Google Shopping product ratings: getting stars on your listings
How star ratings appear on Google Shopping ads and listings, the product ratings programme, review feeds vs aggregators, minimum review counts, and product vs seller ratings.
The stars under a Shopping ad come from Google's product ratings programme, and they measurably lift click-through, listings with stars consistently outperform identical listings without them. Getting in requires around 50 reviews across your account, delivered either through an approved third-party aggregator or your own review feed, matched to products by GTIN. This guide covers how the programme works, the two routes in, and the difference between product ratings and seller ratings.
What product ratings are and why they matter
The row of stars under a Shopping ad ("★★★★½ (1,204)") comes from Google's product ratings programme. It's not a feed attribute you set; it's an aggregation Google builds from review data submitted by merchants, review platforms and marketplaces, keyed to the product's identity.
The commercial case is blunt: stars are the strongest visual differentiator a Shopping listing can carry short of price. Two identical tiles, one with stars and one without, do not perform the same: ratings lift click-through rate meaningfully (double-digit percentage improvements are commonly reported), and because Shopping auctions reward engagement, better CTR compounds into better effective visibility. Stars also pre-qualify the click: a shopper who chose the 4.6-star listing arrives warmer than one who clicked on price alone.
Ratings appear on paid Shopping ads and free listings alike: see how Shopping listings work for where each surface draws its data.
The two routes into the programme
Route 1: an approved third-party aggregator
If you collect reviews through one of Google's approved review partners (the established review platforms most merchants already use), the partner syndicates your reviews to Google for you. You keep collecting reviews as normal; the plumbing is their problem. This is the right default for most merchants: no feed to build, no separate approval, and the partner handles moderation standards Google requires.
Route 2: your own product review feed
If you collect reviews yourself (native platform reviews, a custom system), you can apply to the product ratings programme directly and submit an XML review feed to Merchant Center. Each review entry carries the reviewer, timestamp, rating value, review text, and (critically) product identifiers that let Google match the review to the product. You're taking on moderation obligations too: reviews must be genuine, unedited, include the negative ones, and follow the programme's content policies.
| Aggregator | Own review feed | |
|---|---|---|
| Setup effort | Low (enable syndication) | High (application, XML feed, ongoing maintenance) |
| Control | Partner's pipeline | Full |
| Cost | Platform subscription | Engineering time |
| Best for | Most merchants | Large catalogues with an in-house review system |
Whichever route you take, the reviews themselves have to exist: which means asking for them, systematically, post-purchase. A shop doing 300 orders a month with a decent post-purchase email flow reaches the programme threshold in weeks; a shop that never asks stays at zero forever.
Minimum review counts: the two thresholds
There are two numbers, and merchants regularly confuse them:
- ~50 reviews account-wide to be accepted into the product ratings programme at all. This is a gate on you, roughly: enough volume to demonstrate genuine review collection.
- ~3 reviews per product before that specific product's stars display. This is a gate on the product: Google won't show a rating built on one data point.
The practical consequence: joining the programme doesn't put stars on everything. A 5,000-product catalogue with 400 reviews concentrated on 60 best-sellers shows stars on those 60: which, conveniently, is usually where the traffic is anyway. Because ratings aggregate across merchants, popular branded products you resell may show stars from day one (other sellers' reviews already exist for that GTIN), while your own-brand products only ever show what you collect yourself.
Seller ratings vs product ratings
Two separate programmes that merchants conflate constantly:
| Product ratings | Seller ratings | |
|---|---|---|
| What's rated | A specific product | Your shop as a business |
| Where it shows | Shopping ads and free listings, under the product | Text ads, and alongside your store name on Shopping surfaces |
| Data source | Product review feeds, aggregators, marketplaces | Google Customer Reviews, approved partners, Google's own signals |
| Shared with other merchants? | Yes (aggregated per product across sellers) | No, yours alone |
| You control it by | Collecting product reviews | Delivery, service, and collecting shop reviews |
The distinction matters for diagnosis. "Why don't my ads have stars?" has different answers depending on which stars: product stars need the ratings programme and per-product review depth; the shop rating needs your business-level review volume and quality. And they compound: a 4.7-rated product from a 4.8-rated seller is the strongest tile in the auction.
