Smartphone Feature Optimization (Marketing Mix)

This synthetic but realistic dataset contains 90+ customer reviews for 6 smartphone models (from Apple, Samsung, and Google), along with:

Product specifications (Price, Screen Size, Battery, Camera, RAM, Storage, 5G, Water Resistance) Customer reviews (Star Ratings, Review Text, Verified Purchase Status) Sales data (Units Sold per Model) Potential Use Cases: ✅ Feature importance analysis (Which specs drive ratings?) ✅ Sentiment analysis (NLP on reviews) ✅ Pricing strategy optimization ✅ Market research (Comparing Apple vs. Samsung vs. Google)

Smartphone Customer Satisfaction Survey Objective: Understand how product features influence purchasing decisions and satisfaction.

Section 1: Demographic & Purchase Behavior Which smartphone brand did you purchase?

☐ Apple ☐ Samsung ☐ Google Maps to brand column. Which model did you purchase?

Apple: ☐ iPhone 14 | ☐ iPhone 15 Samsung: ☐ Galaxy S22 | ☐ Galaxy S23 Google: ☐ Pixel 7 | ☐ Pixel 8 Maps to model_name column. Where did you purchase the phone?

☐ Online (e.g., Amazon, Brand Website) ☐ Physical Store Justifies verified_purchase (assumed online = verified). Section 2: Product Feature Ratings How would you rate the following features? (1 = Poor, 5 = Excellent)

Battery Life: ⭐⭐⭐⭐⭐ Camera Quality: ⭐⭐⭐⭐⭐ Screen Size: ⭐⭐⭐⭐⭐ Performance (RAM/Processor): ⭐⭐⭐⭐⭐ Aggregates into star_rating (average of these). Which feature is MOST important to you?

☐ Battery Life ☐ Camera Quality ☐ Screen Size ☐ Performance ☐ Price Explains review_text keywords (e.g., "battery" mentions). Section 3: Price & Satisfaction How do you feel about the price of your phone?

☐ Very Affordable ☐ Fairly Priced ☐ Slightly Expensive ☐ Too Expensive Maps to price vs. star_rating correlation. Would you recommend this phone to others?

Column Details (Metadata)

Column Name (Type) Description "Example"**

model_id (Integer) Unique ID for each phone model 1 (iPhone 14)

brand (String) Manufacturer (Apple, Samsung, Google) "Apple"

model_name (String) Name of the phone model "iPhone 15"

price (Integer) Price in USD 999

screen_size (Float) Screen size in inches 6.1

battery (Integer) Battery capacity in mAh 4000

camera_main (String) Main camera resolution (MP) "48MP"

ram (Integer) RAM in GB 8

storage (Integer) Storage in GB 128

has_5g (Boolean) Whether the phone supports 5G TRUE

water_resistant (String) Water resistance rating (IP68 or None) "IP68"

units_sold (Integer) Estimated units sold (for market analysis) 15000

review_id (Integer) Unique ID for each review 1

user_name (String) Randomly generated reviewer name "John"

star_rating (Integer) Rating from 1 (worst) to 5 (best) 5

verified_purchase (Boolean) Whether the reviewer bought the product TRUE

review_date (Date) Date of the review (YYYY-MM-DD) "2023-05-10"

review_text (String) Simulated review text based on features & rating "The 48MP camera is amazing!"

Suggested Analysis Ideas to inspire data analysis: A. Feature Impact on Ratings Regression: star_rating ~ battery + camera_main + price Key drivers: Does battery life affect ratings more than camera quality?

B. Sentiment Analysis (NLP) Use tidytext (R) or NLTK (Python) to extract most-loved/hated features. Example: r library(tidytext) reviews_tidy <- final_data %>% unnest_tokens(word, review_text) reviews_tidy %>% count(word, sort = TRUE) %>% filter(n > 5)

C. Brand Comparison Apple vs. Samsung vs. Google: Which brand has higher average ratings? Price sensitivity: Do cheaper phones (e.g., Pixel) get better value ratings?

D. Sales vs. Features Correlation: units_sold ~ price + brand Premium segment analysis: Do iPhones sell more despite higher prices?

CC0

Original Data Source: Smartphone Feature Optimization (Marketing Mix)