CRO / Reviews
How to Ask Customers for Reviews (Timing and Wording)
// the reviews you don't have are the ones you never asked for
By Fracto Solutions
October 26, 2026
The short answer
Ask every customer automatically, once the product has arrived and been used. Keep the request short, link straight to a one-tap rating, and strip out every extra step. On incentives the law is now specific: you may reward an honest review of any rating, never one that expresses a particular sentiment.
When is the best time to ask?
After the customer has received and used the product, not at the moment of purchase. For most products that means a few days after delivery; for items that reveal their quality over weeks, longer. We should be upfront that this is reasoning rather than measured data.
Ask too early and the customer has nothing to say; ask too late and the experience has faded. Anchor the request to delivery, then add enough time for genuine use. We looked for a published study on request timing against response rate and could not find one: the timing advice circulating online comes from review-platform help docs, which give sensible defaults rather than measured curves. Treat our default as a starting point and test it on your own orders. This request usually lives inside your post-purchase email sequence, and feeds the broader reviews strategy.
What is measured is that asking at all changes the outcome. Across 57,000 anonymous and 65,000 verified-buyer reviews, Spiegel found prompted, verified buyers averaged 4.34 stars against 3.89 for unprompted anonymous reviewers, and that a verified-buyer badge improved the odds of purchase by 15%. So the case for a systematic ask is solid; the case for day four specifically is not.
How should I word the request?
Briefly, personally, and with one clear action. Thank the customer, say why their review helps other shoppers, and give one obvious link. Keep the wording neutral about what you want them to say, which matters legally as well as ethically.
Do
Avoid
Short, warm, one clear ask
Long message with several requests
Explain it helps other shoppers
Any hint that you want five stars
One-tap link straight to the review
Making them log in or hunt
Name the product they bought
Generic, impersonal template
Keep the ask neutral on rating
"Loved it? Leave a review!"
The last row is not just tone. A request that frames the ask around a positive experience edges toward conditioning a review on sentiment.
How do I remove friction?
Make leaving a review as close to one tap as possible. Link directly to the review form for the specific product, let people set a star rating inline, keep text optional, and do not force an account. Each step you remove lifts your response rate.
Friction is where review requests quietly die. If the customer has to log in, find the product, and fill a long form, most will not bother. Let them tap a star straight from the email, make written text optional, and keep the whole thing mobile-first. Encouraging a photo is worth doing because it doubles as user-generated content, though we should note the widely quoted figures on photo reviews converting better all come from review vendors with no published method, so we treat that as a reasonable bet rather than a known result.
What are the rules on incentives?
Clear, and stricter than the old “do not buy five-star reviews” framing suggests. Since 21 October 2024 this is US federal law under 16 CFR Part 465, with civil penalties of $53,088 per violation. This is general information, not legal advice.
Section 465.4 prohibits giving anything of value “in exchange for, or conditioned expressly or by implication on” a consumer review expressing a particular sentiment. Read “or by implication” carefully, because it is what catches the soft versions: a discount advertised next to happy-customer imagery, a loyalty reward described as a thank-you for a great experience, or a contest entry promoted only to buyers who rated highly.
What remains allowed is rewarding the act of reviewing. A small discount or loyalty points for an honest review of any rating is fine, and you should say “any rating” in the offer itself so the neutrality is on the record. Two more parts of the rule bite here:
§465.5, insider reviews. Reviews from officers, managers, employees or their relatives need clear and conspicuous disclosure of the relationship. This is the usual shortcut for seeding a new product’s first few reviews, and it is the one most likely to go wrong.
§465.2, fake reviews. Creating, buying, selling or disseminating them where you knew or should have known, with AI-generated reviews explicitly named.
Outside the US: the EU has banned fake and unverified reviews under the Omnibus Directive since 28 May 2022 and requires you to disclose whether and how you verify that reviews come from real purchasers. The UK’s DMCC Act has banned them since 6 April 2025, with CMA fines up to 10% of global turnover. If you sell into either market, build the disclosure into the review widget rather than bolting it on later.
What about customers who don't respond?
A single gentle reminder is worth sending; beyond that, stop. Most reviews come from the first ask and one follow-up, and pestering customers damages the relationship for diminishing returns.
Set the request and one reminder, then let it go. The goal is a steady, automated stream from every order, not squeezing the last review out of a reluctant customer. Consistency across all orders beats intensity on a few, which is the same compounding logic behind good retention. As with timing, the one-reminder cap is our operating default rather than a measured optimum.
