Myntra Private Label Sentiment Analysis Case Study | Actowiz Solutions

Executive Summary

The "Private Label Paradox" suggests that while consumers are drawn to store brands for price, they often return to national brands for perceived quality or status. Actowiz Solutions was commissioned to analyze this transition by scraping and processing ~60,000 reviews from Myntra's top-performing private labels. Using a combination of Aspect-Based Sentiment Analysis (ABSA) and Emotion Recognition, we pinpointed the exact "friction points" where consumers felt the private label failed to meet the standards of a national brand.

The Challenge: Managing "Review Fatigue" at Scale

Analyzing 100,000+ reviews manually is impossible; even a sample of 30,000 per category requires sophisticated data cleaning to remove "shallow reviews" (e.g., "Good product," "Nice") that offer no strategic value.

The Specific Goals included:

• Identifying "Switching Triggers": Why did a customer who previously gave a 5-star rating to Roadster suddenly give a 2-star rating to a new purchase?
• Gender-Based Divergence: Do men care more about durability (Roadster/Mast & Harbour) while women focus on fabric feel and fit accuracy (Anouk/Dressberry)?
• Brand-National Comparison: Extracting mentions of national brands within Myntra reviews to see which competitors are winning back the customer.

The Actowiz Solutions Process

Step 1: Intelligent Data Collection (Web Scraping)

Actowiz utilized custom scrapers to extract verified purchase reviews. Unlike standard scrapers, our tool captured:

• Reviewer History: (Where available) to identify repeat vs. one-time buyers.
• Metadata: Including images uploaded by users, which our AI analyzed for "Color Mismatch" or "Stitch Quality" issues.
• Product Attributes: Size, color, and price at the time of purchase.

Step 2: Data Cleaning & Deduplication

We filtered out non-informative reviews. Out of your 30,000 sample, we typically find that 15–20% are noise. Our cleaning process ensures you only pay for analysis on "high-substance" text.

Step 3: Sentiment & Emotion Analysis

We go beyond "Positive/Negative." We use Emotion Analysis to categorize the why:

• Frustration: Related to sizing inconsistencies.
• Disappointment: Related to fabric quality after the first wash.
• Joy/Surprise: Related to the value-for-money proposition (The "Trial Phase").

Conclusion: Turning Data into Strategy

Actowiz Solutions doesn't just provide a list of reviews; we provide the "Why." By identifying that sizing inconsistency is the #1 reason women switch from Dressberry to national brands, or fabric longevity is the pain point for Roadster, we empower you to give Myntra actionable advice on how to retain their private label customers.