Combining Big Data and User Experience Analysis for the IT@Cornell Website
The IT@Cornell website provides information about IT services, service announcements, and technical documentation to the Cornell community. Our goal is to continuously improve information and support for students, faculty, and staff. How can the IT Communications and Documentation (Comm and Doc) team use big data and user experience (UX) analysis to measure and improve community use of the website?
The first step is to measure the impact of communications campaigns. We will compare Google Analytics data for the IT service page before and after the campaign. Did visits to the service page increase? Did users take the recommended action?
The next step is to combine Google Analytics data, User Experience (UX) analysis, and Siteimprove heat maps to optimize page usability. This is a new process for the Comm and Doc team. The goal is to develop a prototype process and evaluate whether it is an effective method for enhancing our pages. We will do the following:
- Examine Google Analytics data, Siteimprove heat maps, and conduct UX testing on a single IT service page
- Work with the service owner and manager to identify measurable goals for user behavior on the service page
- Modify the service page based on the data and UX analysis
- Control: Make the changes when there is no marketing campaign scheduled for the service, to prevent data contamination between communications efforts and web design improvements
- Measure whether the desired user behavior increased on the page
- If successful, follow up with a Communications campaign to take advantage of the enhanced service page
- Goal: Develop a prototype for using this process on other service pages
Questions:
- Is there any measurable improvement, or is a communications campaign also required to change user behavior?
- What improvements could be made to this prototype? Are there additional data sources that could be included for better results?
- What constraints or challenges did we find when we used this method?
- Is this an effective method for improving service pages?
- What are the costs and time commitment required to use this method?