Retail customer feedback
Analyse product reviews with a word cloud
A word cloud from 2,000 clothing reviews shows the language shoppers use when they discuss fit, size, fabric and comfort.
The source
2,000 unique review bodies
A fixed CC0 sample of unique review bodies from the Kaggle dataset, selected without ratings or product metadata.
Kaggle: Women's E-Commerce Clothing ReviewsHow we made it
We analysed the review text with standard English stop words, frequency scoring, and single-word terms. The cloud displays the 60 most frequent remaining terms.
Sources and preparation notesRead it carefully
Word size represents occurrence frequency. It does not measure sentiment, satisfaction, or the importance of a product issue.
What this cloud can show
- Fit, size and fabric are recurring topics, giving a quick starting point for product and merchandising teams.
- Words such as small, large, petite and waist suggest useful questions about sizing guidance and product descriptions.
- Positive words such as love, perfect and comfortable appear often, but need reading in context before they are treated as a finding.
Try this with your own text
- 1Create separate clouds for each product category, season or rating band.
- 2Add product names and generic retail terms to Stop Words when they obscure the comparison.
- 3Use the Summary and Text views to inspect the original reviews behind an interesting term.
Start with a question, then read the text.
Use a cloud to find patterns worth investigating, then inspect the underlying comments or papers.