Case Study: Globe Site Survey Design and Presentation for easier readability and increase completion rate
Project Overview: CoCo Temp is a open-source platform to track geographic temperature fluctuations. This project is part of an ongoing enterprise project for the Humane Interface Design Enterprise (HIDE) at Michigan Tech in collaboration with Arizona State University, where my main role as the Lead UX research and designer was to create user personas, understand the user journey, and assist the development team in implementing their data survey in a 3 month semester.
Typical User: With various educational backgrounds, a user could span the ages of 18 – 65, as a scientist and/or environmentalist from the United States. This information was sourced from our stakeholders.
Page without NOA Marker
Research Setup: These were the questions we were looking to answer with the Globe Site Survey
Survey Layout and Interaction
Who is the user taking this survey?
Why is completing the survey helpful?
Would a person need to have internet access to complete the survey?
What feedback is the user receiving about their response, and expected time?
Is there transparency feedback?
How might someone be overloaded with information?
How can the presentation of information be simplified?
How can the process of completing the survey be optimized?
Survey Design
Who is the user taking this survey?
What is the clear, attainable goal of taking the survey?: User & Stakeholder
What questions are most important to be answered? (ordering effects)
What verbiage is unclear?
How long should this survey take?
Is the length of the survey an issue?
Are questions clear and concise?
Are users being shown unnecessary questions based on previous responses?
How will this information (quantitive and qualitative) be used?
Research Method:
I reviewed the content and design of the Globe Site Survey and found that not only were some of the questions unclear or wordy, but the survey was not appropriately giving the team quality data they were looking for. The lead developer and I immediately scheduled a call with our stakeholder to address the validity of the survey and, with the stakeholder, developed clear and direct questions to ease understanding of the topics. Capturing the most relevant data for both the user and stakeholders was crucial in evaluating the list of questions that were written, and determining if any changes were necessary or what could be improved.
Individuals were not filling out the survey. We conducted 1X1 interviews with enterprise participants not on the Coco temp team to understand first how they would navigate to the survey, how they interpreted the questions, the context of when they would be taking this survey, and why they did not finish the survey? The survey was an after-thought when creating a profile on the website and was not brought to the user’s attention. The fix we implemented was to notify the user of the Global Site Survey when creating their profile, and if they were able to fill it out at that present time, show how completing the survey would optimize their profile to have other users find their data. Since, this Globe Site Survey was optional, ordering the most pressing items first increased the likelihood of achieving the stakeholder’s goal.
Common language and defining vague terms were crucial in collecting accurate responses particular to question 10: “Is the sensor located in a depression or riparian area where water can collect?” To define a riparian area and give context, a graphic was included with the question.
Survey with Improvements
Key Insights
Internet Accessibility
Current pain points include the typical user may not have access to the internet while in the field when sourcing their data, which may lead them to forget about filling out the survey. Giving users the option to fill the survey out in the field and then have the file upload to the site with logging in resulted in more completed surveys.
Profile Completion
To bring the survey to the user’s attention, we increased the frequency of when the survey was mentioned (i.e. multiple times throughout creating a profile), nudged users toward filling out a survey when logging in by showing how their data related to others in the area and throughout the country,
Common Language
Defining vague terms was crucial in collecting accurate responses particularly to question 10: “Is the sensor located in a depression or riparian area where water can collect?” To define a riparian area and give context, a graphic was included with the question.
Survey Layout on Page
We simplified the visual display of information to avoid information overload by presenting each question separately. We made sure to include a progress bar to show their completion and how many questions were left to give the user a sense of how long each question would take.