How to ask for a review without sounding desperate

The fear behind this question is real, and it is worth naming. Owners think asking makes them look like they are struggling. So they either never ask, or they overcorrect into a paragraph about how much reviews mean to a small family business, which is the version that actually looks desperate.

Customers do not experience a review request as a favor. They experience it as a small task. Your job is to make the task obvious, quick, and easy to decline.

Am I even allowed to ask?

On Google, yes, and Google says so in its own words.

Its review generation help page tells businesses to remind customers to leave reviews, and says you can ask customers to visit a Google link or scan a QR code. The Maps user generated content policy is more precise about the boundary. Merchants are permitted to solicit or encourage the posting of content that does represent a genuine experience, without offering incentives to do so or attempting to influence the rating or the contents of the review.

Three limits are named on the prohibited side of that same policy, and every one of them shows up in a script somebody sold to a restaurant last year.

No pressure on the premises. The policy says merchants should not require or pressure users to leave ratings or write reviews while on the premises. A staff member standing over a guest with a tablet is on the face of it pressure. A QR code printed on the receipt is not.

No requesting specific content. The policy names requesting that specific content be included, and it explicitly names asking for reviews that identify a staff member. So "mention that you came in for the deep clean" is out, and so is "mention me by name."

No staff quotas. Merchants requesting that staff solicit a certain number of reviews is enumerated as a violation.

The federal rule points the same direction from a different angle. The FTC's Rule on the Use of Consumer Reviews and Testimonials, 16 CFR Part 465, effective 21 October 2024, carves out a safe harbour at section 465.2(d) for reviews that resulted from a business making generalized solicitations to purchasers to post reviews about their experiences. Generalized is the operative word. Ask everyone, condition on nothing, script nothing, and you are inside the carve out.

What makes an ask sound desperate?

Length, apology, and repetition. In that order.

Length. Anything longer than two sentences reads as a pitch. The moment you start explaining how reviews help your ranking, you have moved the request from "small favor" to "please participate in my marketing."

Apology. "Sorry to bother you, I know you're busy, but if you have a second..." Every hedge in that sentence tells the customer this is an imposition. They will treat it as one.

Repetition. One reminder is acceptable. Three is a collections process. The customer who ignored the first two is not going to be moved by the third, and you have converted a happy customer into a slightly annoyed one.

The ask should cost the customer less attention than deciding what to tip.

There is also a tone problem specific to owners. When you write the request yourself, you know what it is worth to you, and that leaks into the wording. Write it, then cut it in half, then cut the part that explains why.

What does a good ask actually sound like?

Plain, specific, and attached to what just happened.

In person: "If you have a minute later, a Google review really helps us. No worries if not." Hand the card. Walk away. Do not stand there.

By text, after service: "Thanks for coming in tonight. If you have a minute, here's the link." Then the link.

By email, after a job: "Job's done and the site is clean. If you were happy with it, a Google review would help the next person deciding. Link's here."

Three things are doing the work. It references the actual visit, so it does not read as automated. It gives an exit, which removes the pressure that makes people avoid you. And it is short enough that the link is visible without scrolling on a phone.

What to leave out: your star rating goal, the word "please" more than once, anything about supporting local business, and any mention of what a review does for your rankings. That last one turns a personal request into a transaction the customer did not agree to.

Use the review link Google generates from your profile rather than a URL you built by hand. The generated link is tied to the profile's place ID and survives address changes, profile merges and rebrands. Hand assembled search links break silently, and a broken link produces a month of asks and no reviews. Google documents the procedure, including the detail that review QR codes can only be generated on a computer browser rather than on a phone.

When is the right time to ask?

While the experience is still current, and while the person who provided it is still in front of them.

The strongest ask comes from the person who did the work. The server who ran the table, the technician who finished the install, the hygienist who saw the patient. That request lands because there is a relationship, however brief.

The second strongest comes within a day, by text, referencing the visit specifically.

After a week, response rates fall and the reviews you do get are vague, because the customer no longer remembers the detail that made the visit good. Vague reviews are less useful to the next person reading them.

Do not ask during a problem. Do not ask while service recovery is still in progress. And do not ask on the way out the door if the customer is clearly rushing, because that is the version that feels like an ambush.

