Feedback Analysis for an activewear giant’s T-shirt: A Case Study

This marketplace giant is perhaps the best-known brand when it comes to activewear and shoes. Any organization of this magnitude is bound to get millions over millions of feedback by its customers across hundreds of different channels of sale. An in-depth feedback analysis can deliver real value to such organizations.

ParallelDots ran a conceptual study running a feedback analysis for such a gigantic brand. The analysis was done for the customer reviews on Amazon. This article discusses the results of the feedback analysis in the form of an in-depth case study.

The Challenges with Traditional Feedback Analysis

Traditional feedback analysis is vulnerable to human biases and errors. In addition to this, the company of this magnitude generates so many reviews that the task itself becomes extensive in terms of both time and cost. As a solution to the problem, ParallelDots ran a mock analysis on all the customer reviews for a T-Shirt by this activewear brand on Amazon.

We used our in-house tools to perform these tasks. Our standard text classification product is SmartReader.

About SmartReader

feedback analysis
A Glimpse

SmartReader is a simplified SaaS solution where you can do all this in one go! All you’ll have to do is upload your survey responses in Excel and wait for the AI to do its job. You can customize the themes, give your own keywords, test the results and tweak the project until your results are good.

Click here to schedule a demo of SmartReader on your data.

Overview of Results

feedback analysis

Key Insights Generated as Result of the Feedback Analysis

Our product successfully analyzed over 1,00,000 customer reviews. The analysis produced several valuable insights. SmartReader not only identified the major themes but also classified the reviews.

feedback analysis

Insight 1: What is the general sentiment of buyers in regards to the T-Shirt?

ParallelDots used its Sentiment Analysis API to track the primary 3 sentiments- Positive, Negative or Neutral expressed in verbatim customer reviews. The results of the analysis are shown below:

The feedback analysis revealed that the general sentiment towards the T-shirt is positive. Out of all the reviews, over 80% carried a positive sentiment while less than 3% showed a negative sentiment.

Insight 2: What is the underlying emotion that reviewers display in regards to the brand and the product in question?

ParallelDots’s Emotional recognition API detects whether a piece of text or feedback displays excitement, happiness, sadness, anger, boredom, or fear. With this information an organisation can create an effective retargetting strategy.

Upon further investigation, our study revealed that 53% of the reviews displayed excitement as well as happiness. These customers can be retargeted with deals and special discounts. It is also to be noted that 22% of the reviews displayed a sad emotion which points to a disappointed customer.

Insight 3: What are the key themes that customers talk about in their reviews and what themes can be considered as parameters of success or failure?

SmartReader upon through the reviews generated 7 key themes around which the reviews revolve. These themes are listed above.

The net sentiment score for each theme was calculated and this result brought to light key areas of improvement and satisfaction. The classification was done by the AI itself so the user can effectively sit back and relax while the machine performs its magic.

Insight 4: What is the individual sentiment towards each of these key parameters?

Based on the net sentiment scores around each of the themes we figured out the key areas of satisfaction as well as key areas that need improvement.

As you can see from the results above, the products fit had the highest positive sentiment attached to it while quality is the area that suffered and caused the maximum dissatisfaction.

Insight 5: What are the most frequently used negative keywords in order to gauge the key areas of improvement?

A negative keyword cloud was generated in order to exactly figure out the issue with Quality that caused the negative sentiment. Such an analysis lets your organization hear and then act on the “Voice of the Customer”. As was revealed by our research this brand can improve its overall quality by working on the thinness and stickiness of its product’s fabric.


ParallelDots’ AI-based solution has the power to echo the customers’ voice. SmartReader undertakes a very deep analysis and provides targeted results which would not be possible with conventional methods of survey analysis. In contrast to typical tracking studies, which are limited to simply measuring satisfaction levels, ParallelDots yields not only intelligence about the actual factors driving satisfaction (the “Whys”) but also quantifies the extent to which each factor actually influences satisfaction. An organization like the one under analysis can use the insights generated by ParallelDots to improve their product’s quality in an efficient manner and create an effective retargeting strategy.

We hope you liked the article. Please Sign Up for a free ParallelDots account to start your AI journey now. Create your own classification model with SmartReader and discover interesting insights from your customers’ voice. You can also check out free demos of ParallelDots AI APIs here.



Follow Me

Leave a Reply