Deep dive Identifier matching: why GTINs decide whether your reviews count
Product ratings are aggregated per product, across every merchant and review source selling it. That aggregation needs a key: something that says "this review, this listing, and that other retailer's listing are all the same kettle". That key is the product's identity, and in practice it's the GTIN.
Follow the mechanics through and several non-obvious consequences drop out:
- A review without a solid product match is a wasted review. If your review feed (or your aggregator's) carries reviews keyed to a SKU that doesn't resolve to the product's GTIN, Google can't attach them to the canonical product. The reviews exist; the stars don't move.
- A wrong GTIN is worse than a missing one. A product carrying another item's GTIN doesn't just risk disapproval: it inherits the wrong product's rating. Your premium kettle showing 3.1 stars because the feed carries the GTIN of a different model is a conversion killer nobody thinks to check for. If your identifiers are shaky, fix that first: see GTIN issues.
- Variant identity decides where reviews pool. Reviews keyed to a parent product versus per-variant GTINs behave differently: 40 reviews spread across 8 colour variants may leave every variant below the display threshold, where the same 40 pooled on a properly-related product family clears it. How your feed expresses variants and their identifiers directly shapes whether stars appear.
- Resold products inherit the market's rating; own-brand products don't. For branded resale, clean GTINs plug you into an existing pool of reviews: instant stars. For own-brand goods, you are the pool, which makes systematic review collection an asset competitors can't share.
This is why ratings work is feed work before it's marketing work. The review platform can be flawless, but if the identifiers underneath are missing, malformed or misassigned, the matching fails silently: no error in Diagnostics says "your reviews aren't attaching". The audit worth doing: for your top 50 products by traffic, confirm each has a valid, correct GTIN in the feed, then confirm the review platform keys its reviews to that same identifier. Enrichment that validates and fills identifiers across the catalogue does the same job at full scale: the stars are downstream of the data.
Making ratings work harder
- Concentrate collection where it counts. Post-purchase review requests on your highest-traffic products clear the 3-review threshold where it moves the most revenue. A best-seller sitting at 2 reviews is the cheapest win in this whole area.
- Don't fear moderate ratings. A 4.3 with 500 reviews outperforms a suspicious 5.0 with 6. Volume is credibility; the programme requires you to submit negatives anyway.
- Stack the tile. Stars combine with price annotations and promotions on the same listing: a rated product with a visible offer is the strongest possible presentation.
- Point spend at the strong ratings. A product whose stars and review depth beat its competitors' converts above its category average: a genuine signal when deciding which products deserve a bidding push, alongside the benchmarks in the price competitiveness report.
Frequently asked questions
How many reviews do I need for stars on Google Shopping?
Roughly 50 product reviews across your whole account to join the programme, and typically at least 3 reviews on an individual product before its stars show. The account threshold is the gate; the per-product threshold decides which listings actually display stars.
How do I get product ratings on Google Shopping?
Two routes, collect reviews through a Google-approved third-party aggregator that syndicates them for you, or apply to the product ratings programme and submit your own XML review feed to Merchant Center. Either way, reviews must be genuine, moderated, and matched to products (GTIN is the reliable key).
What's the difference between seller ratings and product ratings?
Product ratings are stars for a specific product ("this kettle, 4.6 from 210 reviews") aggregated across every merchant selling it. Seller ratings score your shop as a business, delivery, service, experience, and appear on text ads and your merchant name. They're separate programmes with separate data sources.
Do product ratings include reviews from other retailers?
Yes. Google aggregates reviews for a product across sources and merchants, keyed on the product's identity. A shopper may see stars fed partly by reviews left elsewhere, and your reviews likewise strengthen the product's rating wherever it shows.
Can I show ratings from my own website reviews?
Yes, if they're genuine, tied to the specific product, and submitted correctly, via an aggregator or your own approved review feed. You can't hand-type a rating into the feed; review data goes through the ratings programme, not a product attribute.