Most review requests fail for a boring reason: they arrive at the wrong moment, ask for too much, and sound like a form. A good request is the opposite. It is timed to a feeling, short enough to answer from a phone, and personal enough to feel like a person sent it.
Ask at the moment of completed value
Timing is most of it. The right moment is when the client’s customer feels the value, not a week later when the feeling has faded. For home services, that is the hour the job is finished and the problem is gone, not the day the invoice clears. Ask while the relief is fresh and people say yes. Ask late and you are interrupting someone who has already moved on.
This is why the request should fire off a completed job, not a calendar. When the work is marked done, the ask goes out. The trigger is the event, not a batch someone remembers to send on Friday.
Short, personal, one ask
The message itself should do one thing. Not three. A greeting by name, one sentence that references the actual job, and a single link. That is the whole message. Every extra sentence lowers the odds it gets read, and every extra ask lowers the odds any of them happens.
“Hi Sarah, glad we got the hot water sorted today. If you have twenty seconds, a quick review really helps a small business like ours: [link]”
Notice what it is not. It is not a paragraph about how much reviews mean to the team. It is not a five-star graphic. It is a person, thanking a person, asking once.
The negative-feedback pattern, done honestly
This is where a lot of review tactics cross a line, so it is worth being precise. The legitimate goal is to catch an unhappy customer before they are stuck, and give them a fast way to be heard privately. The illegitimate goal is to stop a bad review from ever being public. The first is good service. The second is fraud, and it is not something we build.
The honest pattern works like this. Ask about the experience first, before you ask for a public review. If the answer is good, invite the public review. If the answer is not good, route that person to a private resolution path instead: a real reply, a callback, a chance to make it right. That is interception before posting, and it is legitimate because it helps the customer, not because it hides them.
What the pattern never does is gate, filter, or block a public review. If someone wants to post their experience in public, good or bad, that is their right, and nothing in the flow stops them. A tool that quietly drops the one-star reviews is not managing a reputation. It is faking one, and the fake falls apart the first time a customer compares notes.
“Asking an unhappy customer to talk before they post is good service. Stopping them from posting is a lie. Only the first one is worth building.”
Respect the opt-out, always
One rule sits underneath all of it. When someone asks not to be contacted, that request is honored everywhere, immediately. A review ask is a message, and the same consent that governs every other message governs this one. Someone who has opted out does not get a review request, no matter how happy they were with the job. The opt-out is not a setting to talk your way around. It is the floor.
Put those together and review requests stop feeling desperate. They go out at the right moment, in one honest line, to people who have not asked to be left alone, and the unhappy ones get a private door before a public one. That is a system a client can trust with their name, which is the only kind worth running under yours.