Customer Reviews Insights: How to Analyze a Buyer’s Sentiments for Amazon Products
 

In the last couple of decades, the revolution has come because of the internet in the shopping experiences. Having nearly everything accessible online, consumers need to just open the app, click a few times and that’s it! The product will reach their doorsteps!

Though, with hundreds of alternatives accessible in all categories, this can be an intimidating task to search all the options and choose the best option. Product reviews always prove to be differential when comes to making definitive decisions.

On the other side, user review data is similarly important for businesses to observe as well as predict trends, study consumer attitudes as well as create pursued action plans.

Since 2012, X-Byte Enterprise Crawling is gathering product reviews across an extensive range of product categories and e-commerce platforms to provide go-to solutions for all the data scraping requirements.

Inspiration

Thanks to larger inventories, superior user experience as well as customer services, Amazon is always the leading retailer in a lot of countries. The majority of products are well-reviewed having many in tens of thousands.

Using the assistance of Machine Learning as well as Language Processing methods, we’ve examined lots and lots of Amazon user reviews. Our objective is to know consumer sentiments as well as help in improving decision-making.

Collecting Data

For the use case, we have collected reviews of the external monitoring from three top brands — Dell, LG, and HP— for Amazon’s US as well as UK’s web stores.

collecting-data.jpg

Our dataset has incorporated 12 products from the given three brands as well as the following data fields:

  • Product Details
  • Product Name
  • ASIN
  • Price
  • Brand
  • Reviews Details
  • Seller
  • Review Titles
  • Ratings
  • Complete Review Text
  • Review Date
  • Reviewer’s Name
  • Useful
  • Verified Purchase

After some rounds of the QA (deduplication, removing null records, etc.), our last dataset included 10,786 records from which the first review was given in December 2014.

reviews-over-time.jpg
Best Brands Sharing Similar Insights

Dell, LG, and HP are the three most reliable brands in the electronics and computer accessories industry. Because it is anticipated with the top brands, the highest ratings of 5.0 had maximum reviews of 7,168.

total-reiviews-by-ratings.jpg

After analyzing the chart above, we can observe that these brands got the overall number of reviews as well as for every rating.

review-by-brands.jpg
total-reiviews-by-ratings-break-down.jpg

These three brands are also having related average ratings for the monitors at Amazon.

average-rating-by-brands.jpg

Depending on numbers, you can assume that all these brands have roughly a similar market stake on Amazon as well as are also equally popular. Potential buyers might further need drill-downs about review content to make the last decision.

Analyze the Best Reviewed Products

HP VH240a is the 23.8-inch Full HD Monitor, which was the maximum-reviewed product in the dataset having 3,559 reviews. So, we have chosen to do individual analysis. This monitor model is also having average ratings of 4.27.

~100% Certified Purchases
verified-purchases.jpg

For the selected monitor, nearly all the product reviews were taken from the verified purchases to enforce the dataset quality for more analysis.

Quarterly Average Ratings over the Time

The initial review for the HP VH240a monitor was given in July 2017. This product has achieved the maximum quarterly average ratings of 4.40 in the Q4 of the year 2019.

 

The average ratings have been dropping since all-time highs that might be down for some factors like:

Better options either in the same competitor or a brand

New technologies in the hardware display panel with power management, picture quality, etc.

87% of the Customers are Happy!

With nearly two-third (2,352) reviewers providing that 5 stars, you can say that largely, customers were pleased with the purchase, and merely 12% of customers have given poor ratings of 1-2 stars.

happy-customers.jpg
Reviews at the Opposite Ends of Becoming More Useful

With Amazon, the registered can “like” the review to exhibit that they have found the detail useful. These “likes” counts are a very good indicator about the type of details that echoes with the potential customers as well as helps them in whether going with the products in question or discard them.

 

Depending on the review percentage, which was considered “helpful”, as given in the above chart, it is evident that the higher percentages of the users have found negative reviews to be the most useful during the buying procedure. This percentage was considered as:

Total reviews marked useful * 100 / Total reviews about those ratings

Although, while considering the real numbers of “likes”, reviews of two extravagances (ratings of 5.0 or 1.0) looked to be the most useful.

help-reviews-by-ratings.jpg
Length of the Reviews

Depending on our study, we have found that many users left reviews of comparable lengths for every rating — on average, about 350-400 words. This suggests that, despite the apparent product sentiments, customers like to be equally descriptive about the experiences in the reviews.

Having said it, there is a small increase in the lengths of average reviews with every rating.

Length of the Reviews

Speakers as well as audio quality control the customer opinions

good reviews

Depending on the words ‘cloud generated’ for good reviews, consumers are usually “happy” with the “quality” of the purchase, having “sound”, “screen”, “speakers”, and “picture” often mentioned. Consumers also like the monitors for “office”, “home”, and “gaming” objectives, as well as are ready to “suggest” it to the contacts.

 

In contrast, dissatisfied customers look to be getting “issues” with “power”, “sound”, and “cable” as well as might “return” it.

Having “speakers” featuring importantly in both the review groups is the feature, which is differentiating the customers.

It was only an indication of the type of actionable analysis possible with quality data, which X-Byte Enterprise Crawling offers. Analysts can do more detailed sentimental analysis for scraping extra insights from in-house or competitive review data that can be essential in making well-versed product development judgments.

About X-Byte Enterprise Crawling

X-Byte Enterprise Crawling is a data scraping company specializing in the web data scraping of data from nearly all business sectors. Contact us with your requirements, and we will surely provide the best solution to you!