Ecommerce SEO · Guide

How Do Reviews and Ratings Impact Ecommerce SEO?

Two different kinds of review do two different jobs and they are constantly confused. On a large catalogue, customer reviews are also the only realistic way to get original text onto thousands of product pages.

Updated: July 2026
Written by: Andrew Odgers, Managing Director
Reading time: 11 minutes
Separated before anything else

Two Kinds Of Review

A review of a product and a review of your business are different things. They sit in different places, are collected differently, do different work, then are treated differently in search. Most articles on this subject run them together.

Product reviews. What a customer thought of the item.

These live on the product page. They are content and they are structured data, covered in blocks two and seven.

Business reviews. What a customer thought of buying from you.

These live on external platforms and profiles rather than on your own site. They are a trust and local signal, covered briefly in block three.

Why confusing them causes real problems. Because the advice for one is wrong for the other.

The most common consequence is a shop putting business reviews on its own pages and marking them up expecting stars in search results. Block seven explains why that has not worked for years. It is entirely avoidable once the two are separated.

What a shop needs. Both, collected in different ways.

The cheapest fix available

What Product Reviews Do

Product reviews persuade people, which everybody knows. They also solve the largest content problem on a shop, which almost nobody treats as their purpose.

The problem they solve. Duplicate supplier copy.

Per our guide to product descriptions, thousands of product pages carry the manufacturer's words, identical on every shop selling the item. A page with nothing distinctive has no reason to be chosen.

Why reviews fix it and rewriting does not. Cost and scale.

Nobody is writing five thousand descriptions. Customers will write reviews across the whole catalogue continuously, at no cost per page. Every review is text that exists on your page and nowhere else.

What makes review text unusually valuable. It contains what descriptions leave out.

Customers write about fit, durability, whether something matched the photograph, what they used it for and what went wrong. That is precisely the information a buyer wants and a manufacturer never supplies.

The other effect. Reviews use the words customers use.

A description written by a manufacturer uses industry terms. A review uses whatever the customer calls the thing, which is frequently what the next person searches for.

What this makes reviews. The cheapest content programme on the shop.

Which is why block four treats collecting them as a content activity rather than a customer service one.

Briefly

What Business Reviews Do

Reviews of your business do a different job. They build trust before somebody has chosen a product. For a shop with a physical presence they are a local signal as well.

Where they matter. Before the visit rather than during it.

Somebody deciding whether to buy from an unfamiliar shop looks for what other people experienced: whether delivery arrived, whether returns were handled, whether anybody answered the phone. None of that is a product question.

Why they sit off your site. Per block seven.

Reviews of you, published by you, carry limited weight precisely because you control them. Their value comes from being somewhere you do not control.

For a shop with premises. They also feed local visibility.

Our local SEO guides cover that properly, since it belongs to local search rather than to ecommerce.

What a shop should do about them. Collect them deliberately, on the platforms customers actually check.

Treated as a content programme

Reviews As Content At Scale

If reviews are the only realistic route to original text across a whole catalogue, then collecting them deserves the attention a content budget would get. Most shops treat it as an afterthought handled by an application nobody has looked at.

What changes when you treat it seriously. Volume and coverage.

The difference between a shop with reviews on a few hundred products and one with reviews across the range is usually not the customers. It is whether anybody set up the asking properly.

Coverage is the number that matters. Not the average score.

A shop with a handful of heavily reviewed bestsellers and nothing on everything else has solved the content problem for the pages that needed it least. What matters is how much of the catalogue has any original text at all.

How to increase coverage. Three things, in order of effect.

Ask everybody, automatically. A request sent after every order, without anybody deciding which customers to ask.

Ask at the right moment. Long enough after delivery for the customer to have used the product, which varies enormously by category.

Make it short. Every additional field loses people. A rating with two sentences is worth more to you than a detailed form nobody completes.

What to do about products with no reviews. Ask specifically.

A targeted request to recent buyers of an unreviewed product is the most direct way to fill gaps. It costs nothing beyond the sending.

Three rules. The law behind them

Collecting Them Properly

Three rules govern this. In the United Kingdom they are not merely search guidance. They are the law, with penalties attached.

