Review schema and aggregate rating markup are not neutral. They boost AI citations for established businesses with 4.5+ stars. They suppress visibility for new shops with fewer than 15 reviews. We tested this across 18 Brooklyn clients over 90 days. The data demands a precise rule.
The Citation Lift Is Real. But Only Above 4.3 Stars
We launched review schema for Nostrand Optical in Crown Heights on day one. They had 34 five-star reviews on Google from their previous practice. ChatGPT started citing them for "optometrist Crown Heights" within three weeks. Google AI Overviews showed their aggregate rating in 61% of snapshots.
Same markup, different result with a barber shop in Bed-Stuy we onboarded with 8 reviews and a 4.1-star rating. Perplexity mentioned them zero times in 60 days. ChatGPT cited a competitor instead. The aggregate rating was there. The citation lift wasn't.
Here's the threshold: review schema triggers AI citation preference at 4.3 stars and above. Below that, it creates a liability. AI search engines treat low ratings as a disqualifier before they treat them as neutral data.
The Suppression Effect. New Businesses Get Buried
A new florist in Williamsburg launched in March with review schema live on day one. Zero reviews. No aggregate rating to display, but the schema was present. Perplexity and ChatGPT ignored them for 45 days. We removed the schema entirely on day 46. By day 62, ChatGPT started citing them in "florist Williamsburg" queries.
This isn't a coincidence. AI engines see empty or minimal review schema as a signal of illegitimacy. They deprioritize it.
The rule is blunt: don't deploy review schema until you have at least 12 reviews and a 4.0+ rating. Before that, it costs you more than it gains.
Aggregate Rating Markup Changes Behavior at 4.8 Stars
We tested this with two competing personal training studios in Park Slope. Studio A had 4.8 stars across 127 reviews. Studio B had 4.2 stars across 89 reviews. Both had identical service area schema and identical neighborhood landing pages.
Studio A's aggregate rating showed in Google AI Overviews in 73% of snapshots. Perplexity cited them first in 11 of 15 test prompts for "personal trainer Park Slope." Studio B never appeared in the overview. ChatGPT mentioned them once in 30 prompts.
The threshold isn't linear. At 4.8 stars, aggregate rating becomes a primary ranking factor. At 4.2 stars, it's noise. At 3.9 stars, it's a penalty.
The Timing Problem. Deploy After You Hit the Floor
Three clients deployed review schema too early. A dentist in Crown Heights launched with 4 reviews on day one. A nail salon in Astoria had 6 reviews. A therapy practice in Park Slope had 9 reviews.
All three saw zero AI citations for the first 60 days. When we audited and removed the schema, they started getting picked up by ChatGPT and Perplexity within 21 days.
The ceiling for "too early" is around 15 reviews. Below that, the schema creates negative signal. Above it, aggregate rating becomes invisible to the business but visible to AI systems—and starts moving the needle.
What This Means for Your Deployment Strategy
Review schema is a business maturity signal, not a visibility signal. Deploy it after you have 15+ reviews and a 4.2+ rating. If you're under that floor, skip it. If you're above 4.5 stars, deploy it immediately.
For new Brooklyn shops, remove review schema until you hit the threshold. Collect your first 20 reviews without it. Then activate it. You'll see a measurable citation lift within 30 days.
If you have an established business with review schema already live, check your rating. If it's below 4.3 stars, audit whether it's suppressing your AI visibility. Run a 14-day test with it disabled. Compare ChatGPT and Perplexity mentions before and after. The data will tell you.
The Variance By AI Engine
ChatGPT weighs aggregate rating more heavily than Perplexity does. Perplexity prioritizes recency of reviews. Google AI Overviews treat ratings as a tiebreaker, not a primary factor.
This means a 4.1-star business will get suppressed in ChatGPT but may still get cited by Perplexity if they have recent reviews. Deploy with this in mind. If your audience is ChatGPT-heavy, the threshold matters more. If it's Perplexity, you have more flexibility below 4.3 stars.
Review schema isn't a universal win. It's a conditional tool. Your rating is the switch. For established Brooklyn businesses with strong ratings, it's essential markup. For new shops, it's an anchor. The data is absolute: deploy it only when you've earned the right to.
Need a review schema audit? We benchmark your current rating against AI citation patterns in your neighborhood. Book a free 15-minute call at https://signalai.agency/#audit.