The first reviews matter far more than the ones that follow: going from zero to five reviews changes the buying decision, while going from 200 to 250 changes nothing. Show the rating near the price, and reviews that mention sizing near the size selector.
The benchmark
Count your product pages with no reviews at all, and the share of traffic they get. That's where the gain is, not on your best sellers that are already well rated. Research from the Spiegel Research Center at Northwestern University points the same way: a product with five reviews is 270% more likely to be purchased than one with none, and most of the effect happens within the first ten.
The entry cost is low. At a premium shoe brand the authors worked with, no product page showed reviews or a rating that Google could read, on baskets of €500 to €850. The project fits in a day: six actual hours at one brand, nine estimated at another, for the same setup.
The rule
Ask every buyer for a review, after delivery and after some time using the product. Show everything, including average reviews, and respond to the bad ones.
A project manager on the team summed it up in a weekly meeting: "
You start a flow three days after the package arrives, and if they haven't left one, you follow up a week later."
She added a caveat about getting started:
"If people see there are only three, that hurts trust too. Usually, you collect reviews for a while and then you publish them."
The steps
- The trigger. Set the request to go out based on the delivery date, not the order date.
- The target. Start with products that have no reviews, five per product.
- The placement. Put the rating under the title and on the thumbnails on collection pages.
- The filters. Let people sort reviews by size or by use.
- The legal notice. State whether reviews are verified and how, show the date of each review, and give the reasons for rejection: the French Consumer Code has required this since 2018. And never claim that reviews come from buyers without having verified it.
A sentence to say to your leadership, in your own words:
"We ask for a review three days after delivery, we publish everything, we show the date and the verification method, and we live with a 4.5."
The trap
The trap is the perfect rating. A 5.0 across 300 reviews inspires less trust than a 4.6 with a few well-argued critical reviews and a reply from the brand: purchase likelihood peaks between 4.2 and 4.5 stars, then drops as it nears 5.0, again according to Spiegel.
It's the hardest sentence to get across internally, and we let the opposite reflex slide for a long time.
An executive described it to us at a project kickoff:
"We've never had negative reviews. What we often do is publish only the ones above a certain rating."
The reflex is understandable. Yet since 2022, the French Consumer Code has listed it among deceptive commercial practices. The other mistake is never reading your reviews: the objection that comes up ten times belongs on your product page.
