Parse — A UX Case Study

Introduction

Staying up to date with the news can be difficult due to the overwhelming amount of information from countless sources online, resulting in people struggling to find reliable, relevant, and personalized news. We posed the question, how might we create a news platform designed to provide a personalized experience while still promoting the diffusion and discovery of differing perspectives from credible sources? During the Spring of 2025, our design team sought to solve this issue through a six-week design sprint.

Our goal was to streamline the news discovery process by creating an app that prioritizes content personalization while balancing credibility so that users would have direct access to news that matters to them with guaranteed accuracy at a glance.

Through in-depth user research, iterative design, and comprehensive testing, we developed Parse — a news aggregator app designed to inform younger audiences of current events with personalized content delivered in a format they know and trust.

Responsibilities and Timeline

The timeline was as follows:

1. Week 1: User Research
2. Week 2: Synthesis & Ideation
3. Week 3: Mid-fi Prototyping
4. Week 4: Mid-fi Prototyping & Usability Testing
5. Week 5: Hi-fi Prototyping
6. Week 6: Final Presentations

Research

User Surveys

Our target audience was Gen Z, in particular, young adults who want to stay informed of the current events around them. During the research phase, we first started by performing user surveys and gathering data from 72 respondents to get an initial idea of what methods people use to stay informed. We found several strong findings:

• 88% of our respondents typically consume news through social media such as TikTok and Reddit
• 55% often check the credibility of news articles
• 57% typically consume news through a mixture of short and long-form content

We defined short-form content as videos lasting a few seconds, such as Youtube Shorts and Instagram Reels.

These findings helped us better understand user behaviors, guiding us in reimagining news consumption to be more accessible and familiar to younger audiences. Specifically, we explored how to integrate the engaging qualities of social media, its ability to capture attention and encourage regular use, while preserving credibility, so users can trust both the information presented on our platform and its community.

User Interviews

After our initial survey, we interviewed 6 of our respondents. From these interviews we discovered four main ideas: the need for diverse perspectives in news to prevent echo chambers, the mixture of short and long form content to fit user’s preferences and consumption styles, the ability to fact-check to prevent the rampant spread of misinformation, and clarity on potential biases in the news so that users can make objective interpretations.

This data reinforced our findings from our surveys and further outlined the issues people have with their current method of news consumption and what they would need in a news aggregator app like ours.

Competitive Analysis

We conducted four competitive analyses, two on dedicated news apps and two on social media apps that users reported using for news. We chose to analyze both types of platforms to gather insights and inspiration for our project, which aims to become a hybrid of news and social media features. The apps we chose to assess were Apple News, The New York Times, TikTok, and Twitter.

Based on these insights, we planned to build a platform that combines a clean, visually minimalist design with personalized features such as a “For You” feed, article swipes, and a daily digest. Our goal is to offer a mix of long and short-form content to keep users informed while promoting mindful engagement and to foster a sense of community through interactive comment sections on article pages.

Ideation

For our ideation phase, we wanted to implement ideas that we discovered from our research, mainly how we can implement short-form content and features that appeal to a Gen Z audience

User Personas

Our user personas mainly consisted of younger people, as that was our target audience. We created our user personas to match the type of people who could use our app, for example, people interested in communities, broadening their perspectives, and reliable and easily consumable content.

Affinity Mapping

We conducted affinity mapping to figure out what features people are drawn to when consuming news-based content. We organized three key categories that contained key things that appealed to our audiences: User Habits, Content Preferences, and Credibility. For user habits, we found that people were drawn to short-form content, wanted to browse personal algorithms, and that they often got informed via social media. For content preferences, people thought that short-form content was more accessible and grabbed their attention more, that long-form has more details and content, and that users preferred a balance between the two types, long-form and short-form. For credibility, users desired more fact-checking features and transparency to increase trust

HMWs — How Might We?

From our affinity mapping, we created some key HMWs to target what our users wanted most from our app. These HMWs were:

• How might we design a personalized experience while still promoting the discovery of differing perspectives?
• How might we create a news app for users that offers a wide variety of perspectives and fosters a balanced community?
• How might we reinforce user trust through database and community features?

Solution Sketches

From our HMWs we were able to create sketches of what we thought the app should include, as the HMWs guided the kinds of features that we should build towards

Goals for the sketches:

• Accessible, community-oriented features
• Many different mediums for news, like articles, shorts, and audio
• Ways to gain various different perspectives on topics
• Transparent credibility

Some features that we came up in our sketches incorporated what we thought users wanted based on our HMWs. Some notable features from our sketches were ‘For You’ and ‘Top Stories’ pages to create personalized recommendations for users and allow them to keep up to date with important topics. We created dedicated pages for the different news mediums, like a dedicated article page, short-form content page, and audio page. We sketched out a comment system to increase perspectives across a large audience. We also created a bias meter and fact-checker to show people what bias the news articles have and if the facts contained inside on them are true or false.

