Asking for a review by text: the permission you actually need

A review request by text works. People read texts, they read them quickly, and a link in one gets tapped far more often than a link in an email. That is exactly why so many businesses send them on permission they do not have.

This guide is about the permission, because it is the whole question. The sending is the easy part.

The box they ticked when they booked does not cover it

If your booking page has a consent checkbox, read its wording. A well written one says something close to: messages about this appointment, including confirmations, reminders and changes. Ours says that, and it says explicitly that it does not cover promotions, review requests or reactivation campaigns.

That sentence is not lawyerly throat-clearing. It is what makes the consent worth having. Consent in US texting law is specific: it covers the purpose it was obtained for, and a request to review your business is a different purpose from telling somebody when their appointment is. A reminder serves the transaction they are already in; a review request asks them to do you a favour afterwards, and under the FCC's rules that lands on the marketing side of the line.

So using the appointment box for a review request is not a grey area. It is using permission for something it was never given for, and the evidence that you did is the wording you yourself showed them.

What "its own opt-in" means in practice

It means one more unticked box, and some words next to it. Four things make it worth having:

1. It is separate. A single checkbox covering appointment messages and review requests is not two permissions, it is one permission with a drafting problem. Somebody who wanted reminders and not marketing had no way to say so.
2. It is unticked. A pre-ticked box is not consent in any regime that requires consent, and a default of yes anywhere in your stack undoes the whole thing.
3. It names you. "Text me about my experience" is weaker than "Text me occasionally about my experience with Bright Smile Dental". The identification is part of what makes the agreement meaningful, and it is what a customer needs in order to recognise the message when it arrives.
4. Its exact words are kept. Not a flag saying consented, the sentence they were shown, stored next to the timestamp. If you change the wording next year, the record of what this person agreed to must not change with it. This is the part most systems get wrong, and it is the only part that answers the question a complaint actually asks.

Keep the frequency promise honest while you are at it. If the box says occasionally, asking after every visit for a course of six treatments is not occasionally.

Get the wording read by a lawyer, and know why

The words are doing legal work, so they are worth the hour. Three things to ask about specifically, because they are the ones that are easy to get almost right:

• The purpose, stated narrowly. A notice that claims less than you do is a problem; a notice that claims more than the customer understood is a bigger one.
• How somebody gets out. Reply STOP has to work, be honoured immediately, and be honoured across everything you send that person, not just the review asks.
• Spanish, if you serve Spanish-speaking customers. The exact words are the evidence, so a translation is not a formatting detail. Both versions need reading.

Until then, email is open, and it is not a consolation prize

Email marketing in the US runs on an opt-out basis under CAN-SPAM: you may write to an existing customer without prior permission, provided you identify yourself honestly, include a real physical postal address, and make unsubscribing work and honour it. That is a genuinely different legal regime from texting, and it is why a review request can go out by email today while the same message by text waits on a checkbox.

Three things to do rather than treat email as a fallback:

• Ask at the right moment. A few hours after the visit, not at the desk while they are still paying, and not a week later when they have moved on.
• Throttle it. A regular customer asked after every visit learns to delete anything from you. Ninety days between asks is a reasonable default, and it should be enforced in the system rather than remembered by a receptionist.
• Check eligibility at the moment of sending, not when the visit ended. Somebody who unsubscribes in between must not be asked.

What never becomes acceptable, on either channel

Do not route unhappy customers somewhere private. Sending the happy ones to Google and the unhappy ones to a feedback form is review gating. The FTC has fined businesses for it, Google's own policy forbids it, and it is the single fastest way to lose a listing you have spent years earning. Everybody gets the same link.

Do not offer anything for a review. A discount in exchange for a review buys you a review that is worth nothing and a policy violation that is worth less than nothing.

Where Frontiva is, plainly

Frontiva sends review requests by email after a completed visit, with the postal address and the unsubscribe that email needs, a throttle so a regular customer is not asked twice in ninety days, and a record of every ask including the ones it refused to send and why.

It does not ask by text, and the thing in the way is the checkbox and its approved wording rather than the sending. It also cannot read your Google or Facebook reviews back, chart your rating or draft replies to them, because that needs access to those platforms' own APIs. Asking and monitoring are two different products sold under one word, and it is worth knowing which one you are buying.