Thirty reviews is the starting point, not the finish line.
A grocery product can have a respectable average rating and still give shoppers a poor first impression if the first reviews they see are old, unrepresentative or say little more than “nice” or “good product”.
That is why we look beyond the headline star rating. For a SKU to be To Standard in a retailer, our working benchmark is straightforward:
- At least 30 reviews per SKU per retailer
- The Top 3 visible reviews are less than six months old
- Each of the Top 3 is within 1.5 stars of the product’s average rating, above or below
After that, the words themselves matter. Reviews should help another shopper understand what the product is actually like.
Shoppers do not just read stars. They read stories.
Average ratings are useful. They give shoppers a quick signal. But once a product has established credibility, the visible review text starts doing much more of the work.
A five-star review saying “lovely” gives the next shopper very little to go on. A review explaining the taste, texture, portion size, ease of use or occasion gives them something useful.
This is the thinking behind our Stories not Stars approach. Individual reviews matter because shoppers use them to answer practical questions: Will my family like it? Is it worth the price? Is it too sweet? Does it feel premium? Will it work for the occasion I have in mind?
Our own reviewer guidelines reflect this. We ask shoppers to write at least two sentences about their experience, stay focused on the product and give an honest, unbiased opinion as if they were talking to friends and family.
That produces something much more useful than chasing review numbers for their own sake.
Why 30 reviews alone is not enough
Getting to 30 reviews matters. Our research shows that the first 30 are particularly important in establishing credibility.
But review health changes.
A SKU that looks strong this month may have an old Top 3 six months later. A new one-star or five-star review may appear that is well outside the average. A product may be strong on Tesco but weak on ASDA or Ocado.
This is why portfolio averages can be misleading. Reviews need to be managed at SKU by retailer level.
Our Always On renewal work shows how quickly standards can decay: 85% of SKUs in any retailer will no longer be To Standard within six months of stopping an Always On programme.
The job is not to reach a review number and walk away. It is to keep checking what shoppers are actually seeing.
Freshness changes the impression shoppers get
Old reviews are not necessarily wrong. They are simply less useful as a picture of the product being sold now.
Recipes change. Packs change. Claims change. Shopper expectations change.
If the latest visible feedback is well over six months old, a shopper may be reading opinions about a version of the product that no longer reflects what is on shelf.
That is why freshness sits alongside volume in our standard. The Top 3 are the reviews most likely to shape the immediate impression of a product, so keeping them current matters more than building a very large archive and assuming the job is done.
The same issue becomes particularly clear with reformulations. A product that has been improved needs current feedback about the improved version, not a page dominated by comments on the old recipe.
Fair does not mean positive
There is an important distinction between a fair review picture and a positive one.
CheckoutSmart does not try to increase a product’s average rating. We do not remove negative reviews and we do not ask shoppers to leave a particular star rating.
Once a product has at least 30 published reviews, we look at whether any of its Top 3 reviews sits more than 1.5 stars away from the average, in either direction.
If it does, we regard that review as unrepresentative of the wider picture. The review stays where it is. We then source more genuine reviews so shoppers get a broader and more current set of opinions.
The average itself may go up or down as those reviews arrive. That is the point. The aim is a fair representation of real shopper experience, not an engineered score.
And if several shoppers keep raising the same problem, the answer may not be more reviews. Repeated complaints about taste, packaging, formulation or usability are useful product feedback and may need action from the brand.
How CheckoutSmart reviews are generated
The way reviews are collected matters because your brand reputation is attached to every one of them.
CheckoutSmart shoppers choose products that interest them, buy them themselves in-store or online and provide a receipt as proof of purchase.
They pay for the product upfront. Cashback is linked to the purchase, not to writing a review. Once the cashback has been awarded and the shopper has had the opportunity to use the product, they can be invited to leave an honest review.
Writing the review is optional. Shoppers are not paid more for doing it, penalised for declining or encouraged to leave a positive rating.
That gives brands feedback from verified purchasers who have used the product in a normal home environment rather than people being told what to say.
Why Always On works differently from a one-off review campaign
A one-off campaign can solve an immediate gap. It cannot stop reviews ageing afterwards.
