A store owner told me her review app was broken. She'd collected several thousand reviews over two years, her product pages showed them, but no star ratings appeared in Google — and when she asked ChatGPT about her own products, it never mentioned they were well-reviewed at all.
The app wasn't broken. It was doing exactly what most review apps do: keeping her reviews inside its own widget instead of writing them into the place Shopify, Google, and AI assistants actually read.
This is a gap most merchants never think to check, and it's quietly expensive. Your reviews can render perfectly on the page for a human visitor and be completely absent from the structured data that every machine deciding whether a stranger trusts your product depends on.
The symptoms look like unrelated bugs
- No star ratings under your product titles in Google search results.
- No rich review snippets — even though competitors with fewer reviews have them.
- AI assistants (ChatGPT, Perplexity, Google's AI surfaces) describe your products with no sense that they're well-reviewed.
Owners usually blame the theme, or assume Google is slow, or — like my client — conclude the review app is faulty. Sometimes it really is the theme. But increasingly often, the reviews simply never made it into Shopify's native review data in the first place.
Shopify has a spec for this — and most apps skip it
This is the part that surprises people. Shopify defines a standard, reserved place for review data, specifically so it's portable and readable across the whole ecosystem:
- Two standard product metafields —
reviews.rating(the average star rating) andreviews.rating_count(the total number). These are the values a well-built theme reads to render stars, and the values Google reads to show ratings in search. They're API-managed and not even visible in your admin — which is exactly why most merchants never realise whether they're populated. - A standard product-review metaobject — a full, structured format for the individual reviews themselves: rating, title, body, author, verified-buyer status, language, and more. Reviews written here become eligible to appear in Shopify's own Shop app, and they're stored in a consistent shape any theme or tool can use.
Here's the catch, and it explains everything: writing to that standard format is a program review apps have to opt into and be approved for. Most don't. They keep your reviews in their own database and render them through their own widget — because that keeps you dependent on them. There's a whole spec from Shopify for it; most apps just choose not to use it.
Your reviews exist. They're just sitting in a private silo instead of the public, machine-readable layer where buying decisions get made.
Why this happens: the app's incentive isn't your visibility
A review app's business model rewards lock-in. If your reviews live only inside its widget, switching apps means losing your social proof — so you don't switch. Writing your reviews into Shopify's portable standard format does the opposite: it makes the data yours, and yours to take with you. Some apps do it well. Many quietly don't, and the marketing never mentions the difference.
The cost lands at the worst possible moment: discovery. Star ratings are one of the strongest trust signals in a search result, and we're moving into an era where AI assistants — not just blue links — decide which products get recommended. Those systems read structured data. Reviews trapped in a JavaScript widget are reviews that don't count in the rooms where buying decisions increasingly start.
What to do about it
1. Check whether the standard values are populated. This is a quick, definitive test — query the reviews.rating and reviews.rating_count metafields on a few products. Either they're there or they aren't. If they're empty while your widget shows dozens of reviews, that's your answer.
2. If your current app can write the standard data, turn it on. Some apps support it but don't enable it by default. The fix can be a settings change plus a small theme tweak to read the native values — not a migration.
3. If it can't, weigh switching — but do the math first. Moving review apps has real switching costs; you don't want to lose your review history or break collection. The right move depends on catalog size, how many reviews you have, and how much organic and AI-driven discovery matters in your category. For some stores it's worth it; for others, a custom approach is cleaner.
4. For full control, the reviews can live in your own store data. If you need design freedom, multi-market and translation support, or to group reviews in ways no app offers — by vendor, for example — reviews can be modelled in your own metaobjects and rendered straight from the theme. Fully owned, fully portable. It's more work up front, and worth it for stores where reviews are a core asset.
The bottom line
You spent two years earning those reviews. Whether they show up as stars in Google and get cited by AI assistants comes down to a piece of plumbing you can't see in your admin and were never told to check. It's one of the cheapest, highest-leverage fixes in the structured-data category — the reviews are already won; this is just making them count where it matters.
If you're not sure whether your reviews are reaching Google and the AI layer, that's a five-minute thing to verify and usually a same-week thing to fix. Grab a slot and I'll tell you which it is.
