Do Reviews Affect Whether AI Recommends My Security Company?
Reviews do shape what ChatGPT and Google say about an alarm company, but star rating is the smallest part. Here is what matters and how to ask the honest way.
A homeowner asks ChatGPT who to call for an alarm system. The answer names two companies and says one of them is "known for fast installs and clear pricing." That description came from somewhere. Very often it came from what customers wrote in reviews.
So the short answer is yes. Reviews do affect what ChatGPT and Google say about your company. But the star rating is the smallest part of it. What counts more is how many reviews you have, how new they are, what the words say, and how many places they show up. This post covers each one, and then how to ask for reviews without breaking the rules.
One honest note first. Neither Google nor OpenAI publishes a formula for which companies get named. What follows is what a dealer can see by reading the answers, plus plain common sense about what evidence is out there to read.
Free, no call needed
Free: Does ChatGPT Name Your Company?
Four questions a property owner would actually type, asked today with live web search. You see every company name that came back, in order, and whether yours was one of them. Four draws on one day, not a ranking. Leave your email and we will send it over, then walk you through what it found.
How AI uses your reviews
When someone asks for a recommendation, the assistant needs proof that a company is real, still active, and good at the job in question. Reviews are some of the plainest proof on the open web.
Google's AI answers can draw on your Google Business Profile, and your reviews are part of that profile. ChatGPT reads the pages it can open when it searches the web, and those pages often include directories and review sites where your customers have written about you.
Here is a made-up example. Two alarm companies both have a 4.8 star rating. Company A has 35 reviews, the newest is from 2023, and most say "Great company!" Company B has 90 reviews, four from the last month, and they mention camera installs, taking over an old system, and a technician by name. The stars are tied. The evidence is not. Company B gives the AI far more to work with.
Recent reviews show you are still in business
A review from three years ago proves you did good work three years ago. It says nothing about today. Owners sell, techs leave, and service changes. Fresh reviews are the simplest sign that the company a buyer calls this week is the same one the old reviews describe.
A steady pace beats a burst. Five new reviews every month for a year looks like a busy company. Sixty reviews in one week followed by silence looks like a campaign, and it can look suspicious to Google's own filters.
A good target is simple: never let your newest review be more than 30 days old.
The words matter more than the stars
An AI answer is made of words, so it leans on words. "Five stars, great job" gives it nothing to repeat. "They took over our old panel, added two cameras, and had us running the same day" gives it three facts it can connect to three kinds of questions.
You cannot write reviews for customers, and you should not hand them wording. You can ask a better question. Compare these two requests:
- "Would you leave us a review?"
- "Would you leave us a review and mention what we did for you today?"
The second one tends to bring back reviews that name the work. Over a year, that builds a record of the jobs you want more of, written in your customers' own words.
Reviews in more than one place
Many dealers put all their effort into Google. Google should come first, because it is where many buyers look first. But a company whose only reviews are on Google has one source vouching for it. A company with reviews on Google and on one or two other sites has several.
Pick one second site your buyers already use, such as Yelp, the Better Business Bureau, Facebook, or Nextdoor. For one month each quarter, send the review link for that second site instead of Google. If you close 25 jobs that month and 1 in 5 customers follows through, that is 5 reviews on the second site each quarter and 20 in a year. That is enough to show a real pattern.
How to ask without gaming it
The rules here are clear, and breaking them can cost you the reviews you already have.
Google's review policies do not allow these:
- Rewards for reviews. No gift cards, no discounts, no free month of monitoring.
- Asking only happy customers. Sending the review link to people you know are pleased, and steering everyone else to a private form, is against the policy.
- Reviews from the inside. Owners and employees may not review their own company.
On top of that, a Federal Trade Commission rule that took effect in 2024 bans fake reviews and bans paying for reviews on the condition that they are positive.
The honest method is also the easy one:
- Ask every customer, after every install and every service call.
- Ask the same day, by text, with a direct link to the review form.
- Keep the message to two sentences.
- Send one reminder a few days later, then stop.
The math is simple. Say you finish 25 jobs a month, you ask all 25 customers, and 1 in 5 leaves a review. That is 5 new reviews a month and 60 in a year, with no tricks. Your own numbers will differ, so count your jobs and track your real response rate for two months.
Our free review request kit has the text messages ready to copy.
Reply to every review
Google lets the business owner reply in public to each review. Many dealers skip this, and it is a missed chance.
A reply adds a few more plain words about the work. "Thanks, Maria. Glad the new door sensors and the app are working well for you" confirms the job in your own voice. It also shows anyone reading, person or AI, that somebody is paying attention.
For a bad review, reply within two business days. Say what you did or will do to fix it, give a phone number, and stop there. Do not argue. In our business there is one more rule: never confirm details about a customer's system, address, or account in a public reply. That is a security problem for them, even if they brought it up first.
In our view, a two-star review with a calm, useful reply does less damage than a two-star review left sitting alone.
What to track
Once a month, write down four numbers:
- Total Google reviews.
- The date of your newest review.
- Total reviews on your second site.
- How many reviews you have replied to.
Then, every 90 days, check whether the AI answers are changing. Ask six buyer questions twice each in ChatGPT and twice each in Google, and count how often your company is named out of 24 answers. This guide walks through the steps, and our free ChatGPT naming check runs a version of it for you.
Nobody can promise that more reviews will get your company named by a certain date. Reviews are one kind of evidence among several. What steady, honest reviews do is make the case for your company easier to find and easier to believe. That holds up no matter how the AI tools change.
Where to start this week
Set up the same-day text. That one step takes about ten minutes, and everything else in this post builds on it.
To see how your company looks to AI right now, run our free AI visibility audit. For the wider plan beyond reviews, read how to become a most recommended alarm company.
Free, no call needed
Free: Does ChatGPT Name Your Company?
Four questions a property owner would actually type, asked today with live web search. You see every company name that came back, in order, and whether yours was one of them. Four draws on one day, not a ranking. Leave your email and we will send it over, then walk you through what it found.
