Industries
Reviews page on your website: what to put on it (55 live pages measured)
Yes, a dedicated reviews page on your website is worth having once you have more reviews than one carousel can show: it gives visitors who want to check you a single place to read everything, and it gives you an address to link from the menu, the star rating in your header and your emails. The list itself is the easy part, because any review widget can draw it. What decides whether the page works is what you put around the list: a heading that says whose reviews these are, a short paragraph of your own text, a clear next step above the list, and a sort order that does not bury recent reviews. On most of the 55 live reviews pages we measured, that frame was missing or a few words long.
Data collected on 11 October 2026. Crawler numbers cover the 30 days before that.
How we found and measured 55 reviews pages
Our starting list was every site where one of our widgets drew reviews on 5 or more dates within the past month: 373 sites once previews, staging copies and test hosts were removed. For each, we fetched the start page and the sitemap and kept every address whose path contained a word for reviews in roughly twenty languages: reviews, testimonials, feedback, opinie, bewertungen, avis, opiniones and so on. 331 start pages answered. 142 sites had at least one matching address. We took up to three per site, 212 in total, rendered each in Chromium with a 1440 by 900 window, went all the way down the page and saved what it contained.
75 distinct pages on 67 sites contained a list of reviews from our widget. 55 of them, on 48 sites belonging to 42 customers, used the Reviews Page widget, which is built for exactly this job. The other 20 used a testimonials carousel or a summary instead. The figures below are about the 55 unless we say otherwise.
One limit up front: we only found pages whose address contains a review word. A page at
/about-us/ that happens to hold the full list was invisible to this method, so 67 of 331 sites,
about one in five, is a floor for how many have a dedicated reviews page, not a count.
What a crawler gets: the server’s HTML against the rendered page
Our widget, like most review widgets, draws the list in the visitor’s browser after the page has loaded. That has a consequence worth measuring: the HTML your server sends and the page a person sees are very different documents.
| Measure (55 Reviews Page pages) | Median | Spread |
|---|---|---|
| Words in the server’s HTML | 292 | 156 to 751 (middle half) |
| Words in the server’s HTML outside header, menu and footer | 90 | 31 to 495 |
| Words on the page after rendering and scrolling | 2,560 | 1,311 to 3,535 |
| Words inside the review widget after scrolling | 2,420 | 1,087 to 3,274 |
| Share of all rendered words that belong to the widget | 90.8% | 82% to 96% |
On 45 of the 48 sites, the server’s HTML contained none of the review text. We checked by taking fragments of the first reviews shown on screen and searching for them in the raw HTML. The other three sites put review text into the HTML on the server.
Whether that matters depends on who is reading. Google renders JavaScript. Over the last 30 days, requests carrying the Googlebot user agent asked our API for review data 9,990 times, for the widgets of 258 accounts; Bingbot did so 22,058 times and Applebot 6,920 times. Those requests only happen when a crawler runs the widget’s script, so these crawlers do see the list. Our logs record only requests that miss the Cloudflare cache, so the real numbers are higher.
A crawler that does not run JavaScript never makes that request. For it, the median reviews page in our scan is about 90 words of content plus a menu. That is not a reason to avoid widgets. It is a reason to write the text around the list as normal page text, because that text is the only part every crawler, link preview and screen reader gets on the first request.
The frame around the list is where most pages are thin
We split the rendered text that is neither the widget nor the site’s header, menu and footer into two parts: what sits above the widget and what sits below it.
Above the list, the median page had 24 words of its own. 26 of the 55 pages had 20 words or fewer, which in practice means a heading and nothing else, and 7 had no text at all before the reviews started. Below the list, 30 of the 55 pages had no text at all.
The bottom of a reviews page is a poor place for anything important anyway. 52 of the 55 pages load 20 reviews at first and fetch the next 20 as the visitor scrolls. In our own scroll test, twelve screens deep, the median page grew to 80 reviews and over 14,000 pixels. A call to action, an FAQ or a contact block placed under the list keeps moving away from the visitor who is looking for it. Put the next step above the list, next to the rating summary, where it stays put.
The heading needs the same attention. The widget renders a rating summary and the reviews, but no heading tag of its own, so the page’s main heading has to come from you. 13 of the 55 pages had no H1 at all. 33 had an H1 that contained a word for reviews or testimonials, and 49 had such a word in the page title, so most site owners name the page correctly in the browser tab and then leave the page itself without a title. Our own reviews page was in the same state until an audit on 28 July 2026 found it had no H1. It now opens with a one-line heading and a short paragraph that says where the reviews come from, and the widget below it is unchanged.
How site owners set up the list
The widget’s own settings tell us how people configure a reviews page in practice. We read them from the requests each page sent to our API while it loaded.
| Setting | What the 55 pages chose |
|---|---|
| Sort order | 23 best rated first, then newest; 10 best rated first, then longest; 22 newest first |
| Review sources | 10 one source, 25 two sources, 20 three or more (median 2) |
| Reviews available to the page | Median 368; 18 pages had 1,000 or more; 6 had 50 or fewer |
| Language filter | 3 pages filter reviews by the page’s language |
| Review form on the page | 12 pages show a button that lets visitors write a review |
| First batch | 52 pages load 20 reviews, 3 load 50 |
Three things stand out. First, 33 of 55 pages, 60%, put the best ratings first. That is a display choice, not a problem, as long as every review is still reachable by scrolling and you do not describe a filtered list as all your reviews; our guide to the FTC rules on displaying reviews explains where the line is. The cost of best-first is that a visitor checking whether you are still good today has to scroll past older praise to find out. If your recent reviews are strong, newest first is the more convincing order.
