Star ratings and review badges on a landing page. Drawn to shiny things

· Founder of Magpie AI

Published · 7 min read

In short

Show a star rating, review count or award badge only when a visitor can check it: name the site it comes from, link to its page there, keep the figure as that page shows it today, and retire a badge once its season is long over. A true 4.6 can sell better than a polished 5.0, and in the UK a misleading rating is banned outright.

Contents

Five stars and a number fit on one line under a headline, and a visitor reads them in a second. That's why they get copied from page to page, rounded up a little, and left in place long after the figure behind them has moved on.

On 11 November 2026 we read 87 of the landing pages we'd been testing our reports on. It's a hand-picked set rather than a fair sample of the web, so take the share loosely, but about 37 of them showed at least one rating, review count, award badge or rating hidden in the page's code for search engines. Among the ones that wouldn't survive a check were stars that named no source and ratings kept in the code where no visitor could see them. Both are quick fixes.

Why a 4.6 can sell better than a 5.0#

Researchers at Northwestern University's Spiegel centre found in 2017 that purchase likelihood typically peaks for ratings between 4.0 and 4.7, "and then begins to decrease as ratings approach 5.0", across product categories. The same report found a product with five reviews was 270% more likely to be bought than one with none.

That was shop products, not software or services, and the study doesn't say why. My guess is that a buyer who sees 5.0 starts wondering who left the bad reviews out. So I'd show the real 4.6, with its count and its source, before I'd show a round 5.0 with nothing beside it, because a careful buyer is going to wonder about a flawless score regardless.

Name the site every figure comes from#

"★★★★★ Loved by 1,000+ teams" says nothing about who did the loving, or where. A visitor has to take it on trust, and a landing page is about the last place they'll do that.

UK law takes these figures as seriously as the reviews themselves. The Competition and Markets Authority's guidance on the Digital Markets, Competition and Consumers Act (opens in a new tab) counts "overall ratings, overall summaries, review counts and rankings" as consumer review information, and the Act bans publishing it "in a misleading way". A business that shows on its own website the rating it got on a review site is, in the guidance's words, a "second-party publisher" with a duty of its own, one it can't hand over to that site. I'm not a lawyer, and this isn't legal advice, but the guidance is plain enough on that point.

The review sites want the source named as well. Trustpilot's brand guidelines ask for the word "TrustScore" beside the number, and its logo with the rating. G2 wants a badge's "full badge name (including segment or region, if applicable)".

Here's a made-up booking app at tablefor.example, before and after:

BeforeAfter
★★★★★ Rated 5 stars by restaurantsTrustScore 4.6 from 312 reviews on Trustpilot, linked to its profile
G2 LeaderG2 Leader, Restaurant Management, Small Business, Summer 2026, linked to its category
1,000+ five-star reviews4.7 from 1,180 ratings on the App Store, linked to its listing

Baymard Institute's 2023 testing of tours and experiences sites watched people click third-party rating badges, expecting the reviews behind them. Where the badge went nowhere, some left the site to search for the reviews on the rating site themselves. One called it a "dead link". Another said "this is not a link. This is just like a screenshot."

Trustpilot asks for the link in so many words ("Link back to your Trustpilot profile page when using our brand on your website"), and G2 wants a backlink to its category page for every badge. I think it's the same doubt that makes visitors trust quotes on outside sites more than testimonials on your own site, and a link is about the cheapest way there is to answer it on your own page.

Keep the figure as it stands today#

On 14 October 2026 the Advertising Standards Authority ruled against a jewellery retailer whose product page showed "4.7/5 | 2540+ reviews" beside Trustpilot's name. When the ASA looked on 4 August 2026, the Trustpilot profile had "a rating of 4.1 from 1,662 reviews", and the retailer accepted it hadn't updated the claim.

The CMA's guidance says something similar about fake reviews, that a star rating or count not "promptly" updated once fakes are removed is published in a misleading way. Any figure typed into your site's code stops being true the day the real one moves. You can use the review site's own live widget, or check the number against the source by hand once a month. The second is duller, but it works on every page builder going.

Careful with "#1 rated"#

In October 2025 the ASA ruled against an internet provider that called itself the "#1 rated internet provider on Trustpilot". Its own rating was 4.6, and the top provider in that Trustpilot category had 4.9. It only came first after filtering to London and 4.5 stars or more, then sorting by the number of reviews, which took about 400 companies down to 20.

The ASA read the claim the way a visitor would, as highest rated against all its rivals. If yours needs a footnote explaining the filters, it's a smaller claim than it looks, and I'd just make the smaller claim (the same goes for "cheapest" and "the only", which read as claims about every rival at once).