Never filter by expected rating. Asking only customers you expect to be pleased.

Sometimes done deliberately, frequently done accidentally by excluding anybody who contacted support. Both produce a score that misrepresents what customers experienced.

Never incentivise. No discount, no entry into a draw, no free item in exchange for a review.

Never write them. Not yourself, not staff, not a supplier, not anybody commissioned to.

The legal position, checked on 29 July 2026. The Digital Markets, Competition and Consumers Act 2024 made these practices unlawful in the UK from 6 April 2025.

The Act bans submitting or commissioning fake reviews, meaning any review purporting to be based on a genuine experience when it is not. It bans incentivised reviews where the incentive is concealed. These sit among the practices listed in Schedule 20 as unfair in all circumstances. The Competition and Markets Authority published guidance on them in April 2025.

What happens to a shop caught. The Competition and Markets Authority can now impose fines directly, without first going to court, of up to ten per cent of global annual turnover.

This is general information rather than legal advice. A shop with a question about its own practices should take advice on it.

Why our rule on incentives is stricter than the law. The law bans concealed incentives. We say do not incentivise at all.

Disclosure is easy to get wrong. Per block seven, the aggregate score itself is within scope of the legislation rather than only the individual reviews.

A flawless score is a warning

Negative Reviews

Shops fear negative reviews and remove them where they can. Both instincts cost more than the reviews do.

What a perfect score signals. That something has been managed.

Buyers have learned this. A product with a hundred reviews and not one below the top rating reads as filtered rather than excellent, so experienced shoppers now go looking for the negative ones.

What negative reviews contribute. Credibility for the positive ones.

A mixed set makes the whole set believable. The presence of criticism is what tells a reader the praise was not arranged.

What they contribute commercially. Fewer returns.

A review saying something runs small prevents purchases that would have been returned. So does one saying assembly is difficult. That is a saving, arriving as something that felt like a problem.

How to respond. Publicly, once, without arguing.

Acknowledge it, say what you will do, then offer to sort it directly. The response is read by everybody who arrives afterwards. It is worth more than the review it answers.

What not to do. Suppress them.

Beyond the credibility cost, per block five, publishing reviews in a way that misleads is within scope of the same legislation. Removing genuine criticism to protect a score is not a presentational decision.

Verified before publishing

Structured Data And Rich Results

Review markup can produce star ratings in search results, which is why shops want it. What it can produce them for changed some years ago and a great deal of published advice has not caught up.

The position, checked on 29 July 2026. Google does not show review rich results for reviews an entity controls about itself.

Its guidance describes these as self-serving, meaning a review about a business placed on that business's own website. Pages using the LocalBusiness or Organization types. Their subtypes, are not eligible for review snippets where the entity being reviewed controls the reviews. Google announced this in September 2019 and the position sits in its current requirements, which were updated in July 2026.

The part shops most often miss. Embedding a third party widget does not change it.

Where the reviews are collected elsewhere and displayed on your own site, the entity still controls the arrangement, so the pages remain ineligible.

What remains eligible. Product reviews on product pages.

Reviews customers gave about a product, marked up on that product's page, are the primary supported case and remain so. Google also says sites should not aggregate reviews or ratings from other websites.

Because this changes, check it. The guidance was updated within the last month.

Anything a supplier tells you about review markup should be verified against the current requirements rather than against an article, including this one.

The rule underneath all of it. Structured data is an assertion about reality.

Never mark up a rating nobody gave. Per block five, in a shop that is a consumer protection matter and not only a search one, since the aggregate score is itself within scope of the legislation.

Three questions before signing

Third Party Review Platforms

Independent review platforms collect, verify and display reviews for a fee. They solve real problems and they carry a commitment worth understanding before you make it.

What they add. Verification and independence.

A platform confirming a reviewer actually bought the item removes the question a shop's own reviews always raise. That independence is the whole product.

What they cost. A subscription that scales with volume.

Frequently reasonable, though worth comparing against what your platform already provides, since many shops pay for a service duplicating something included.