We automate the ask so reviews arrive from every order, on their own.
We set the post-purchase request at the moment that actually produces reviews, strip the friction down to a tap, keep the incentive honest, and cap the follow-ups, so your review count grows steadily without anyone chasing it manually.
It’s the same measurement-first approach behind our lifecycle and conversion optimization work, where we rebuild the buying path and measure the result .
// we would rather show up in your reporting than in your inbox
Frequently asked questions
When should I send a review request?
After the customer has received and used the product, usually a few days after delivery and longer for items that take time to judge. We could not find a published study measuring request timing against response rate, so treat this as a sensible default to test rather than a proven optimum.
How do I write a good review request?
Keep it short and personal: thank them, explain it helps other shoppers, and give one clear link to a one-tap rating. Keep the wording neutral about the rating you want, which matters legally as well as ethically under the FTC rule on reviews.
Should I give a discount for reviews?
Only for an honest review of any rating, and say ‘any rating’ in the offer. Under 16 CFR Part 465, effective 21 October 2024, giving anything of value conditioned expressly or by implication on a review expressing a particular sentiment is prohibited, with penalties of $53,088 per violation.
Can I ask my team or family to leave the first reviews?
Only with disclosure. Section 465.5 of the FTC rule requires clear and conspicuous disclosure of the material relationship when officers, managers, employees or their relatives review the business. Seeding a new product with undisclosed insider reviews is the most common way stores fall foul of the rule.
How many times should I follow up?
Once. Most reviews come from the first request and a single gentle reminder, and beyond that you annoy customers for diminishing returns. This is our operating default rather than a measured optimum, so cap the sequence and rely on consistency across all orders.
Sources
Review volume and purchase likelihood: Spiegel Research Center, Northwestern University, “How Online Reviews Influence Sales”, published June 2017. Purchase likelihood for a product with five reviews is 270% greater than for a product with no reviews; nearly all of the increase occurs within the first ten reviews and the first five drive the bulk of it. Conversion increase reached 190% for lower-priced products and 380% for higher-priced ones. Verified-buyer badges improved the odds of purchase by 15%, and verified buyers averaged 4.34 stars against 3.89 for anonymous reviewers. Samples: 111,460 SKUs across 22 categories at two lower-priced retailers, plus one high-end gift retailer (15.5 million page views, 1,800 products, 7.8 million users over a year), plus 57,000 anonymous and 65,000 verified-buyer reviews across about 13,500 products. Two caveats we keep visible: the data was supplied by PowerReviews, a review-software vendor, and the price-tier findings rest on a single specialty retailer. https://spiegel.medill.northwestern.edu/wp-content/uploads/sites/2/2021/04/Spiegel_Online-Review_eBook_Jun2017_FINAL.pdf
US law on reviews: FTC Trade Regulation Rule on the Use of Consumer Reviews and Testimonials, 16 CFR Part 465, effective 21 October 2024. It bans creating, buying, selling or disseminating fake reviews including AI-generated ones (§465.2); compensation “in exchange for, or conditioned expressly or by implication on” a review expressing a particular sentiment (§465.4); undisclosed officer, manager or relative reviews (§465.5); company-controlled sites presented as independent (§465.6); and review suppression, including representing that displayed reviews are most or all of those submitted when reviews are filtered by rating or sentiment (§465.7). Civil penalties run to $53,088 per violation, and the FTC began issuing warning letters under the rule in December 2025. https://www.ecfr.gov/current/title-16/chapter-I/subchapter-D/part-465
EU and UK law on reviews: the Omnibus Directive (EU) 2019/2161 added fake and unverified reviews to the blacklist in Annex I of the Unfair Commercial Practices Directive, in force since 28 May 2022, and requires traders to disclose whether and how they verify that reviews come from actual purchasers. In the UK the Digital Markets, Competition and Consumers Act 2024 banned fake reviews from 6 April 2025, imposes a positive duty to take reasonable steps to prevent and remove them, covers aggregate ratings, and lets the CMA fine up to 10% of global turnover without going to court. https://eur-lex.europa.eu/eli/dir/2019/2161/oj
Stated as Fracto’s position, not research: that the right moment to ask is a few days after delivery, that photo reviews are worth encouraging, and that a single reminder is the sensible cap. We found no published study on review-request timing against response rate, and no independent measurement that photo reviews convert better than text-only ones. The circulating figures on both come from review vendors with no stated method. These are our defaults from client work, offered as a starting point to test, not as evidence.