Information Architecture

We created an information architecture to figure out features that we wanted to prioritize going into our mid-fi phase

We targeted specific pages that would be included in our news aggregator app. These pages were a For You/Top Stories home page, a discovery page, a short videos page, an audio page, and maybe an account page.

We gained new insights as a result of the information architecture:

• We created formatting for each page and how they would connect
• We established hierarchies for each feature and how we could differentiate each medium effectively
• We created a flow that would allow users to be able to navigate our app in an understandable and efficient way while still putting news and informing users at the forefront
• Found homes for features included in our sketches that didn’t have a place on any of our pages yet, for example, our Article Swipe feature

Mid-fi Prototyping

To begin our Mid-fi Prototyping, we refined our low-fidelity designs to solidify our five main pages:

1. Home: View articles based on a personalized algorithm or trending topics
2. Discover: Discover articles across diverse categories and perspectives
3. Shorts: View news in a scrollable short-form medium
4. Audio: Listen to long-form, auditory news hands-free
5. Article: Read articles with a bias meter and community features

As we moved forward, we experimented with various layouts and organization in order to create a design that best fit user-centered needs. We also implemented light prototyping to simulate the user experience and demonstrate how navigating the app would feel in practice.. This step prepared us for usability testing.

Usability Testing

During our usability testing phase, we conducted six user tests with participants between the ages of 20 and 22, all of whom were university students. We asked them to complete these main tasks:

1. Navigate through new user onboarding
2. Access your For You page and your Daily Digest
3. Access the Explore page and your Article Swipe
4. Read an article
5. Play an audio and watch a reel

From our tests, we discovered three overarching issues our users were facing:

1. Unintuitive Features
2. Disorganization of News Topics
3. Hierarchy and Readability Issues

Unintuitive Features

Users had difficulties understanding how to use or interpret certain features in our design.

For example, 4/6 testers didn’t understand the controls of Article Swipe.

3/6 testers found that the bias meter design didn’t clearly tell them the political lean of an article.

6/6 testers didn’t understand what the percentages represented in our comment sections.

To solve these issues, we included instructional and informational panels that would provide ample details to inform users how to use this feature.

We also reworked the layout based on specific user feedback to create an experience more tailored to what users would want.

Disorganization of News Topics

Our next issue had to do with the disorganization of topics. Our For You and Discover page feeds were continuous and would display articles of varying topics in one flow, meaning there were no categories for specific topics.

Users explained that they were bothered by the idea of unrelated news being jumbled together, as it would be more difficult to find news under a certain topic. 5/6 testers thought that categories would help them consume news.

To solve this, we added specific topic headers within the For You and Discovery Pages. On the home page, this would be catered to a user’s specific algorithm, while on the Discovery page, all categories would be available to view.

Hierarchy and Readability Issues

Our last issue had to do with hierarchy and readability. Our primary typeface, Franklin Gothic, was uncomfortable to read for long periods of time and created an issue when used for article headlines. Users reported that Franklin Gothic was too stylized to read frequently. Another problem we had was with confusing iconography on the Shorts page. 6/6 users did not know what the thumbs up/thumbs down icon would do. Lastly, we introduced horizontal scrolling cards but 6/6 testers found them too narrow, making them uncomfortable to read.

To solve these problems, we used our secondary font, Roboto, instead of Franklin Gothic for headlines for a more neutral typeface that is easier to read. We swapped out the original icon on the shorts page for a simplified heart icon. And lastly, we adjusted the width of horizontal scrolling cards to be easier to read and less condensed.

Hi-fi Prototyping

Design System — Colors, Typography, Iconography, Components

For our finalized design system, we focused on a modernized, sleek aesthetic while preserving visual clarity. We achieved this by designing around a dark mode foundation and sticking to neutral colors for the majority of our components. To implement some personality, we chose purple as an accent color because it serves as the median between red and blue, which represent the two major political parties.

Our typography consists of Franklin Gothic Medium Condensed and Roboto for heading and body text, respectively. Both typefaces are relatively simple and maintain readability, with enough stylistic distinction to differentiate them.

Our iconography and key components adopt a cohesive principle of simplicity while utilizing the main color palette we established. We also implemented various gradients and glow effects to emphasize highlighted sections, and these techniques help elevate the modernistic aesthetic and create more personality.

