Helping a task force aimed at eradicating hunger on Staten Island improve their mobile website experience by analyzing user data
BEHAVIOR ANALYTICS
Project summary
Staten Island Hunger Task force is a coalition of food pantries, soup kitchens, temples, churches, and other organizations that want to ensure that Staten Islanders get enough food. They have a two pronged aim of connecting community members in need of food to pantries and also connecting the pantries to government resources. The aim of this study was to evaluate the usability of their current mobile website using behavior analytics tools and provide recommendations to make it more user friendly.
Being able to locate nearby pantries allows the users to get enough food for them and their families (Image credit - Unsplash)
DURATION
4 weeks
TEAM MEMBERS
Ariella Brown, Johna Shi, Samika Rastogi
ROLE
UX researcher
TOOLS
Google Analytics & Hotjar
PROBLEM STATEMENT
How do you find the user pain points while locating nearby pantries without any direct testing and relying on behavior analytic tools?
THE PROCESS
Gathering data from Google Analytics & Hotjar
We used Web Analytics to collect user data in this study. We looked at key metrics like the number of users, sessions, avg. session duration, bounce rate, behavior flows etc. by using Google analytics. We also used Hotjar to get more detailed and qualitative data like where were the users clicking, how far were they scrolling and also session recordings.
THE PROCESS
Analyzing the data
We started the data analysis process by framing four research questions. These questions guided the synthesis of information and determined which aspects of the website we focused on. This process also made us aware of the limitations of our tools as we could see quantitatively what our users were doing but didn’t exactly know the reason behind it. We focused on the homepage and the pantry’s page as they were the most visited pages. After going through the data and recordings we found many insights which we organized in an affinity map.
Affinity map of the insights generated by the team
Overall findings
After going through the user behavior data, we found that the STHTF’s mobile website had really valuable pantry resources information. However, it needs to be shown in a more user centric and mobile friendly format. There were three key areas where the data indicated that the users were facing a problem and we decided to focus on them and come up with potential solutions.
FINDING 1
The Site’s Landing page does not appear to users as a Home Page
The current homepage is not responsively scaled for a mobile website. When a user lands on the website they see a lot of empty space and mailing list information as shown in homepage(/) below. This makes them think they are not on the correct page and they click on the hamburger menu to find it.
While analyzing the google analytics data we also noticed that in the behavior flow diagram where we saw that they clicked on the index page to look for the homepage. This showed confusion amongst users and not being able to find the information they need.
RECOMMENDATION 1
Prioritize information above the fold to familiarize users with the organization
Our recommendation was to restructure the homepage in order to display the key information in the very first fold. An image of the organization's work along with a brief about who they are would help users more motivated to interact with it. The key actions which is finding the pantry could also be shown as a quick link.
FINDING 2
Users appear to be confused about where to locate pantry information
While analyzing the behavior flow we also noticed that a lot of the users were going back and forth between the Home Page, Pantry List and the Area code pages. This meant that they were having difficulty locating a pantry which was near them. We also found a hot jar recording of a user unable to find a suitable pantry
A Hotjar recording that shows the user trying to:
1. Navigate to the homepage from the homepage by clicking on the logo & the menu
2. Click on the pantry names on the pantry list page in order to find more
3. Going back and forth between the homepage, pantry list & zipcode page
RECOMMENDATION 2
Restructuring and consolidating pantry information into one section
Our recommendation was to firstly rename the pages to make it clearer which information is going to be nested there. For example, 'Pantry List' could be renamed to 'Find a Pantry' which could help users know which information was going to be there.
The second recommendation was to provide a map and list view as tabs on the same page so that the users didn't have to navigate to different pages. Quick links could also be provided for the zip codes so that they don't need to scroll that much.
FINDING 3
Finding relevant details on an extensive pantries list is difficult for users
The pantry list is very long and requires users to scan a lot of irrelevant information to find important data. As shown in the heat map most users drop by the 3/4 th of the page which leads to a lot of valuable information going unread. They also were trying to click on pantry names as they seemed like links.
RECOMMENDATION 3
Organizing the pantry list page based on zip codes to make it easier to locate nearby resources
The pantry list could be based on area codes that auto scrolls to relevant sections. This would remove the endless scroll. Secondly the pantry title would be clearly shown an non clickable through the typographic usage and the links would take to google map locations.
A/B test plan
We proposed to our client to run an A/B test on the existing homepage and the version recommended by our team which prioritized key information and was mobile friendly. Our metric for determining the success of the test would be reduction in the behavior flow interactions:
-
from “/” to “.index.htm”
-
users bouncing between multiple pantry finding resources
Conclusion
This project helped us understand both the value as well as the limitation of digital analytic tools. It exposed us to a massive amount of data and we could judge the usability of the mobile website by identifying patterns in it. However, because we didn’t directly interact with the users there were gray areas which would have benefitted from user participation and testing.
We presented our findings to our client through a slide deck report and a one page visual summary. They liked our presentation and found the insights useful. A key takeaway on their end was optimizing the homepage for mobile view. Their feedback was also really helpful for us to understand how could we further improve our study.
