on demographic groups or cohorts, driving behavior change through a cohort effect. For example, college students living in dorms have different impacts than college students in apartments, who also have different impacts than non-college students living in similar apartments. The information provided would not be specific to any individual but an aggregation of the data of college students living in said dorm or apartment complex.
Our third idea was a location-based resource usage information and comparison app. It allows users to enter their location on a map (and optionally, more specific information like specific providers and monthly utility bills) and visualize where their resources are coming from. It also could act as a hub for users to manage and view their information regarding resource use and providers.
Our team decided to pursue the third of the application ideas listed above. Our rationale was that the first option is already being attempted at another company and also involves many complicating factors. One such concern is security. Linking users' credit cards to get information about environmental expenses (as OroEco does) requires a vast degree of consumer trust. Another frustrating aspect of data-tracking is the inaccuracy of personal results that results from using averages when the app's algorithms inevitably cut corners. Therefore, with respect to the second idea, it is not clear whether it is possible to find enough accurate data to create sufficiently varied and comprehensive cohorts for the app to attract a broad audience. Our team wanted to avoid the same problem of unsatisfying pseudo-personalized statistics.
Our third option, however, employs the least amount of personalization and long-term tracking, so the margin of error surrounding accuracy in public-information data and impact assessment can acceptably be much higher. This means that the app can firmly stand behind the data it does have, without leading to doubt on the part of the users as to authenticity and validity. The information hub also allows us to use more techniques to drive behavior change as it combines visual information with various forms of quantitative data (distance, cost, other measures of environmental impact).