High-Fidelity Prototype — Onboarding

The main purpose of the onboarding experience is to let users customize their account and preface the personalized aspect of the app. We achieved this by presenting users with prompts to select the news categories, political stance, and publishers that they want recommended. To ensure that users have full control over the curation of content, these prompts are accompanied by a “search” feature and a skip button. These elements allow the algorithm to seamlessly adapt to a user’s preference and lets them quickly get to discovering curated news content without having to extensively browse on the app.

High-Fidelity Prototype — Home

Home represents the central landing page that users will be initially introduced to, so it has to harbor our app’s main browsing tab in the form of For You. This tab presents curated news articles based on the user’s personalized taste, organized by categories. It also features Daily Digest, which synthesizes the most significant headlines based on the algorithm and displays them on one frame, simulating the front page of a physical newspaper. The other tab on Home is Top Stories, which doesn’t run based on an algorithm and simply presents trending headlines, with the top 10 most popular articles displayed first. Designing this tab on Home emphasizes the discovery of differing publishers and topics, such that we prevent an echo chamber of information consumption, which was a key concern voiced in our surveys and interviews.

High-Fidelity Prototype — Discover

The Discover page gives users access to a search bar to find any headline, publisher page, topic, etc., as well as presenting new categories of articles that they may have missed on Home. A common opinion from our surveys voiced that interactive elements like games or puzzles on news apps were effective for engagement and providing a break from constant reading, but disconnected from the general purpose of the platform. Therefore, we designed a dynamic feature called Article Swipe, which presents 10 random articles that users can swipe left on to skip, or right to save if they are interested in reading them. We also provided a brief instruction pop-up and info button at the top to explain how the feature functions, as it was unclear to our usability testers. After going through all the articles, users are given a results page that lists the headlines they saved. This design provides an interactive discovery feature that creates a break from otherwise endlessly scrolling through headlines.

High-Fidelity Prototype — Shorts

Shorts function very similarly to existing social media short-form video platforms, where users can vertically scroll through short video content, with the addition of swapping between the For You tab and the Top Stories tab. However, a key difference is that our shorts model is designed such that each video is connected to a particular article, meaning the video functions as an extension of the headline to hook users into reading the full story. Thus, each shorts entry links to the full-page article through the “Read the Article” button. The implementation of short videos demonstrates our attention to a younger target audience, who generally consume information through social media platforms, with short-form videos being the modern trend for information delivery.

High-Fidelity Prototype — Audio

The inclusion of audio-exclusive content reflects our desire to expand on accessibility, diversifying the medium of information delivery, and to align with the business standard of our competitors. The audio page matches a similar design principle to our homepage, with a split For You and Top Stories tab, and content is organized through categories. Playing an audio piece brings up a “Now Playing” bar at the bottom, which can be interacted with to full screen the content, allowing users to pause, play, skip forward or backward, etc. We also implemented a queue page that lets users set up multiple audio pieces to continuously run without manually having to play each one.

High-Fidelity Prototype — Article Page

Our extended article page prototype represents what an article would look like when fully opened. The unique top navigation bar offers various elements like save, the publisher name, which works as a button that links to the publisher page, a toggleable highlight feature, and an additional drop-down menu. Other features allow the user to navigate to the attached short video (if supported), as well as enable text-to-speech. We also implemented a bias meter that indicates how the article politically leans. Enabling highlights displays annotations throughout the article, with each highlight representing a section that users directly commented on. Thus, interacting with a highlight will pull up a tab of the associated comments, automatically sorted by most liked. A general comments tab can also be found at the end of an article, which combines both highlighted and non-highlighted entries. Every comment is given a trust score if it is making a factual statement, and users can directly attach source links to back up any claims. Highlighted comments have a corresponding symbol, which functions as a button that scrolls users to the highlighted section that is being referred to in the article. A common criticism from usability testing was the sense of overwhelmingness given the density of features and the lack of explanations, so we added info buttons to our more elaborate features like the political bias meter and the comment section. After reading through an article, users can browse alternative viewpoints with different political stances to see how other publishers are reporting on the topic, as well as a recommended list of new headlines that don’t relate to the current article.

Presentation Day

On Presentation Day, we presented Parse to a panel of three judges and received valuable, insightful feedback on our visual design, prototyping, and creative problem solving. We’re grateful to the judges for their comments on our work, as their feedback is essential to further improvement and growth of our product.

Challenges

Some challenges we encountered in designing our product included visualizing information flow in a concise, manageable way to create a seamless experience free of overwhelm for our users. We were also eager to solve many different user pain points, and thus managing the amount of features in our design, especially under time constraints, was also a challenge.

Next Steps

For our next steps of the project, we’d love to add more features that would further manage the flow of information, such as filters and options for profile customization. Additionally, we’d like to conduct further user testing on our high-fidelity prototype for more insight into personalization preferences that would assist in enhancing our product. To make our product more accessible to users, we would also work on designing a website view.