Amazon Web Services Data Exchange (ADX) is a hub for companies to share data assets with consumers for easy accessibility. Partnering with BrainStation for our 24 hour Industry Project, ADX tasked us with providing a solution to a current user issue with their website. Over the course of 24 hours, teams consisting of UX designers, software engineers, and data scientists worked together to design a solution.
ADX provides a marketplace to support organizations of all sizes with a centralized place to access datasets.The current marketplace structure can be filtered, but is not searchable, user-friendly and the information about the datasets is hard to find and compare.
A cost-effective new search layout with visualized dropdown filters and indicators of the user’s journey through the site, leading to a selected data product page with a metadata-based recommendation engine that utilizes data from users’ past searches.
Once we were provided with the problem statement and our teams, we immediately converged and created a name for ourselves, The Titans of Industry!
After getting further acquainted and understanding our roles, we began to develop a work flow and schedule. From there, our design team determined the parts of our process we felt necessary to develop a solution, including:
After establishing our workflow and approach as a team, we jumped right in with the How Might We question we were given:
With this defined, we wanted to better understand the key issues directly from users by researching reviews and conducting a user test with a data acquisition associate.
“It could be easier to navigate through the functionality.”
“Confusing, buggy, and misleading UX. Confusing documentation that is often not detailed enough or erroneous.”
“Some filters feel redundant and I’m not sure what’s available in each dataset from these blurbs. I’d like to compare other similar sets, see what else is out there for what I need.”
Based on our user test, we conducted our own design audit from the user’s experience. Noting these main areas of confusion and hesitation:
Some of the available filters felt redundant and confusing, and the list appearance becomes overwhelming to the user.
There is no clear indication where the user is currently on the site, leaving them feeling lost and unable to place themselves.
Many of the products lack essential information such as a data dictionary, necessary for the user to find the right product for their needs.
Taking into account ADX’s key competitors, we walked through their sites and compared experiences.
“Google Cloud Platform optimized the dashboard and some features that make it easy to use.”
“Snowflake is a very good tool for analysis, easily used.”
With the research, we felt that a clear user would help us better understand how to approach the problem space and narrow down our scope due to time constraints.
As we developed our solution, our team laid out the key components of our approach we aimed to place an emphasis on. Understanding AWS and their business goals, our team wanted to create guidelines for our solution that would fit AWS' brand. Our 3 main guidelines became:
With the guidelines and our approach settled, our team narrowed down our concept to key elements we felt could be implemented and others approved upon to create a minimum viable product within 24 hours:
Conferring with our software engineer, we decided on a flow consisting of a few pages that would be feasible given the time. Someone like Zoe Parker would want to locate a dataset quickly and be ensured that it meets all of the requirements necessary for her to achieve her goal.
Skipping sketches to speed up the process, we laid out wireframes just for desktop as our main user would likely be utilizing the site through their desktop version as opposed to mobile.
Here are the main changes our team focused on, utilizing assets pulled directly from AWS’ website like the navigation bar and logos, to save time.
With the clock ticking, we transitioned the wireframes into a high-fidelity prototype in Figma and passed along the mockups to our software engineer to make our solution a reality.We had a working prototype and live version produced by our software engineer with just 10 minutes to spare!
Given just 24 hours, there is so much more work to do to bring this prototype to the next level and successfully incorporate all of our ideas.
It’s so important to check in with your team and ensure everyone is at a point their are comfortable with, allowing people to express any concerns they have. Having a stand up every hour or so proved to be very helpful as things can get lost in communication, especially with such a tight deadline.
Packaging and setting information for engineers as UX designers is key for ease of transition as well as keeping them updated on the UX process and where the design stands.
