How to Automate Google Review Requests After a Job or Appointment

Review requests often fail at the handoff. The electrician finishes the job, the receptionist closes the appointment, or the cleaner leaves the property. Everyone means to ask for a review. By the next morning, nobody remembers which customers were asked.

The software can send the message; the business still has to tell it whom to contact and when. Start by following one completed job into the customer record. If that path is vague, automate it later. For a one-off message, the request templates are enough.

The six decisions behind an automated request

1. Choose the trigger. Decide what your team records when a job or appointment is complete. It might be a customer added to your review tool after checkout, a daily CSV export, or a manual addition by the person who closed the job. “We'll remember later” is not a trigger.
2. Define who is eligible. Use real customers who may lawfully receive the message. Keep contact permissions and opt-outs with the customer record. Do not filter for people you think will leave a high rating.
3. Pick a channel you can actually run. Email is easy to schedule once you have a valid address. QR cards work for walk-in traffic. A personal WhatsApp or Viber message can work when that is already how you talk with the customer, but it still needs someone to send it.
4. Set the delay. A same-day request can suit a simple visit; a more involved service may need time for the customer to experience the result. Pick an initial delay, then change it based on your own outcomes rather than a universal “best time.”
5. Write one clear ask. Name your business, thank the customer, invite an honest review, and include one working link. Avoid a long marketing email or a request for five stars.
6. Prevent repeats and measure. Track who has already been contacted, respect unsubscribes, and review how many requests were sent, delivered, and followed by public reviews or private feedback.

Write the handoff on one card

Before you turn on an automatic send, fill in this short operating card. If a field is blank, your team will make a different choice each day.

Consider a repair finished on Monday and entered into the tool on Thursday. With a two-day delay measured from the customer record, the email becomes eligible on Saturday, five days after the repair. Moving a slider from two days to one will not repair a three-day import lag. Check the trigger your software actually uses and arrange the handoff around it.

Make the customer list safe to send

A useful import has fewer columns than most customer databases. For a first run, prepare an internal customer reference, the email address, the relevant location and the date you added the customer. Keep evidence of contact permission and opt-outs in the source system. The sending tool also needs a way to recognise someone it has already contacted, whether that is an existing customer record or a previous request record.

Before uploading a batch, ask the person who exported it to check three cases by hand: a repeat customer, a customer who opted out, and a customer whose job is still open. A repeat should not receive two invitations for the same visit. An opt-out should never be reintroduced because a fresh CSV created a new row. An unfinished job should stay out of a completion-based send. If the export cannot reliably distinguish those cases, repair the source list before increasing volume.

TrustMint's scheduled email delay is measured from when an eligible customer record is created in the tool. It does not watch your job-management system for a completion event. Its available delay settings are one, two, three, five or seven days. Pick the delay with your actual import timing in mind, and check the queued and sent records after the first batch.

Check the message and the audience together

Write a short invitation with the business name and one tested review link. Send it to eligible customers without screening for likely praise. Discounts or gifts in exchange for reviews, and lists limited to likely five-star reviewers, violate Google's policy. Keep a route for someone to raise a service concern, while leaving the public review option open. Check the contact rules that apply where you operate before importing a list.

Test with a small batch before you switch it on

1. Send a test to yourself. Check the sender name, subject, mobile layout, unsubscribe link, and where the review link opens.
2. Use a small group of recent, eligible customers. Confirm that no one is included twice and that a previous opt-out is excluded.
3. Ask the person who closes each job to show you the exact handoff. If a customer never enters the system, automation cannot reach them.
4. After a week, compare eligible customers with requests sent. Investigate gaps before deciding the message needs different wording.

Keep the first check concrete. If 25 jobs closed, 21 customers were eligible, 18 reached the import and 16 requests were sent, the immediate investigation is the five eligible customers missing from the import or send. It is too early to judge the wording by the number of public reviews. Check whether the missing records were held back by permission, plan limits, duplicate detection or a late handoff, and record the answer next to the process card.

Decide who handles exceptions

An automatic queue still needs an owner. Give that person a short weekly list: customers with no usable email address, messages that bounced, records held because of an opt-out, and completed jobs that never reached the tool. An opt-out is not a data-cleanup problem to work around; keep it excluded. A bounced address needs correction in the customer system before a later send is considered. A job missing from the import points back to the closeout process.

For the first month, note the date each batch entered the tool and the date requests were actually sent. That makes a weekend or holiday delay visible. It also prevents the team from mistaking “queued” for “delivered.” Once the handoff and exception list are stable, the process can run with a brief check instead of daily rescue work.

Keep a manual path for exceptions. A customer who prefers a personal conversation may appreciate one short message in the channel you already use together. Do not promise an automated channel your tool does not provide.

Check what your tool actually sends

In TrustMint, automatic requests are emails after you add or import customers and enable sending. The system checks contact restrictions and existing request records before queuing a message. WhatsApp and Viber are manual shares from your phone; SMS and WhatsApp are not sent automatically. Check the current feature details against the channel your customers actually use.

Whatever tool you choose, compare the completed-job list with requests actually sent. Open a real test email on a phone and follow its link. If those two checks pass, you have a working process worth measuring over a longer period.