Most businesses that struggle to get reviews think they have a wording problem. They rewrite the message. They add a please. They try being funny.
They don't have a wording problem. They have a timing problem.
There's a window after a job where a customer feels good about you and will do you a favor without thinking hard about it. That window is short. Ask inside it and reviews come easily. Ask outside it and the best-written message in the world gets ignored, because the feeling that would have powered the review is gone.
The window is when the relief lands
Think about what a customer actually feels during a job.
Before: mild dread. Something is broken, or overdue, or expensive.
During: uncertainty. They're watching a stranger work in their space and hoping it goes fine.
At the end: relief. The thing works. It cost what you said. Nobody made a mess. The problem that was on their mind is off their mind.
That relief is the fuel for a review. It is the moment a customer feels most positively toward you they will ever feel. And it starts decaying immediately.
Two days later, the problem being fixed is just the normal state of the world. A week later they've forgotten your name. A month later, asking them for a review is asking a stranger for a favor.
So: ask at the relief.
The practical rule
For most service work, the best moment is within a few hours of the job being finished and confirmed good.
Not before the customer has seen the result. Not after they've moved on with their life. In that band where the fix is fresh and they're still a little grateful it's over.
If you take payment on-site, the ask goes right after payment, because payment is the emotional end of the transaction.
If you invoice later, the ask goes with the invoice being paid — not with the invoice being sent. Nobody feels warm about you while they're deciding to spend money. They feel warm after it's done.
The trigger events worth using
Set the ask on an event, not a date. Dates are guesses. Events are facts.
Job marked complete. The default. Works for most businesses.
Invoice paid. Better than "invoice sent." Paid means they were satisfied enough to release money.
A positive signal in a conversation. Someone texts "looks great, thank you!" — that is the customer volunteering that they're happy. That is the single best moment to ask, and almost nobody does, because it happens in a text thread and nothing is watching that thread. If you do nothing else after reading this, start asking every time someone thanks you.
A second purchase. Someone hiring you twice has told you more than any survey could. They will absolutely leave you a review.
A problem resolved well. Counterintuitive, but a customer who had an issue and watched you fix it fast is often a stronger advocate than one whose job was uneventful. They've seen you under pressure. Wait until the fix is confirmed, then ask.
When not to ask
Timing cuts both ways. There are moments where an ask does damage.
When anything is unresolved. If there's an open complaint, an unpaid balance in dispute, a callback scheduled, or a part on order, do not ask. You're inviting them to publish their frustration.
Immediately after you've asked for something else. If you just asked for a referral, or an upsell, or payment, don't stack a review request on top. One ask at a time. Stacking makes you a taker.
During the job. Asking while you're still standing in their house is pressure, and they know it. Customers who feel watched leave inflated reviews or none at all, and inflated reviews read as fake to everyone else.
On a bad day. If the job ran long, the price went up, or something went wrong that you couldn't fully fix — skip it. Fix the relationship first. The review can wait or never come. It's not worth the risk of prompting someone to write down how they feel while they still feel it.
How long you have, roughly
Nobody can hand you a universal number. But you can find yours.
For the next thirty jobs, log two things: how many hours after completion you asked, and whether you got the review. Bucket them — same day, next day, 2 to 3 days, a week or more.
You'll see the drop-off with your own eyes. It's usually steep and it's usually earlier than owners expect. Then you'll know where to put your trigger, and you'll know it from your own data instead of from a blog post.
That's a spreadsheet exercise. It costs you nothing but the discipline to record it.
One reminder, then let it go
If they don't act on the first ask, send exactly one reminder, two or three days later, and make it lighter than the first.
Then stop.
A third ask is nagging and it converts almost nobody. Worse, the person who caves and leaves a review because you wouldn't quit is not writing you a warm one. They're writing you three sentences of nothing, which does you no good, and they're remembering that you were annoying, which does you harm.
Ask twice. Move on. There will be another job.
Make the timing automatic
Here's why most businesses ask at the wrong time: the right time is a busy time.
You finish a job at 4:40, you're covered in whatever you're covered in, you've got one more stop, and you're not going to stand in the driveway crafting a review request. So you tell yourself you'll do it tonight. You don't. Two weeks later you remember, and by then the window is closed.
The fix isn't discipline. It's that the ask should not depend on you.
Mark the job complete in whatever system you already use. That's it — that's the only action a human takes. The message goes out a couple of hours later on its own, with the customer's name and the actual work in it. If they don't respond, one nudge follows in three days. If they replied to say something was wrong, the ask is suppressed entirely, because the last thing you want is an automated "how'd we do?" landing on someone who just told you they're unhappy.
That's a small piece of plumbing, and it turns a good intention into a number that goes up every month.
The compounding part
Reviews are one of the few things in a small business that compound. Each one makes the next customer slightly more likely to call, which makes the next review slightly easier to get.
Businesses with 200 reviews didn't run a review campaign. They asked every single customer, at the right moment, for years, without ever forgetting. The volume is a byproduct of consistency, not effort.
Consistency is what systems are for.
Start Monday: for one week, ask every customer within two hours of finishing. Just that. See what happens to your count.
If it works and you'd rather not think about it ever again, that's a short conversation.