Consumers Energy

Dates:
Jan 2014 – Apr 2014

Skills:
Interview, persona, and scenario
Survey
Heuristic evaluation
Interaction map
Usability test
Comparative analysis
Final presentation video

Software:
Microsoft Office Suite
Google Docs
QuickTime

Course:
Needs assessment and usability evaluation (SI 622)

Collaborators:
Katrina Lanahan
Michael Grisafe
Yu-Ting Lu
Wei Wang

Overview

Consumers Energy is the primary public utility company in Western and Northern Michigan, providing energy to more than 6 million of Michigan’s 10 million residents. Recently, Consumers Energy decided to overhaul its website. In order to gather best practices to be incorporated in this overhaul, Consumers Energy contacted the University of Michigan School of Information for assistance. During the semester, our team authored five full-length reports offering constructive recommendations for improving the usability of the site.

Stakeholder Interview

We conducted two stakeholder interviews in order to better understand our client’s goals, motivations, and constraints. Our takeaways from these two interviews are the following:

1. Recommend changes for navigation, overall look and feel, and user flow
2. Focus on educational resources, transactional features related to bill payment, and the ability to report power outages
3. Craft personas and scenarios
4. Concentrate on residential customers only

Reports

To evaluate the Consumers Energy website, we utilized five different research methods, which built on each other and incrementally enhanced our understanding of the website.

1. Interview, persona, scenario
2. Survey
3. Comparative analysis
4. Heuristic evaluation
5. Usability test

Interaction Map

Our team crafted an interaction map to help us better scope relevant areas of the website that we examined in further detail in our subsequent usability reports. The areas we selected were based on our stakeholder interviews.

Interviews, Personas, and Scenarios

We conducted five Consumers Energy customer interviews across a broad spectrum of demographics. From these interviews, we generated three personas and scenarios to help the web design team focus on their end users’ attitudes, goals, and motivations.

Survey

We sent out a web survey via the Consumers Energy newsletter. This survey was sent to 600,000 customers; opened by 130,000; and responded to by 53. This corresponds to a 0.038% response rate.

Comparative Analysis

Within this analysis, we included the following comparators: one direct, one partial, one indirect, two parallel, and one analogous. This comparative analysis allows us to gather best practices used in similar websites and recommend potential areas for improvement on the Consumers Energy site.

Heuristic Evaluation

A heuristic evaluation followed this in which five evaluators carefully went through the Consumers Energy website and noted problems according to Jakob Nielsen’s 10 Heuristics for User Interface Design. Problem areas were then rated in severity from 0–3.

Usability Test

We recruited five Consumers Energy customers to conduct usability testing on the website. During this testing, users spent 30–45 minutes conducting predetermined tasks on the Consumers Energy website. One of our associates acted as the proctor for these usability tests while a second carefully observed, noting areas in which users had difficulty with tasks.

Limitations and Discussion

It is important to note some of the limitations of our study:

1. Our client did not have demographic data on the Consumers Energy user population. However, the company informed us that its user base roughly mirrored the general population in Michigan. Although we attempted to obtain a stratified sample for each portion of our study matching Michigan’s population, this was often not possible. Because of this, our samples may not have been representative of the general Consumers Energy population.
2. Our investigation was incremental, and as we completed each section of our investigation, we may have gradually formed preconceived notions of some of the website’s problems. This potentially made some of our conclusions more biased toward finding certain problems as the process went along.
3. At our client’s request, business customers were not included in our analysis. Because of this, our analysis cannot address the website needs of this population.