|Skills||Sketching, design thinking, product development, prototyping|
|Tools||Balsamiq, Invision, Adobe Illustrator|
During a 12-week UX/UI Design course through the New York Code + Design Academy, I produced a start to finish UX design project, including: ideation, user research; sketching, wire-framing & testing prototypes; and creating an interactive mockup of a MVP.
The product that I developed is a mobile application for searching academic databases that reimagines the search process to tailor the experience to the needs of undergraduate students. Most academic databases are tailored to the needs of expert users and present an overwhelming number of options of unclear function. Academic librarians spend significant time and resources teaching undergraduate students how to navigate these databases, often with only marginal improvements in effectiveness and efficiency. Instead of working to find ways to improve our instruction, I wanted to explore the possibility of improving the interface to reduce the learning curve and need for training sessions.
An ideal solution would involve changes to the databases themselves, but as an interim step, I conceptualized an app that could act as an intermediary between the end user and the existing database features and content. In addition to improving the search experience itself, many databases lack a viable mobile interface, so the app could act as a stopgap for this issue as well. The project goals included striking a balance between ease of access and learning more advanced research skills that could be used elsewhere.
Competitive Landscape & Feature Prioritization
An analysis of existing products and complementary approaches illustrates the gap that currently exists. When plotting learning curve against the quality and control offered by the search mechanism, options either offer sophisticated access to high-quality information at the cost of developing expertise or improved accessibility but significantly reduced quality and/or control. The sweet spot of refined and flexible access to high-quality materials with a minimal learning curve presents an opportunity for product development. A feature audit of competitors, user interviews, and consideration of user goals and pain points were used to generate a pool of potential features. These features were prioritized and narrowed down to focus on creating a minimal viable product.
Testing & Iteration
The initial concept was for a wizard that would help students design the “perfect” search. The intention was to reduce cognitive load by presenting decisions one at a time and provide guidance to allow students to take advantage of advanced search features. Initial testing conducted by soliciting user input orally and manually entering the resulting search query revealed a number of faulty assumptions and problems. The fundamental flaw is that expert searchers don’t meticulously plan but rather iterate responsively. Additionally, student topics don’t typically require intricate search queries, especially in early stages of research, and the wizard-process becomes prohibitively cumbersome for intermediate users who are familiar but still need occasional help.
Segmented approach displays only the most relevant search options based on user feedback
As a result, the new focus became tools for refining with the goal of creating a simplified search interface with only the most used and useful tools, and providing support to diagnose and improve search results with goal-oriented feature groups and point-of-need tips. This early set of sketches shows an approach to displaying displaying a subset of tools based on current goal. The process models strong search techniques including moving from general to specific and making iterative refinements but constrains the available options to those that are most likely to be beneficial at their current stage. A second round of user testing with the updated design was conducted using Balsamiq wireframes loaded into POP by Marvel for presentation on a mobile device.
The final prototype of an initial MVP was designed in Adobe Illustrator and turned into a clickable prototype for demonstration using Invision.