What conversion rate should I expect?

Nobody credible publishes one, and you should be suspicious of anyone who quotes you a number.

The figures in circulation come from the companies selling review software. Reputation states that 77% of consumers are willing to leave a review if requested but only 5% will do so otherwise. Swell reports roughly 10% of invites converting to a review, with case studies between 9% and 24%. None of those come with a disclosed independent methodology, and every one of them is published by a business whose product is the invite.

Treat vendor conversion figures as marketing, then go measure your own. Yours is the only number that can tell you how many asks a week you need, and it takes one month to produce.

Here is the arithmetic. A restaurant runs about 380 checks a month and prints the review QR code on every one of them, so the ask count is 380. Month one produces 22 new reviews. That is roughly one review for every seventeen asks, and it is your rate, not an industry rate.

Now work backwards from the target. Say the three competitors above you in the map pack are each collecting about five reviews a month, so your target is six. Six reviews at one per seventeen asks means about 102 asks a month, or 24 a week. You are already doing 380, so the ask is not your constraint and you should stop worrying about the script.

Run the same arithmetic with a weaker channel. If your rate is one review per forty asks, six reviews a month needs 240 asks, which is 55 a week and a genuinely different operating problem. A dental practice seeing 300 patients a month can hit that. A specialist seeing forty cannot, and that practice needs a better moment and a better channel rather than a bigger list.

The number that matters is asks per week, and it falls out of two figures you can measure yourself in thirty days.

Do the rules differ by platform?

They differ enough to matter, and getting it wrong wastes the reviews you earned.

Google permits asking, with limits on how you incentivize it. Yelp does not permit asking at all. Its own guidance for business owners tells merchants not to ask anyone to review the business, naming customers, mailing list subscribers, friends and family, and adds that staff should never compete to collect reviews. It also warns that proactively asking may hurt your Yelp rating, because Yelp's automated software may not recommend reviews that appear to have been prompted by the business.

That is a categorically different rule, not a stricter version of the same one, and what is normal practice on Google can work against you on Yelp.

Industry specific platforms have their own rules, and some marketplaces prohibit asking customers to leave the platform entirely. Read the terms once for every platform you actually care about, then write your script to the strictest one you use.

The safest general script asks for feedback without specifying a rating and without steering by sentiment. It is also the script least likely to sound rehearsed.

Where does the short ask stop working?

Two places, and neither is a wording problem.

When the customer has no account and no context. A one line text with a bare link, sent to someone who does not know what a Google review is, reads as a phishing attempt. Older customer bases and first generation immigrant customer bases in particular respond much better to a person handing them a card and saying the words. Shortness is a virtue in a channel the customer already trusts, not everywhere.

When the ask is short because it is automated and nobody owns it. A three word text from a number the customer does not recognise is short and also anonymous. The brevity that works is brevity from a person they just met. If your only channel is a bulk sender, the fix is the sender, not the sentence.

And there is a floor under all of this. Google's policy prohibits pressure and scripting because those distort the record. A short ask that nobody feels obligated to answer is not just more polite, it is the version that stays inside the rule.

What to do this week

Write your ask in one sentence. Read it out loud. If you would feel awkward saying it to a stranger, it is too long or too grateful.

Pick the single moment it happens and attach it to an existing step. Check drop. Invoice. Appointment confirmation. A separate task that somebody has to remember will not survive a busy week.

Test it on ten customers before you roll it out. You will hear the awkward part immediately.

Set the rule for reminders at one, and hold it.

Count asks and reviews for thirty days and compute your own rate. That single number replaces every conversion statistic anyone will ever quote you.

Then be patient with the outcome. Review flow is a compounding signal rather than a switch, which is part of the broader reality that local search results move on a timeline most owners find slower than they expected.

Questions about your review request wording?

Email me at eric@seod.com with the exact text your team says or sends right now. I will send back a shorter version, tell you which line is costing you responses, and flag anything that could get the reviews filtered. It takes me a few minutes and there is nothing attached to it.

If a review already came in badly worded and you are not sure how to answer it, the response structure is a separate skill worth having ready.

Otherwise keep reading the reviews and reputation library.