Product reviews

Find out what reviews say about each product and pick the ones you can quote on the product page.

Run this on your file Download the template CSV

Columns to upload

Name the columns in your CSV like this. Extra columns are fine: every column of the row is sent with the questions.

Column	What it holds
`product`	Product name
`rating`	Star rating, 1 to 5
`review`	Review text

Questions

Each question adds a column to your file. The answer is one of the listed values, or Unsure when the model's confidence is below the cutoff: 70% for pick-one and scale questions, 80% for yes or no.

New column	Question	Possible answers
`aspect`Pick one	What does the review mainly talk about?	taste or quality How good the product is fit or size Size, fit or dimensions shipping Delivery speed or condition on arrival price Price or value for money packaging Box, bottle, wrapping service Customer service
`sentiment`Scale	How does the reviewer feel about the product?	very negative negative neutral positive very positive Lowest to highest
`defect`Yes or no	Does the review report a defect, damage or a wrong item?	yes It reports a defect, damage or wrong item no It does not
`quotable`Yes or no	Could you quote this review on the product page: positive, about the product, specific, no profanity?	yes It is positive, specific and clean no It fails at least one of those

Example row

Column	Value
`product`	Domaine Ganevat Chardonnay 2022
`rating`	5
`review`	Salty, lemony, fresh. Opened it with oysters and the table went quiet. Buying two more.

Answers from a test run on this row

Column	Answer	Confidence
`aspect`	taste or quality	100%
`sentiment`	very positive	100%
`defect`	no	98%
`quotable`	yes	95%

More ecommerce use cases

Use case	What it does	New columns
Product catalog tagging	Fill in missing tags and filters on a product catalog from titles and descriptions.	`audience` `season` `giftable` `description_quality`
Return reasons	Code the free-text reasons on returns and find the ones better product pages could prevent.	`reason` `preventable` `warehouse_fault`
Order notes	Pull the order notes that need someone to act before the order ships.	`note_type` `needs_action` `date_given`
Competitor product matching	Check whether the competitor listings you matched to your products are the same item before you compare prices.	`match` `same_size` `bundle`
Product listing quality check	Audit product titles and descriptions for missing attributes and mismatches before a marketplace or Google Shopping feed goes live.	`title_complete` `category_fits` `promo_text` `description_matches`