Second, most pages combine sources. Google was on 46 of the 55 pages and Facebook on 34, and Google plus Facebook was the most common pair, on 15 of the 25 pages with exactly two sources. Mixing sources on one page is what a dedicated page is good at, because a homepage carousel has no room to show where each review came from.
Third, the language filter is almost unused. Only 3 of 55 pages filter reviews by the page’s language. On a site with separate language versions, a German page full of English reviews reads as an afterthought, and that is exactly the case the filter exists for.
Indexing basics: mostly right, with a few self-inflicted gaps
Most pages had the plumbing in place. 43 of the 55 were linked from their site’s homepage, 43 were listed in the sitemap, and 45 declared themselves as the canonical address. The gaps were small but complete: 5 pages carried a noindex instruction, in a meta tag or a response header, so search engines are told to leave them out entirely, and 10 had no canonical tag. A noindex on a reviews page is sometimes deliberate, but more often it is a setting copied from a template or left over from a staging copy. It takes a minute to check.
Do not build the page expecting review stars in Google results. Google does not show star snippets for reviews a business publishes about itself on its own site; it calls them self-serving and has excluded them from rich results since 2019. We covered the details in our guide to review schema markup. A reviews page earns its keep with visitors who are already checking you, not with stars in the results.
What to put on a reviews page
| Element | Why it matters | What we saw on 55 pages |
|---|---|---|
| H1 with your brand and the word reviews | The widget has no heading of its own | 13 had no H1 |
| 50 to 150 words of your own text above the list | The only content every crawler gets on the first request | Median 24 words above the list |
| Where the reviews come from, and that you show all of them | Answers “are these curated?” before it is asked | Rarely stated |
| Rating summary with the review count | “4.8 from 368 reviews” carries weight that “4.8” alone does not | Shown by the widget |
| One next step above the list: book, buy or contact | Content below an endlessly loading list is out of reach | 30 had no text below the list |
| The full list, in batches | Fast first load, nothing hidden | 52 load 20 at a time |
| A link from the menu, the header rating and your emails | A page nobody reaches persuades nobody | 43 linked from the homepage |
The paragraph above the list does not need to be clever. Say whose reviews these are, which platforms they come from, that you show every review the source shows, and what the visitor can do next. That is 60 to 100 words, and it turns a page that is 91% widget into a page with a topic.
Do you need one at all?
Not always. If you have a dozen reviews, a testimonials block on the homepage shows all of them and a separate page adds a click for no gain. Only 2 of the 55 pages in our scan had 20 or fewer reviews available, so most site owners already reach for a dedicated page once the count grows.
A reviews page also does not replace reviews where the decision is made. It is the place you send people who want to read everything; the star line in your hero and a review block near your offer still do most of the persuading. Our study of where sites put reviews found the first review block on a homepage a median of four screens down, so most sites need both.
Where we fit
The reason to use us for this is that one snippet builds the whole page inside your own site, on your domain: the rating summary with the review count, an optional form for collecting new reviews, and the full list that loads in batches as the visitor scrolls. You pick the sources, Google, Facebook, Booking, Etsy and the rest, the sort order and, if you run several language versions, the language filter. Our own reviews page is the widget running live, with the heading and paragraph we recommend above it. Plans and limits are on the pricing page.
The caveats, stated plainly. The Reviews Page widget is part of Pro and Ultimate, not the free or Plus plans, with a 7-day trial. It renders in the visitor’s browser, so the reviews are not in the HTML your server sends: Google, Bing and Apple render them, but a crawler that does not run JavaScript sees only the text you write around the widget. And the widget does not write that text for you; the heading, the intro and the next step are yours to add.
If you already have enough reviews to fill a page, it is a short job. Create an account, connect the sources your customers review you on, put the widget on a page of its own, and write the heading and the first paragraph before you publish it.
FAQ
Is a reviews page good for SEO?
It can be, but not because of the review list alone. A widget that loads reviews with JavaScript adds nothing to the HTML your server sends: on 45 of the 48 sites we measured on 11 October 2026, the server's HTML contained none of the review text. Google renders JavaScript and does see the list, but crawlers that do not run JavaScript see only what you wrote around it, which on the median page was 90 words outside the header, menu and footer. A heading with your brand and the word reviews, plus a short paragraph of your own text, is what gives the page a topic for every crawler.
How many reviews should a reviews page show?
All of them, loaded in batches. Most of the pages we measured load 20 reviews at first and fetch the next batch as the visitor scrolls, which keeps the page fast without hiding anything. A dedicated page is worth having once you have more reviews than fit in one carousel: the median page in our scan had 368 reviews available, and only 6 of 55 had 50 or fewer.
Should the reviews page be in the main menu?
Yes, or at least linked from the homepage and the star rating near the top of the page. A reviews page nobody can reach does not reassure anyone. 43 of the 55 pages we measured were linked from their site's homepage. Add the address to your sitemap as well, and to the emails where you ask customers about their experience.