The Complete Guide to Product Reviews

Key takeaways

• Products with five or more genuine reviews convert noticeably better than products with none.
• The FTC's 2024 rule on reviews and testimonials bans fabricated, AI-generated, and undisclosed-insider reviews, with penalties up to roughly $51,744 per violation.
• Ask for reviews right after a real result, not before, and never pay for a positive-only review.
• Review schema (AggregateRating and Review markup) is what lets star ratings show up in search results and AI answers.
• Never suppress or selectively hide negative reviews you solicited. It's both a trust failure and a compliance risk.

Why product reviews move the needle on sales

A shopper comparing two nearly identical products almost always picks the one with visible, specific, recent reviews over the one with none, even when the review count is modest. Reviews do the reassurance work a product description can’t, since they come from someone with no reason to oversell the product.

The effect isn’t uniform. A handful of detailed, specific reviews (mentioning fit, durability, a real use case) does more persuasive work than a large pile of generic five-star ratings with no text. Volume matters less than most sellers assume. Specificity matters more. A review that says “runs a half size small, true to the sizing chart on the fabric care page” earns more trust than fifty reviews that just say “great product.”

What the FTC actually requires now

The FTC’s Consumer Reviews and Testimonials Rule, 16 CFR Part 465, took effect in 2024 and gives the agency direct enforcement power against fake and manipulated reviews, with penalties that can run up to roughly $51,744 per violation. It bans a specific list of practices: reviews for products never actually used or received, reviews generated or substantially written by AI on a company’s behalf, reviews from insiders (employees, close relatives) not clearly disclosed as such, and buying only positive reviews while suppressing negative ones you solicited.

The rule doesn’t ban asking for reviews or running a review program. It bans faking the outcome. A store that asks every customer for honest feedback and publishes what comes back, good or bad, is exactly what the rule is designed to allow.

How to actually collect real reviews

Ask after a real result, not before or immediately at checkout. A customer who’s actually used the product for a few days has something honest to say. One asked to review a product they haven’t received yet has nothing to say but a guess.

• Automate a follow-up email or SMS timed to when the product typically arrives and gets used
• Make leaving a review genuinely easy: one click into a pre-filled form, not a multi-step login wall
• Ask a specific question or two (fit, durability, what surprised them) instead of just a star rating, since specific prompts produce more useful, detailed responses
• Send a single reminder if there’s no response, then stop asking so it doesn’t read as pressure

Incentives are allowed under the rule as long as they’re not conditioned on a positive review and are clearly disclosed if given. A small discount for leaving any review, positive or negative, is fine. A discount only if the review is five stars is not. Document the incentive terms in writing somewhere internal, so if a customer or regulator ever asks how the program works, the answer is already on record rather than reconstructed after the fact.

Where and how to display reviews for conversion

Star ratings belong near the price on both the category and the product page, since that’s where a comparing shopper’s eyes actually go before they click in. On the product page itself, surface a few of the most detailed reviews above the fold rather than burying all of them in a tab nobody opens.

Let shoppers filter and sort by rating, recency, and whether the review includes a photo. Photo reviews specifically carry outsized trust because they’re the hardest kind to fake convincingly, and shoppers know it. Don’t hide low ratings behind extra clicks. A visible 4.3 average with some critical reviews reads as more credible than a suspiciously perfect 5.0 with none. Shoppers have learned to be suspicious of a page with dozens of reviews and not one complaint anywhere on it.

Review schema markup: getting the stars to actually show

Star ratings in Google search results come from structured data, specifically AggregateRating and Review markup on the product page, added as JSON-LD. Without it, your reviews exist on the page but never appear in the search snippet, which means shoppers comparing search results never see them before clicking.

The rating markup has to reflect what’s genuinely visible on the page. Google has cracked down repeatedly on schema that claims a rating not actually shown to visitors, and mismatched schema is grounds for a manual action, not just a missed opportunity. This is also the same structured data AI answer engines pull when citing or comparing products, so getting it right pays off in more than one channel. Validate the markup with a structured-data testing tool after any theme or platform update, since a plugin change can silently break it.

Reviews vs. Testimonials vs. UGC: what the rules cover

The FTC rule covers reviews and testimonials broadly, not just star ratings on a product page. Social proof pulled from social media, video testimonials, and case studies all fall under the same disclosure and authenticity requirements. A glowing quote pulled from a customer’s tweet without permission or context is the same category of risk as a fabricated product review.

Treat every piece of customer-sourced content the same way: get real permission, don’t edit a quote to change its meaning, and disclose any relationship (employee, affiliate, paid partnership) that a reasonable reader would want to know about before trusting it.

Handling negative reviews without breaking the rules

A negative review is not a problem to hide. It’s a credibility asset if handled well, and a compliance risk if handled badly. Respond publicly, briefly, and without defensiveness, and offer a real fix through a private channel rather than arguing the point in the review thread.

What’s not allowed: deleting a negative review because it’s negative, threatening a customer for leaving one, or setting up a review flow that routes unhappy customers to a private form while only publishing the happy ones. That last pattern, sometimes called selective solicitation, is explicitly called out under the FTC rule as a form of review suppression. Train whoever handles the review inbox on this distinction directly, since it’s an easy rule to break by accident while just trying to be helpful.

Reviews and the rest of your ecommerce build

A strong review program doesn’t sit in isolation from the rest of the store. It connects to e-commerce fundamentals like page speed (a review widget that slows the page down works against the trust it’s supposed to build), product photography that matches what a review describes, and checkout flow, since a shopper convinced by reviews still needs a fast, simple path to actually buy.

Reviews are proof. They still need a page built well enough to let that proof do its job.

Product review platforms and how they're different

Most stores collect reviews through a dedicated app or platform (Yotpo, Judge.me, Okendo, or a native Shopify/WooCommerce plugin) rather than building a system from scratch. These handle the request emails, moderation queue, photo uploads, and schema markup automatically, which removes most of the technical work from getting a compliant program running.

Pick based on how the platform handles verified-purchase tagging (a real trust signal shoppers look for specifically), whether it supports photo and video reviews, and whether it exports clean AggregateRating schema without extra developer work. A platform that makes verification and disclosure easy is doing half the compliance work for you before a single review comes in.

More e-commerce guides