Review Signals

Review signals are the review-related data points that influence how a business ranks, both in Google's traditional local pack and in AI citation systems — and the two contexts weight them differently enough that it's worth treating them separately rather than assuming what works for one automatically works for the other.

Review signals in the local pack

In local pack ranking, review count and rating feed into prominence, one of Google's three declared ranking pillars. The relationship here is gradual, not a gate — more reviews at a solid rating tends to move a business incrementally further up the pack over time, and there's no published cutoff where a business flips from "not competitive" to "competitive" at some specific review count. Google has never disclosed a threshold, and no independent, verifiable one exists either, so any number presented as a hard cutoff is a guess dressed up as a fact.

Review signals in AI citations

AI citation systems appear to behave somewhat differently. For "best" and "recommended" style queries, review count seems to function less like a smooth gradient and more like a rough legitimacy check — a business with a handful of reviews is less likely to get surfaced as a confident recommendation than one with a substantial, active review base, though again, no engine has published an exact number and none should be assumed. Once a business clears whatever informal bar an engine is implicitly applying, further differentiation among cited businesses appears to run on velocity and rating rather than on count alone.

The signals that make up the category

Total review count carries different weight in each context — a general activity and legitimacy signal for AI systems, one input among several into prominence for the local pack.

Average rating acts as a quality filter across both contexts. Lower ratings correlate with reduced visibility in both local pack ranking and AI recommendation eligibility, though the relationship is again a gradient rather than a documented cutoff.

Monthly velocity, the pace of new reviews arriving, signals that a business is currently active rather than dormant — this is review velocity specifically, and it carries meaningful weight in both local ranking and AI citation eligibility.

Recency matters somewhat independently of velocity: a handful of reviews from last month generally carries more weight for an activity assessment than the same number of reviews spread thin from three years ago.

Review text itself functions as a kind of entity enrichment for AI systems specifically — reviews that mention specific services, treatments, or product names give AI engines more concrete material to associate with the business's entity profile than generic one-line reviews do.

How this varies by platform

For Google AI Overviews, Google reviews are the primary review data source, layered on top of Google's own indexed web content. For ChatGPT and Perplexity, Google reviews factor in indirectly, while Practo, Zomato, and JustDial reviews come into play more directly whenever those specific pages get retrieved and browsed as part of answering a query.

India context

For most Indian local businesses, review count sitting below whatever informal bar AI systems are applying is the single most common AEO gap across every category Angryturtle works in — more common than missing schema, more common than incomplete GBP attributes. Building a steady, ongoing review velocity is usually the highest-leverage AEO investment available, precisely because it moves the review signal set on both fronts, local pack prominence and AI citation eligibility, at the same time. There's no magic number to hit; the direction that matters is a business's velocity trending up rather than sitting flat or declining.

Why this is a gradient, not a checklist

It's tempting to want a target number — "get to X reviews and you're set" — because a fixed target is easier to plan around than an open-ended gradient. But treating review signals as a pass/fail gate at some invented threshold produces bad prioritisation: a business that hits a made-up number and stops investing in review generation will drift backward relative to competitors who keep growing theirs. The signal that actually matters is the trend, sustained over months, not a one-time count crossed off a list.

Related terms: Review velocity → · Review sentiment → · Reputation management → · Prominence → · GBP AI optimization → · Answer engine optimization →

GBP AI Optimization → | Review Generation Engine → | Managed Local SEO →