You sent review requests on Friday and saw no new reviews the following week. The tempting response is to send a reminder. First, take three actual customer records and follow the request from completed service to the customer's phone. You may find that the message never left the queue, arrived at an old address, or opened the wrong Business Profile.
Work through this diagnostic if you already ask customers. For a new request routine, see the four-week launch plan. The review link guide covers link creation.
Follow one request through five checkpoints
- Eligible: Was this a real customer you were permitted to contact?
- Recorded: Did someone actually log the completed service and the customer contact?
- Sent: Is there a sent event, or did the request stay in a draft or queue?
- Delivered and usable: Did the email arrive, and does its link open the correct review form on a phone?
- Outcome: Did the customer choose to review, reply privately, or take no visible action?
Trace three recent customers through the checkpoints. A scheduled timestamp proves only that something was queued; a sent timestamp does not prove it reached an inbox. For a QR card, record that it was offered and scan one of the printed cards yourself. Fix the earliest gap you can verify.
Choose records from different days or staff members. Three requests from the same imported batch may all share one technical fault; a mixed sample is more likely to expose an inconsistent handoff. Record where each trail stops rather than marking every unanswered message as a failed request. A customer can receive a perfectly clear invitation and simply decide not to review.
If customers never get the request
Compare completed jobs or appointments with customer records. If ten jobs closed and only six customers entered the request system, message wording is not your first problem. Find who owns the four missing records. For the customers already present, inspect invalid addresses, opt-outs, queued or failed messages, duplicates and any sending limit in your tool.
For automated email, verify what starts the timer. In TrustMint, the delay starts when an eligible customer is added or imported, not from a separately recorded job-completion timestamp. A late import can therefore produce a late request. The automation setup guide includes a handoff card you can adapt.
If the message arrives but the path breaks
Send yourself the exact message a customer sees. Open it on a phone, tap the link, and check the business name and location. For a multi-location business, a link to headquarters is not good enough. Google's Business Profile instructions explain how to get a review link or QR code. You can also create and test one in the free link generator.
Remove avoidable steps. A request with several buttons, a survey before the review link, or a demand to log in to your own website gives the customer extra decisions. Keep the business name, a short honest invitation, and one clear review link. Google's own account requirements for posting are outside your control, so do not promise that every recipient can review in one tap.
If one branch gets reviews and another does not, test both links before rewriting either message. Branch mix also matters: a location serving occasional emergency customers may get a different response from one with repeat appointment clients. Compare the journey and the people involved before treating the branches' public counts as a contest.
If delivery and link both work
Now inspect timing and wording. Was the request sent before the customer had used the product, or weeks after a simple visit? Test one sensible window against another for the same service type. Change only one thing at a time. The request templates give short examples, but no subject line can guarantee a response.
Keep the same invitation available to eligible customers whatever you expect them to say. Screening out likely critics would turn a delivery investigation into a biased request process; Google's review policy prohibits selective positive-review requests. Handle any service complaint on its merits.
Decide whether a reminder would help
A reminder is useful only when the first request was delivered, enough time has passed for the customer to use the service, and a further contact is appropriate under your contact rules. Do not send one because the public count did not move in a few days. You may not know whether a particular customer already reviewed under a different display name; a repeated “you have not reviewed us” message can be both wrong and irritating.
If you do send a single follow-up, make it easy to ignore: “In case the link was lost, here is the Google review page for {Business name}. Sharing your experience is entirely up to you.” Record the send date and stop there. Test the link and the handoff first; a reminder sent to a bad address or a broken page only repeats the original failure.
Judge the process with the right numbers
Each week, record eligible customers, requests sent, confirmed deliveries where available, and new public reviews. Treat these as separate counts. A public review may arrive later, may come from someone you never messaged, or may not be identifiable as a specific recipient. You cannot calculate a precise per-customer conversion rate from aggregate public reviews alone.
Once the first failure is fixed, test another small set of requests. If the link was wrong, replace it and test the new one on a phone. A reminder makes sense only after you know the original request arrived and your contact rules allow it.