AI + Power BI:
Turning Amazon Review Sentiment into Actionable Insights

Airtable

AI + Power BI:
Turning Amazon Review Sentiment into Actionable Insights

Experimented with Azure AI and Power BI to analyze review data, revealing product pain points and opportunities through real-time sentiment visualization.

Overview

Overview

Curious about how AI could support user research at scale, I ran a self-initiated experiment using Microsoft’s language AI model and Power BI to analyze sentiment in Amazon customer reviews.

Traditional user feedback analysis often relies on manual synthesis and small sample sets. To scale this process, I used AI to classify review sentiment (positive, neutral, negative, mixed) and visualized the results in Power BI—uncovering patterns in user emotion, product pain points, and recurring themes.

As a Product Designer, this project sharpened my skills in AI-powered research, data storytelling, and system-level thinking—equipping me to translate raw customer voice into actionable product insights.

A/B testing is essential for validating design decisions—but in B2B contexts, small sample sizes and tight timelines often limit statistical confidence.

To address this, I experimented with Microsoft Azure Machine Learning to build a predictive A/B testing pipeline that learns from historical test data. This approach surfaces actionable insights even with limited inputs, enabling faster, evidence-based decisions.

As a Product Designer, I built this ML-powered system using Azure ML Studio and the Boosted Decision Tree Regression algorithm. Trained on real-world e-commerce interaction data, the model compares UI variants and predicts conversion outcomes—empowering teams to prioritize high-performing designs without waiting on full test cycles.

A/B testing is essential for validating design decisions—but in B2B contexts, small sample sizes and tight timelines often limit statistical confidence.

To address this, I experimented with Microsoft Azure Machine Learning to build a predictive A/B testing pipeline that learns from historical test data. This approach surfaces actionable insights even with limited inputs, enabling faster, evidence-based decisions.

As a Product Designer, I built this ML-powered system using Azure ML Studio and the Boosted Decision Tree Regression algorithm. Trained on real-world e-commerce interaction data, the model compares UI variants and predicts conversion outcomes—empowering teams to prioritize high-performing designs without waiting on full test cycles.

Why this Matters

Why this Matters

This project shows how AI can elevate product design from surface-level UI to strategic decision-making—by transforming raw user feedback into scalable, insight-driven iteration. It demonstrates how design can uncover hidden friction points, guide data-informed prioritization, and drive cross-functional alignment that influences not just interfaces, but the direction of the product itself.

Highlights

Results

Results

The sentiment analysis shows a mix of reactions: 35% negative, 20% positive, mixed 30%, and 15% neutral.

Some opportunities and key actionable insights were:

  1. Improve Order Accuracy

  2. Address Side Fumbling in Encabulators.

  3. Enhance Customer Communication and Support.

Execution

Execution

  1. Created “Language” resource from Microsoft’s Azure AI resources.

  1. In Power BI, created a custom function by writing the below code.

  1. Invoked a newly created custom function and created a new column data that has sentiment score result.