Behavioral Decision Research — Red then Black (RtB)

Designing and analyzing user decision experiments to surface product signals for an alpha-stage fintech platform.

Behavioral Research Product Analytics User Research Qualitative Analysis Fintech

Overview

Red then Black (RtB) is an early-stage fintech startup building a transparent, simulation-based personal finance decision platform — helping users analyze major financial decisions (career moves, home purchases, retirement timing) using academic-grade methodology rather than rules of thumb or black-box advisors.

As Financial Research Analyst (Jan 2026–Present), I design and analyze behavioral research sessions studying how users engage with the platform's decision features under uncertainty. The work sits at the intersection of user research, behavioral economics, and early-stage product analytics.

Research Design

I designed and conducted 9 one-on-one user behavior interview sessions to study how users engage with financial decision features. Sessions were structured around:

Interactive decision tasks and uncertainty scenarios
Multi-round calibration exercises measuring confidence vs. accuracy alignment
Structured Q&A probing probabilistic reasoning and decision strategy

Each session was followed by a structured findings report delivered to the founding team, directly informing Alpha product iteration.

Analysis Methods

Transcript-Level Semantic Analysis

I analyzed session transcripts using multi-round clustering to extract: recurring language patterns, emotional trigger moments, narrative framing cues, confidence volatility markers, and belief-action discrepancies. This transforms raw qualitative conversation into structured behavioral models.

Calibration Analysis

A standalone calibration study examined the relationship between user accuracy, confidence levels, and calibration error across rounds. Key findings (not for public disclosure) shaped the design of the platform's UX feedback mechanism — specifically how and when feedback is shown to users.

Decision Divergence Study (4 rounds)

A 4-round divergence analysis tracked where user decisions systematically diverged from expected outcomes. I categorized divergence types (model error, emotional interference, narrative bias, signal misclassification) and produced structured reports after each round with actionable feature enhancement recommendations.

Quantitative simulations of these behavioral patterns are documented in the Finance Concept Analysis project →

Product Signal Framework

A core output of the research is translating behavioral patterns into measurable product signals:

Confidence volatility — how much confidence fluctuates across rounds
Decision reversal frequency — rate of strategy changes between rounds
Calibration tightening under feedback — responsiveness to error signals
Escalation patterns after streaks — risk-seeking behavior following perceived patterns

These signals inform alpha-stage monitoring metrics, intervention timing, and UX feedback loop design.

Deliverables

14+ structured research reports (9 interview reports + 1 calibration report + 4 divergence round reports)
Product signal framework documentation
UX feedback mechanism design recommendations
Executive briefings for founder-level decision-making