Most shops that ask for reviews at all ask at the same moment: the day the parcel is marked as delivered. It is the obvious trigger, because it is the one the system already knows about.
It is also, for a great many products, the wrong one. A customer who has had a mattress for four hours cannot tell you anything useful about it. A customer who has had it for three weeks can tell you the thing every other shopper wants to know.
The question worth asking is not when the parcel arrived. It is when this particular customer will know whether they are pleased.
Match the window to what you sell
Something used immediately. Food, drink, a book, anything consumed within a day or two. Two or three days after delivery is about right, while the experience is fresh.
Everyday goods. Clothing, tools, kitchen equipment. Give it a week to ten days, so they have worn it, washed it, or used it properly rather than just unboxed it.
Anything that reveals itself slowly. Mattresses, garden furniture, paint, skincare, anything where the honest verdict takes time. Three to four weeks, sometimes longer. You will get fewer responses and far better ones.
Consumables and refills. The most useful moment is when they are running low, because the request and the reorder prompt can sit in the same message and each makes the other more welcome.
A service or an installation. Within a day or two of the work being finished, while the relief of it being done is still there.
If you sell a mixture, the fix is not one global setting. Group your catalogue into two or three of the buckets above and set a window for each. Most platforms will let you do this, and if yours will not, doing it by hand for your best sellers still beats one timer for everything.
How many times to ask
Once, then once more, and then stop.
One request, and a single reminder a week or so later to the people who did not respond. A third attempt annoys more people than it converts, and it is the point at which a review request starts to feel like being chased.
Nothing else you send should be competing with it. If the review request lands in the same week as a promotional email and a delivery survey, it loses.
What to ask for
“Please leave us a review” is a chore. A question is a conversation.
Ask something specific and easy to answer. How did it fit. What did you use it for. Would you buy it again for the same job. The answers to those are exactly the sentences future customers want to read, and they are much easier to write than a review from a blank page.
Make it one tap. The link should land them on the review form itself, with the product already identified and the star rating clickable from inside the email if the platform allows it. Every step between the request and the finished review costs you responses.
Keep it short and sound like a person. A three-line email from a named human outperforms a designed template with a banner, particularly for small businesses, because the whole appeal of buying from you is that you are not a warehouse.
Ask for the review of the product, and the review of you, separately. A customer can love the item and have been let down by the delivery, or the reverse. Rolling both into one request produces reviews that are hard for the next shopper to interpret, and it hides which of the two you need to fix.
The rules worth knowing before you start
Two things you cannot do, and they matter more than they used to.
Do not offer anything in exchange for a positive review, and do not offer a reward that is conditional on the rating. An incentive for any honest review is on safer ground than one that buys a good one, but the safest position by a distance is not to pay for reviews at all.
Do not filter. Sending the request only to customers you expect to be happy, or holding back anything below four stars, is the practice regulators have taken a much dimmer view of in recent years. It also does not work: a page of unbroken five-star reviews reads as fake to most shoppers, and a handful of fours with a sensible complaint makes the rest believable.
Who owns the reviews once they exist, and what you are allowed to display where, is a subject of its own: who owns your reviews, and what you can do with them is worth reading before you build a page around them.
What to do when they arrive
Reply to them. All of them if you can, and certainly the critical ones. A calm, specific reply to a two-star review does more for the next shopper than any number of glowing ones, because it shows what happens when something goes wrong.
Do not argue, do not explain at length, and do not use the reply to sell. Acknowledge it, say what you have done, and offer to sort it out privately.
Then use them. Reviews are the raw material for better product pages, and the language customers use in them is usually better than anything you would write yourself. That is covered properly in the reviews you already have, and what they are telling you.
Where to start
Find out what your current trigger actually is. In a lot of shops nobody knows, because it was set during the build and has run untouched ever since.
Then pick your three best sellers, decide honestly how long it takes a customer to know whether they are pleased with each, and set the window to match. Leave it a couple of months and compare the response rate and the length of the reviews you get. Longer reviews are the better signal, because they mean people had something to say.
If you would rather have the whole review flow set up properly, from the timing to where the reviews end up on your product pages, get in touch, or read more about how we set up and look after an online shop.
Where these facts come from
The rules on incentives and filtering are set out in law and in the regulator’s own guidance, so here they are if you want to read them yourself.
- The legislation is the Digital Markets, Competition and Consumers Act 2024 (opens in a new tab).
- The Competition and Markets Authority’s fake reviews guidance (opens in a new tab) explains, in plain English, what it expects of businesses that collect and publish reviews.