SmartReputation Always On monitors agreed SKUs every day. New reviews are triggered when a product falls below one of the three standards: fewer than 30 reviews, an ageing Top 3 or a Top 3 review that no longer fairly reflects the average.
Once the product is back to standard, collection can stop. Monitoring continues, and the process starts again when required.
That means the brand team does not need someone manually checking hundreds of retailer pages every week.
The commercial case is not theoretical. In our Lactalis Always On case study, we supported 22 SKUs and delivered more than 200 reviews per SKU-retailer over the year. The average rating moved from 3.7 to 4.5 stars, although improving the rating was not the objective and will always depend on shopper experience.
More importantly, Lactalis saw +13% online sales for Always On SKUs versus non-Always On SKUs over the year. Individual timings and results will vary.
Reviews are becoming useful beyond the retailer product page
Reviews already influence shoppers online and in-store. They also provide brands with a large body of unprompted consumer language.
That text can reveal what shoppers repeatedly value or dislike: flavour, texture, scent, packaging, ease of use, value, specific occasions or unexpected ways a product is being used.
It can also help marketing teams understand whether shoppers describe the product in the same way the brand does.
This matters even more as AI becomes part of product discovery.
Our AI User Consumer Panel captures the questions real consumers ask AI tools and the answers they receive. Across our category work, retailer sites frequently appear as important information sources, while fresh product information and recent reviews strengthen the credibility of the answers being given.
That does not mean fresh reviews alone guarantee an AI recommendation. Product descriptions, claims, FAQs, retailer information and other sources all contribute. But a stale or poorly represented review profile is not a useful signal to put into that mix.
A word on disclosures and UK review rules
Review collection needs to be transparent.
CheckoutSmart’s position is that our model is non-incentivised because the cashback is linked to purchasing the product rather than writing the review. The review request comes afterwards and participation is optional.
We have nevertheless introduced disclosures on our reviews to make the journey clear to readers. These explain, in straightforward language, that the shopper received a saving or reward on the purchase and chose independently whether to leave their comments.
We believe this is the right approach for readers, reviewers and clients. Brands should still seek confirmation from their own legal or compliance team on legislation or guidance relevant to their business.
What grocery brands should do now
Start with the pages shoppers can see, not a portfolio-level average.
Audit every important SKU in every retailer against the same three measures: number of reviews, age of the Top 3 and whether those Top 3 fairly represent the wider rating. Then read the actual words.
Prioritise NPD, core sellers and products receiving significant media or shopper support. There is little sense paying to drive shoppers towards a product page that gives them weak social proof when they arrive.
And do not treat the exercise as finished once everything is green. Reviews move every day. The most effective approach is the one that keeps watching.
FAQs about better grocery product reviews
How many reviews should a grocery product have?
Our working standard is at least 30 reviews per SKU per retailer. The first 30 are particularly important for credibility, but review quality does not stop there. Freshness and the Top 3 visible reviews need to be managed as well.
What makes a good grocery review?
A useful review gives another shopper enough information to make a decision. We ask CheckoutSmart reviewers for at least two sentences about their experience, an honest opinion and comments focused on the product.
Does CheckoutSmart guarantee positive reviews?
No. We never guarantee a particular rating or positive review. Our shoppers are asked to give genuine, unbiased feedback based on their own experience.
What happens if a negative review appears in the Top 3?
We do not remove it. If it is more than 1.5 stars away from the product’s average once there are at least 30 reviews, our Always On process can trigger more genuine reviews so the Top 3 gives shoppers a more representative picture.
Why do reviews need to be managed continuously?
Because they age and new reviews keep arriving. A product that is To Standard today may not be six months from now. Always On monitors that change and acts when a SKU falls below standard.
Can reviews help with AI product discovery?
Our AI User Consumer Panel work shows that retailer content and fresh information, including recent reviews, can contribute to the information AI tools use when answering product questions. Reviews are one useful signal alongside clear product descriptions, claims, FAQs and retailer data.
If you want to see which of your SKUs currently need attention across Tesco, ASDA, Sainsbury’s, Morrisons, Ocado, Waitrose, Boots, Superdrug, Iceland, Amazon and other retailers, contact CheckoutSmart for a review analysis.
