There’s a lot of speculation right now about what you need to do to “rank” in AI-generated answers. One of the more specific claims I’ve heard recently is that you need at least a 4.4 rating on your Google Business Profile (GBP) to have any real visibility.
That sounds unrealistic, given the amount of data I have access to as a leading provider of review response management for the past 15+ years.
But why debate when you can actually test the assumption and see what happens, which is what I did.
The Test: Ask AI a Real Apartment Search Question
I asked ChatGPT a simple question about a search people make thousands of times a year.
“Where’s the best place to live in Rittenhouse Square, Philadelphia?”
The response wasn’t a list of properties sorted by review score. It wasn’t even framed around ratings at all. Instead, the answer broke the market down in a way that actually reflects how people make decisions (and in a better way than I would have, even being quite familiar with the submarket).
What the AI Actually Did
The response grouped properties into meaningful categories:
- Luxury, full-amenity buildings
- Classic, historic Rittenhouse Square options
- Value-oriented rental apartments
It also identified “micro-locations” within the neighborhood.
Then it named specific properties within each category. Here are the ones that made the list:
- 1909 Rittenhouse
- Josephine
- 2116 Chestnut
- The Harper
- The Touraine
- Rittenhouse Claridge
- The Wellington
- Roosevelt Apartments
The AI didn’t say, “Here are the highest-rated properties.” It suggested ways to think about solving the problem and then showed several sets of answers.
That’s a completely different result than you’d find if you did a traditional search.
The 4.4 Rating Theory Falls Apart
Here’s where it got interesting.
The response identified 1909 Rittenhouse as the best overall luxury building in the submarket.
That’s a strong endorsement.
But 1909 doesn’t meet the supposed “requirement.” Its Google rating is around 4.3. So much for the 4.4 threshold!
Multifamily reviews tend to skew lower than other industries. Operational complaints, maintenance issues, and resident expectations based on the price point all factor in. Don’t get me started about reviews left by people who were fired, evicted, or are in a completely different business that we can’t get taken down.
In fact, according to Google, the average Google review rating for apartment complexes (I know we hate this word, but that’s what Google calls them) generally hovers between 3.5 and 4.2 stars.
What This Tells Us About AI Visibility
AI isn’t blindly filtering based on a single metric like review score. It’s doing something more nuanced:
- Understanding how renters evaluate choices
- Grouping options by lifestyle, budget, and preferences
- Selecting properties that represent each category well
- Pulling from a mix of signals (not just reviews)
Reviews still matter a lot, but they’re part of a broader picture. There is no hard threshold.
A Good Example of This Nuance
One thing I appreciated in the response was how it handled positioning.
It mentioned Rittenhouse Claridge (one of our clients) as a strong example of a classic, on-the-square property and highlighted the character and location, not luxury finishes.
That’s exactly how the website positions the community, and because we built the website, I know that there was a great amount of attention and resources spent towards thoughtfully and accurately positioning the community. The community has new amenity spaces, including a fabulous roof deck, but it’s a Grand Dame, not new construction.
ChatGPT didn’t try to force every community into the overused “luxury” category. It understood that different renters value different things, and offered a nuanced answer accordingly.
The Reality: Reviews Really Matter… But They’re Not Enough
Let’s be clear. You need to encourage happy residents to leave reviews and respond to them. They still matter because:
- They build trust
- They shape perception
- They feed into AI training data
You should absolutely be working to improve them.
But the idea that there’s a magic number that determines whether you show up doesn’t hold up.
Where did this number even come from? It’s likely a generalized finding across industries where review behavior is very different. Multifamily is its own special world.
You Can’t Always Afford the Best
1909 Rittenhouse looks like an incredible place to live. It checks every box for modern, luxury, fully amenitized living in the heart of the city.
But not everyone can afford to live here. You’re looking at $3,000+ per month rents even for a studio. This isn’t NYC. That’s a lot for Philadelphia!
Despite that price point and its positioning as the “best overall” in the AI response, it’s sitting at a 4.3 rating, not 4.4. That alone should tell you everything you need to know.
The takeaway is simple. Work on your reviews. They still matter. But don’t assume that missing some arbitrary threshold means you’re invisible to AI because clearly, you’re not.
From the desk of Ellen Thompson, Co-founder and CEO of Respage >> Since its founding, Respage has helped over 10,000 communities attract, engage, and retain residents. Its platform assists properties in generating leads, automating leasing, and managing reputation and social media. Thompson is also the Founder of Results Repeat, a digital marketing agency that has helped hundreds of companies create a digital presence and use SEO and paid marketing to generate more business online.
