A lot of review software still ships with a feature that asks the customer how the job went, then routes the happy ones to Google and the unhappy ones to a private feedback form. It is sold as reputation management. It is against Google's policies, and it is also less effective than the boring alternative.
What gating actually is
Gating is any process where the request to post publicly depends on the sentiment the customer expressed first. The usual shape is a one-to-five star prompt in an email: four and five stars get the Google link, one to three get an apology form that goes to the owner's inbox.
The intent is understandable. Nobody wants a bad month permanently attached to their profile. The mechanism is the problem: it is selective solicitation, and the selection is on expected sentiment.
Why it is a problem
Google's review policies prohibit soliciting reviews selectively from customers based on their likely rating, and discouraging or prohibiting negative reviews. Gating is the textbook example of both. Reviews collected that way can be removed, and enforcement action attaches to the profile — the same profile whose map-pack ranking is the reason you wanted reviews in the first place.
There is a commercial argument too, independent of policy. A profile of nothing but five stars reads as curated. Buyers have learned to scroll to the negative reviews to find out what a company is actually like, and a company with none of them looks managed rather than trustworthy. A handful of middling reviews with a calm owner response underneath does more for a hesitant homeowner than a clean sweep.
What works instead
Ask everyone, consistently, on a fixed delay, with one reminder. That is the entire strategy, and its power is that it is boring enough to actually keep running.
- Ask every completed job. Not the ones you feel good about. Consistency is what produces volume, and volume is what makes any single bad review statistically unimportant.
- Trigger on job completion or payment, so it does not depend on anyone remembering.
- Link straight to the write-review dialog, not to your profile page. Every additional step costs you reviews, and asking a customer to find the review button themselves is the most expensive step of all.
- One reminder, then stop. Two is the ceiling. Chasing stars costs you more goodwill than the star was worth.
- Make it obviously from you. A message that reads like a marketing platform gets deleted like one.
Recency matters as much as count. A profile with forty reviews whose newest is two years old reads worse than one with twelve where the newest is from last week. A steady trickle beats a burst, which is another argument for automating the ask rather than running occasional campaigns.
What to do about the bad review
Respond to it, publicly, once, without arguing. The response is not aimed at the person who wrote it — that relationship is usually settled. It is aimed at the next homeowner reading, who is deciding what you are like to deal with when something goes wrong.
Three rules. Do not dispute the facts in public, even when you are right. Do name the specific thing and what changed because of it. Do not offer money in the response, which reads as buying silence.
A calm reply under a bad review converts better than the absence of the review would have.
The timing question nobody gets right
Send the request when the work has proved itself but the experience is still fresh. Those two pull in opposite directions, and the right point differs by trade.
A drain clear or a repair can be asked about the same evening — the customer knows within hours whether it worked. A water heater install is better at two or three days, once they have lived with hot water. A remodel is worse the sooner you ask, because the customer is still exhausted and finding snags; a couple of weeks after the final walkthrough is far stronger.
Whatever you choose, keep it fixed. A delay you set once and never think about again is the delay that still runs in six months.