Debatable — building a trust layer for online discourse
Back to Work
Product Case Study

Debatable

Building a Trust Layer for Online Discourse

Cornell Tech · Startup Studio Jan – May 2023 Product Management
57%
Bias Consumption
59%
Fake News Concerns
95%
Preferred Argument Maps
100%
Wanted Validity Gauge
HOW MIGHT WE

Create a platform that promotes fact-based, structured debates backed by verified news sources?

Problem

Online debates are broken. We set out to fix that.

57% of people consume biased content without realizing it. 59% are concerned about fake news. And the average person spends 150 minutes a day on platforms that reward outrage over insight. Twitter, Reddit, and Facebook debates are toxic, unstructured, and flooded with misinformation.

There was no platform organizing discourse around verified facts and diverse perspectives. The trust layer was missing entirely.

No structure

Online debates have no framework. Noise drowns out signal; important arguments get lost in unstructured comment threads with no hierarchy.

Unverified claims

Arguments go unchallenged and unsourced. Misinformation spreads unchecked because there's no mechanism to grade or verify what's being said.

Outrage economy

Platforms optimize for engagement, not insight. Rage-bait and partisan content surfaces over evidence-based reasoning, by design.

The What
Structured Debate
Organized format from claim to evidence to conclusion
Both Sides
Surfaces opposing viewpoints by design
Evidence Grading
Arguments ranked by source quality
Generalized Feeds
Cross-topic discovery, not algorithm-driven silos
Clutter Reduction
Filters noise, promotes substance
Fact Verification
Sources checked against verified outlets
The Why

No platform combines structured debate, evidence requirements, and balanced viewpoint promotion.

The Who
Observer
Curious what people think
Debater
Happy to argue any side
Fanatic
Here to promote their side
Fact-Checker
Wants evidence for every claim
Learner
Trying to understand and grow
Bad-Actors
Here to offend and derail
My Role

My role and contributions.

I conducted Market Research and Competitive Analysis, defined Product Positioning with news verification as the core differentiator, designed the Figma Prototype including the news repository feature, contributed to OKR and Roadmap Planning, and pitched the solution to industry experts for validation and feedback.

Team
  • Multi-disciplinary : Operations Research, Product Design, MBA/Product, Computer Science
  • Cross-institutional : Cornell Tech × Parsons School of Design
  • Studio format : Cornell Tech Startup Studio, Jan–May 2023
PM / PMM Skill Where Demonstrated Evidence
Strategy
Market Research Competitive analysis, user needs Mapped competitors across 3 platforms
Product Positioning News repo as differentiator Defined trust layer as core value prop
OKR Definition 4 objectives with key results SMART framework with measurable targets
Roadmap Planning Q1–Q2 product roadmap End-to-end from planning to launch
Research & Design
UX Design Figma prototyping 5-screen flow from mid-fi to hi-fi
Competitive Analysis Landscape mapping, 3 platforms Structured + verified gap identified
Validation & Execution
Risk Analysis 8 risk zones identified Mitigation strategies for each zone
Data-Driven Thinking Validation testing 95% preferred argument maps · 100% wanted validity gauge
Stakeholder Comms Industry expert pitches Presented and incorporated expert feedback
Cross-Functional Collab Multi-disciplinary team Design, product, engineering, research
Discovery

Three problems. Three solutions.

We mapped the core pain points in online discourse and designed targeted solutions for each. Every feature in the product traces directly back to a validated user problem.

Problem
Noise and lost information

Important arguments disappear into unstructured comment threads. There's no way to see the full landscape of a debate.

Solution
Forest View

A visual argument map that surfaces pro/con perspectives at a glance, letting users see the full shape of a debate before diving in.

Problem
Unsubstantiated claims

Arguments go unchallenged and unsourced. There's no way to tell a well-researched point from an opinion dressed as fact.

Solution
Evidence Grades

Every argument is graded on evidence quality, substance, and relevance, sourced from verified outlets, not anonymous posts.

Problem
Clutter and lack of structure

Debate threads are chaotic. There's no consistent format to follow an argument from claim to evidence to conclusion.

Solution
Organization

A structured debate format with clear sections, threading, and hierarchy, so users follow the argument, not the chaos.

Current-state journey map: mapping the online debate ecosystem

Current-state journey map: mapping the online debate ecosystem

Who else was in this space?

I mapped existing platforms against two dimensions: how structured their debate format was, and whether they verified the evidence behind arguments.

Kialo Edu offers structured debates but is limited to educational settings. Twitter, Reddit, and Facebook have debate features but promote echo chambers over evidence. None combined structured argumentation with verified news sources.

Our differentiator: a curated news repository serving as the trust layer, letting users access verified sources from NYT, CNN, Washington Post, and Reuters directly within the debate flow.
EVIDENCE VERIFIED →
STRUCTURED DEBATE →
LOW
HIGH
HIGH
LOW
THE GAP
Structured + Verified
debate platform
Twitter / X
Structured? No. Echo chambers.
Reddit
Community, no verification
Facebook
Outrage-optimized
Kialo Edu
Structured but education-only
News Outlets
Verified, but no debate layer
Social platforms
Debate tools
Media / verification
Opportunity
Strategy

From objectives to roadmap.

We defined 4 OKRs with measurable key results using a SMART framework, built a Q1-Q2 product roadmap from planning to launch, and conducted an 8-zone ethical risk assessment covering everything from misinformation to algorithmic bias and data monetization.

8 Risk Zones ethical and technical risk evaluation

8 Risk Zones: ethical and technical risk evaluation

Design

This is where the product came to life.

I designed the Figma prototype: a mobile-first experience that takes users from topic discovery through evidence-backed argumentation in five clear steps.

01
Choose Topic
Browse trending debates by category.
02
Read Description
Get framing context before diving in.
03
See Arguments
Binary pro/con layout at a glance.
04
Arguments View
Explore the argument map in depth.
05
View Evidence
Verify with graded news sources.

01. Home Feed — Trending topics across categories

Prototype screen 1 Prototype screen 2 Prototype screen 3 Prototype screen 4 Prototype screen 5 Prototype screen 1 Prototype screen 2 Prototype screen 3 Prototype screen 4 Prototype screen 5

These are select screens from the prototype. Explore the complete interaction flow on Figma →

Argument mapping that users actually wanted.

The arguments view uses a binary layout with color-coded pro/con cards, each tagged with evidence grades. Users can see the full landscape of an argument before diving deeper.

95% of test participants preferred this argument map over traditional threaded discussions.

The news repository: our trust layer.

Instead of letting debates devolve into "I read somewhere that...," we built a curated evidence view sourcing from LA Times, USA Today, CNN, Wall Street Journal, Reuters, Washington Post, and more. Each source is graded for relevance. Arguments backed by evidence, not by anonymous posts.

100% of test participants wanted a gauge for comment validity.

Metrics

What success looks like.

95%
Preferred argument maps over threaded discussions
100%
Wanted a validity gauge for comments
57%
Consume biased content unknowingly
59%
Concerned about fake news

We defined north-star metrics across engagement, satisfaction, discussion quality, and platform growth to measure whether Debatable was delivering real value — not just activity.