Who owns the reviews if you leave. The question almost nobody asks first.

Years of accumulated reviews can be an asset you rent rather than own. Establish before signing whether you can export them in a usable form, whether they can be displayed on your site afterwards, then what happens to the profile customers have been looking at.

The related question. Whether the arrangement affects block seven.

Reviews displayed on your own site through a widget are still reviews you control for the purposes of the guidance, whoever collected them. That does not make the platform less useful. It means the stars are not what you are buying.

What we recommend. Decide what you are buying before choosing a supplier.

Verification and independence are worth paying for. Rich results are not available to buy, which is the misunderstanding these arrangements are frequently sold on.

Four, in order of seriousness

Common Mistakes

Buying reviews. The most serious. It is unlawful per block five.

Beyond the penalties, purchased reviews are identifiable by pattern and by language. A shop found to have them loses the credibility of every genuine review it holds.

Marking up an aggregate score collected from nowhere. A number placed in markup with no reviews behind it.

This is a false statement published in machine readable form. Per block seven, the aggregate score is within scope as much as the individual reviews are.

Hiding negatives. Per block six.

It damages credibility, removes the information that prevents returns, then publishing reviews in a way that misleads is itself within scope.

Never asking at all. The most common by far, also the least dramatic.

Most shops have an application installed that could be requesting reviews after every order and is either switched off, sending at the wrong moment or asking in a way nobody finishes. Per block four, that is a content programme sitting unused.

Ecommerce SEO services

The only writers who
work at your scale.

Nobody is writing five thousand product descriptions. Customers will write across the whole catalogue continuously, at no cost per page, in the words the next buyer will search for. Most shops have that switched off.

What is included every month:

Technical health and crawl Site structure work Quarterly technical audits Category page content Schema markup Website management AI optimisation Social, two posts a week

£350 per month, one target area. No setup fee, nothing billed separately.

The full guide series

Twenty-two guides.
One subject.

This guide covers reviews. The rest of the series covers the sequence, structure, category and product pages, technical health, measurement and everything a store owner has to decide.

Questions people ask

Reviews on a Shop

Do reviews actually help our rankings?
Product reviews solve the largest content problem on a shop, which is thousands of pages carrying identical manufacturer copy. Customers write across the whole catalogue continuously at no cost per page. Every review is text that exists on your page and nowhere else. They also use the words customers use, which is frequently what the next person searches for.
Can we get star ratings in search from reviews on our site?
For products, yes. For your business, no. As checked on 29 July 2026, Google does not show review rich results where the entity being reviewed controls the reviews, which covers the LocalBusiness and Organization types and their subtypes. Embedding a third party widget does not change that, since you still control the arrangement. Product reviews on product pages remain the primary supported case.
Can we offer a discount for leaving a review?
Our rule is not to incentivise at all. It is stricter than the law for a reason. In the UK the Digital Markets, Competition and Consumers Act 2024 has banned concealed incentivised reviews since 6 April 2025, so disclosure is the legal test. Disclosure is easy to get wrong. The aggregate score is itself within scope rather than only the individual reviews. This is general information rather than legal advice.
What happens if a shop is caught faking reviews?
In the UK, submitting or commissioning fake reviews has been unlawful since 6 April 2025 under the Digital Markets, Competition and Consumers Act 2024, sitting among the practices listed in Schedule 20 as unfair in all circumstances. The Competition and Markets Authority can now impose fines directly without going to court, of up to ten per cent of global annual turnover. This is general information rather than legal advice.
Should we remove bad reviews?
No, on three grounds. A product with a hundred reviews and none below the top rating reads as filtered rather than excellent, so experienced shoppers now look for the negative ones. Criticism is what makes the praise believable. And a review saying something runs small prevents purchases that would have been returned, which is a saving arriving as something that felt like a problem.
Is a paid review platform worth it?
Decide what you are buying first. Verification and independence are worth paying for, since a platform confirming a reviewer actually bought the item removes the question a shop's own reviews always raise. Rich results are not available to buy, which is the misunderstanding these arrangements are often sold on. Before signing, establish who owns the reviews if you leave and whether you can export them usably.