PROJECT 01
Jul 2025 – Aug 2025
MICROSOFT MONETIZE · INVESTOPEDIA
Publisher
agents
An in-site agentic layer that keeps readers on the page, and makes ads feel useful instead of noisy.
AI-native Design
Interviews
Wireframing
Prototyping
Impact
$1.4M revenue
projected for client
ROLE
Product Designer
TIMELINE
Jul 2025 – Aug 2025
TEAM
Mentor, PM, Engineer, Data scientist
CLIENT
Investopedia
Cover.jpg
Agentic layer
Personalization
Anticipates what you need
Ad delivery
Tailored ads
Earned, not imposed
Personalization
Curated feed
Tailored to reading history
01 PROBLEM
Modern-day readers are frustrated by the
wall of irrelevant ads
and long articles
Publisher websites were still valuable, but the experience often felt cluttered and effort-heavy. Readers had to navigate dense pages, intrusive ads, pop-ups, and scattered recommendations before getting to what they actually needed.
Fig-01.jpg
A sample of Investopedia article and category pages surveyed for ad density and layout friction.
Fig-02.jpg
Screenshot of LLM chat bot screens like ChatGPT and Perplexity
That means fewer clicks, shorter sessions, and
falling revenue
for
publishers
For publishers, this created a direct tension. The more users relied on external AI summaries, the less time they spent on-site — reducing sessions, ad impressions, and monetization opportunities.
Impact.jpg
Business-impact diagram or session drop.
02 INTERVIEWS
from readers
They wanted content that felt
personal, easy
and
proactive
I interviewed 12 people who regularly consume content across lifestyle, finance, technology, travel, and food. The goal was to learn how they discover, read, trust, and use AI alongside publisher content.
Across interviews, readers wanted less effort and more relevance. They valued content that helped them decide quickly, surfaced useful next steps, and felt tailored without becoming overwhelming.
Research.jpg
12
Participants who read publisher sites at least a few times a week
5
Topics: fashion & lifestyle, cooking, technology, travel, and finance
100%
Had used AI tools in some capacity. Most of them use ChatGPT.
03 COMPETITIVE ANALYSIS
the system
AI-native platforms are becoming
more adaptive
Secondary research showed that AI-native platforms were already changing user expectations. Instead of making users browse static pages, these experiences responded to intent, context, and preferences in real time.
04 DIRECTION
The same website,
personalized
to each user
They wanted the website to understand what they were trying to do and help them get there faster.
So I explored an AI-native publishing experience where the site more responsive to user needs. Instead of treating every reader the same, it could adapt to intent, summarize long articles in their proficiency level, surface relevant context, and introduce ads when they are useful.
The same article or homepage no longer had to look the same for every reader.
Traditional.jpg
05 SOLUTION
click through it
Product
walkthrough
Penny learns intent, keeps the brand’s voice in the recirculation, and only introduces sponsored solutions when they help. Same site, different readers.
Feature 01
Flow-01.jpg
Summarizes the article so the users knows what they're getting into
Feature 02
Flow-02.jpg
Tailored and curated content, from customizing the homepage to making a special tailored reading list
Feature 03
Flow-03.jpg
A reading list and tailored insights from prior activity and conversation.
Feature 04
Flow-04.jpg
Sponsored solutions in context — ads that feel useful instead of disruptive.
06 OUTCOME
outcome.md
what changed
Penny created value across
the entire
publishing ecosystem
Penny turned AI into an in-site assistant that benefited every side. Readers got more relevant content and useful solutions, publishers gained engagement and commission opportunities, and advertisers reached users with better context.
Impact
$1.4M/yr revenue
projected for Investopedia
Ecosystem.jpg
07 EXPERIENCE
experience.md
what I learnt
Focusing on the
bigger picture
became just as important as the details
Prioritization
Tight timelines made it obvious: focus on the ideas that created the most impact, instead of trying to solve everything at once.
Feedback
Asking better questions in critiques and feedback sessions helped avoid assumptions and move forward with a clearer direction.
Keep exploring my work!
PROJECTS
Let's Talk
I'm most energized by projects where I can dig into complex problems, collaborate with smart people, and ship things that genuinely improve someone's day.
Reading
Watching
Love Island 🤪
Off Campus
Cooking
ratatoille
Avacado Toast
Biryani
Made with <3 by
Reet
Oberoi
Comment
Reet Oberoi
Open to full-time roles, and interesting conversations about hard design problems.
1