The Number You’ve Been Chasing Is the Wrong One
AI search engines, Google’s Local Service Ads, and the local pack don’t rank by star average anymore. They rank by recency, response rate, content density, and cross-platform distribution. A star average above ~4.3 is now table stakes. The marginal value of going from 4.6 to 4.9 is small. The marginal value of one new review this week is enormous.
Recency does most of the work. A cluster of old reviews followed by months of silence signals to Google that you may no longer be actively serving customers, and your ranking drops. As Ferocious Media’s 2026 LSA analysis put it: a contractor with 60 reviews averaging 4.7 beats one with 12 reviews averaging 4.9. The math the industry has run for a decade is exactly inverted.
60 reviews from this year at 4.7 stars beat 500 reviews from 2021 at 4.9. The number the industry has chased for a decade is the wrong number.
The Four Levers That Actually Rank You Now
Response rate is the second new lever. Google made responsiveness one of the heaviest-weighted LSA factors in late 2024 — quietly. Response within five minutes now correlates with an 8x conversion lift over a 30-minute wait. Most companies don’t have a review-response cadence at all, because the dashboard their marketing director looks at doesn’t surface response rate.
Content density is the third. AI engines read the actual text of reviews as ground truth about what you do. A 200-word review describing “they fixed our 4-ton Trane that was leaking refrigerant and replaced the capacitor in 90 minutes” gives ChatGPT service-specific keywords to surface you for that exact problem. “Great service, thanks!” gives it nothing. Total reviews stopped mattering; median review length started mattering.
Cross-platform distribution is the fourth. Google reviews feed Gemini and AI Overviews; Yelp feeds ChatGPT; BBB and Angi feed the older citation-driven models still active in the long tail. If 95% of your velocity is on Google alone, you’re a thin signal across half the AI surface that drives discovery now.
The strategic implication: stop measuring reviews by star average and total count. Start measuring four things — reviews per completed job last month (target 25%+), median review word count (target 50+), response rate within 48 hours (target 90%+), and platform distribution (target Google 60% / Yelp 30% / other 10%). The dashboard most home services teams look at literally doesn’t show any of these. That’s the gap. (It’s the same recency signal behind last week’s AI visibility audit.)
The question worth asking on Monday
Pull your reviews from the last 30 days. Count them. Read the median one — how long is it? How many did you respond to? If your honest answer is “fewer than 25% of jobs, under 50 words, less than half responded to,” your star average isn’t going to save you. The new game is recency, depth, and response.
From the team at DUO
Most home services dashboards still optimize for the star average that stopped mattering in 2024. We rebuild review reporting around the four metrics that actually move LSAs, AI search, and organic ranking now — plus the FSM automation that gets you there. If your review reporting hasn’t been updated since the AI search shift, we’ll do it.
Rebuild my review reporting →
