When to Ask for a Review (Timing That Actually Works)

Two businesses send the same review request, with the same wording, to the same kind of customer. One collects a trickle. The other collects several a week. The difference is usually not the message. It's the hour it arrived.

Timing is the single largest controllable variable in review collection, and it's the one owners tune last, after months of rewriting the text. This guide is what we'd tell an owner to fix first.

The gratitude window

Every completed job creates a short period where the customer feels actively grateful. Not satisfied, grateful. Satisfaction is the flat baseline that follows; gratitude is the spike, and only the spike produces reviews.

The spike has three properties worth planning around.

It peaks at completion. The moment the AC blows cold, the hair looks right, the drain runs clear, the emotional payoff is at maximum. Everything after that is decay.

It decays fast, at different speeds by industry. An emergency repair at 11pm on a holiday weekend generates gratitude that survives days. A routine oil change generates gratitude that's gone before the customer merges onto the highway.

It doesn't return. Nothing in a follow-up sequence rebuilds it. If you ask a week late, you're not asking a grateful customer, you're asking a busy stranger who once hired you, and the response rate reflects that.

This is why timing beats copy. Good words sent into a decayed window lose to mediocre words sent into a live one.

The good news is how willing people are when you catch them right. BrightLocal's 2026 Local Consumer Review Survey found that 83% of people who were asked to leave a review went on to leave one, and 28% say they will "always" write a review when asked. The constraint on your review count is almost never customer willingness. It's whether the ask arrived while they still cared.

Same day or next morning, and how to choose

Nearly every business lands in one of two correct answers: same day, an hour or two after completion, or next morning around mid-morning.

Same day, one to two hours after completion, when the result is immediate and visible. The customer can see what you did, they're still in it, and the memory needs no prompting. Restaurants, salons, auto service, emergency repairs, anything where the customer experienced the outcome in person and then left.

Why an hour or two rather than instantly? Because a request that arrives while the technician is still packing the van makes the customer feel watched, and because they're often still driving, paying, or wrangling kids. An hour of space converts better than zero.

Next morning, mid-morning, when the job ended late, when the customer wasn't present, or when the result needs a night to be appreciated. A furnace repaired at 6pm proves itself overnight. A house cleaned while the client was at work gets appreciated when they walk in, not when your team locks up.

There's also a plain attention reason. A request landing at 8:30pm competes with dinner, kids, and television. The same request at 9:30am lands during the first coffee, which is the most reliable window a small business gets.

When in doubt, next morning. Same-day only outperforms it when the result was visible and the customer was present.

Timing by industry

The pattern underneath the table: ask when the customer is enjoying the result, not when you finished the work. Those are the same moment for a haircut and 48 hours apart for a bathroom remodel.

Quiet hours are not optional

There is one timing rule that outranks all the others: never text a customer at a rude hour. A review request at 10:40pm doesn't just get ignored, it produces active irritation at the business that sent it, and irritation is the one emotion you're not trying to collect.

ReputeLoop enforces this at the system level rather than trusting the schedule. Texts only go out between 8am and 9pm in the customer's local time zone. If a job closes at 11pm, the request doesn't fire into the night; it queues and goes out the next morning at your configured send time.

That constraint quietly fixes a whole category of problems for businesses that run late, dispatch across time zones, or finish jobs at unpredictable hours. Your late-night emergency call gets a 9am request, which is exactly when you'd want it anyway.

Pick your morning send time deliberately, too. Mid-morning, somewhere in the 9 to 11 range, tends to beat 8am sharp. Eight o'clock competes with the commute and the start of everyone's workday. Ten o'clock lands in a lull.

One reminder, three days later. Not two.

Most people who intend to leave you a review don't. They read the text at a red light, think "yes, I should," and never return to it. That's not rejection, it's friction, and one reminder solves a good chunk of it.

ReputeLoop sends exactly one automatic reminder roughly three days after the original request, to people who haven't responded. Three days is the useful spacing: long enough that it doesn't feel like nagging, short enough that the customer still remembers the job.

A second reminder is where it turns. The customer who ignored two messages has made a decision, and message three converts almost nobody while irritating someone who was previously neutral about you.

If you want the reminder to work harder, make it shorter than the original rather than longer. A one-line "just a quick nudge, here's the link if you have a minute" outperforms a restated pitch.

The cooldown protects your best customers

Timing isn't only about a single job. It's about how often a given person hears from you across a year.

Your repeat customers are the ones most likely to leave a good review and the ones most easily annoyed by repetition. A client on a six-week color schedule, a monthly pest control account, a restaurant regular: every one of them completes jobs on your calendar constantly, and a naive system will ask them every single time.

ReputeLoop enforces a per-contact cooldown of 30 days. Once someone has been asked, they won't be asked again for at least a month, no matter how many jobs close in between. It's a guardrail against your own automation, so you never have to keep a mental list of who was asked when.

For most service businesses, one request per customer per year is plenty anyway. The cooldown is the floor, not the target.

Seasonal and situational timing

A few situations change the answer.

During a heat wave, freeze, or storm surge, your customers are stressed and your inbox is chaos, but the gratitude spike is enormous. The customer whose AC you restored on the third 100-degree day is more grateful than any customer you'll have in October. Keep asking. Just be certain the job is actually finished, because a request that arrives while a follow-up part is still on order reads as tone-deaf.

Right after a price increase, hold the ask for customers who just absorbed the new rate for the first time. Give the value a beat to land.

Around holidays, mid-December through New Year's is a low-response stretch for everything except food and retail. Requests sent then don't hurt you, they just underperform, so don't judge your system by a December sample.

After a service recovery, when something went wrong and you fixed it, wait. Ask a few days after the resolution is confirmed and holding, not the hour you finished the redo. Recovered customers can become your most enthusiastic reviewers, but only after they're sure it stayed fixed.

Email and text run on different clocks

The two channels behave differently and shouldn't share a schedule.

Text is immediate. It gets read within minutes, so it should be sent when you want it read. That makes SMS the right channel for the same-day ask, and it's why quiet hours matter so much: there's no such thing as a text that waits politely on its own.

Email is patient. It sits in the inbox until the person gets there, which makes the exact send time less critical and early morning perfectly fine. Email is the better channel for the job that finished at 9pm, for customers who read at a desk, and for anyone who never gave you a mobile number.

Running both is the practical answer: text where you have a mobile number and consent, email everywhere else.

What to fix first

If your review flow is underperforming, work in this order.

1. 1Move the send time into the gratitude window for your industry, using the table above.
2. 2Make sure requests never fire outside 8am to 9pm local, and that overnight jobs roll to a mid-morning send.
3. 3Confirm exactly one reminder is going out around day three, and that there isn't a second.
4. 4Check that repeat customers are protected by a 30-day cooldown.
5. 5Only then start rewriting the message.

Most owners do that list in reverse and wonder why the sixth version of the text isn't converting.

Two things pair with timing. Send people to a direct review link rather than making them search, which you can generate free with a QR code at reputeloop.com/free-tools/review-link, with the full walkthrough at reputeloop.com/blog/google-review-link-guide. And know where the line is on what you're allowed to ask for: reputeloop.com/blog/google-review-policy-guide.

The last thing worth remembering is why silence is so misleading. Research by customer experience analyst Esteban Kolsky found that only 1 in 26 unhappy customers actually complains; the rest simply leave. The same asymmetry runs the other way, since happy customers almost never think to write a review unprompted. Both groups are waiting to be asked, and the only real question is whether you reach them while they still feel something.