Award badges have a season#

G2's guidelines recommend that references "Reflect current quarter and year", and its real-time rankings should carry "the date of when the rankings were pulled". So a "Leader Winter 2023" badge in late 2026 makes a careful buyer wonder whether the product has placed since, which isn't much to strut like a peacock over.

Launch awards are a bit different. "Product of the Day" names one day and stays true about that day, so it can stay as long as it says which launch site gave it and links there. I'd still move it below the fold once newer proof exists, since it's a fledgling's prize from launch week and says little about the product now.

The rating in your code counts too#

Plenty of sites carry a rating in their structured data, the hidden markup that asks Google for stars in its search results. Google's rules for it are short. "If you use AggregateRating, users should be able to see that aggregate rating on the page", and "Don't aggregate reviews or ratings from other websites." Since September 2019 it has also shown no stars for a business or organisation marked up as reviewing itself on its own site ("a review about entity A is placed on the website of entity A").

Of those 87 pages, 15 carried this markup, and some held figures such as 5.0 from 86,000 ratings that no visitor could see anywhere on the page. I can't see the upside in that. Show the figure with its source, or take the markup out.

Show the good with the bad#

Picking only the five-star reviews makes your rating harder to believe, and in the US and UK it can break the rules. The FTC's rule bars a business from implying the reviews it displays are most or all of those submitted while it suppresses some for their low rating or negative words (16 CFR 465.7). The UK Act names "failing to publish, or removing from publication, negative consumer reviews whilst publishing positive ones" as one way of publishing misleadingly, and Trustpilot's own list of don'ts includes "displaying only five star reviews without a disclaimer".

You can still lead with your best ones. Put the overall rating beside them, linked to the rest, so a visitor can see you're not hiding the 3-star reviews.

What Magpie checks#

Magpie's full report reads every rating, review count and award badge on your site by code, with no model involved. It checks that each names the site it comes from, links to it and is current (a seasonal badge whose season ended over a year ago gets flagged). An App Store rating is compared with Apple's own figure for the app, and a rating in the page's code that no visitor sees is flagged too. Review sites that turn automated visits away, as the big ones do, are taken as shown and never got round.

Questions people ask#

Do I need a paid plan to show a Trustpilot or G2 badge?#

Usually, yes. Trustpilot's guidelines say businesses not on a Trustpilot plan "aren't allowed to display their TrustScore or Star Rating on their websites". G2's free profiles can't show badges from its Summer 2025 reports onwards, though they keep earlier ones and the "Users Love Us" badge.

Should I show a rating under 4?#

I wouldn't put it on the first screen. Spiegel's figures put the peak between 4.0 and 4.7, so below that the number is probably working against you, though I'd add that it was shop products they looked at, and a 3.8 for a niche bit of software with twelve reviewers might read quite differently to someone who knows how few people review that sort of thing. Either way I'd lead with two or three specific, named quotes instead, and keep the rating lower down for anyone who goes looking.

Where should the rating go on my site?#

Near the claim or the button it backs up. On the first screen, one short line with its source beside the main button is plenty, and a wall of badges can sit further down for whoever's still weighing it up.

Sources

  1. Spiegel Research Center, How Online Reviews Influence Sales (Northwestern University, 2017) (opens in a new tab)
  2. Competition and Markets Authority, Fake reviews guidance, CMA208 (2025) (opens in a new tab)
  3. Digital Markets, Competition and Consumers Act 2024, Schedule 20 (opens in a new tab)
  4. Advertising Standards Authority, ASA Ruling on JPD Retail Ltd (14 October 2026) (opens in a new tab)
  5. Advertising Standards Authority, ASA Ruling on Community Fibre Ltd (8 October 2025) (opens in a new tab)
  6. Baymard Institute, Tours and experiences: reviews and third-party ratings (Edward Scott, 2023) (opens in a new tab)
  7. Trustpilot, Legal Brand Guidelines, version 2.0 (2023) (opens in a new tab)
  8. G2, Community Guidelines (opens in a new tab)
  9. G2, Badge Refresh: Important Updates Bringing Greater Exclusivity to Software Vendors (2025) (opens in a new tab)
  10. Federal Trade Commission, 16 CFR 465.7, Review suppression (Cornell LII) (opens in a new tab)
  11. Google Search Central, Review snippet structured data (opens in a new tab)
  12. Google Search Central, Making review rich results more helpful (2019) (opens in a new tab)

About the author

· Founder of Magpie AI

Alex founded Magpie AI and writes about what turns a landing page's visitors into customers, drawing on what its audits find on real sites, from the first screen to the checkout

Related articles

Fix what loses you customers.Fix what loses you customers. Then bring more of them in

  • Free score, no sign‑upNo sign‑up
  • No credit card neededNo card needed