Using ONS retail sales data with customer review software

What to take away

Which ONS retail sales and internet sales series to pull first

Start with the ONS retail sales series, because it is the monthly measure of how much UK households spent in shops and online. The ONS publishes retail industry statistics covering sales volumes and values, split by sector and by channel. The release to pull is the ONS monthly bulletin Retail sales, Great Britain, which carries the retail sales index in value and volume terms and is available to download from the ONS website.

That split is what lets you compare your own review volume against a national spending curve rather than against a vague sense that trade was up or down.

The ONS internet sales series is the second pull, and it is published alongside the retail sales index in the same monthly release. It gives the online share of retail spending, which is the closest national proxy for the traffic that reaches an ecommerce review form. If your review volume rises while the internet sales series is flat, you are probably collecting more reviews per order rather than selling more.

If both rise together, your review growth is mostly a demand story.

Use the ONS digital economy statistics when you need context on e-commerce and online retail trends beyond the monthly retail release.

The ONS business statistics give you wider market scope and demand context. That helps when you are explaining to a board why a category is shrinking even though your own review counts look healthy.

The Bank of England statistics add a view of consumer spending and e-commerce trends when the retail release alone feels thin.

Before you build anything, settle three definitions. First, your reporting month: calendar month or the ONS four or five week period. Second, your sales basis: value or volume, because the two diverge when prices move.

Third, your review basis: reviews submitted, reviews published, or reviews tied to a verified purchase. Mixing these three definitions is the most common reason a benchmarking exercise produces a chart nobody trusts.

A practical starting point is a single spreadsheet with one row per month and four columns: ONS retail sales index, ONS internet sales share, your net revenue, and your review volume. Add your rating and response metrics later, once the first four columns reconcile. This keeps the first pass small enough to finish.

If you already report on commerce reviews key metrics, reuse those definitions rather than inventing new ones for the ONS exercise. Consistency between your internal reporting and your external benchmarking is worth more than a slightly better metric.

Reading BRC-KPMG data alongside ONS releases

The BRC-KPMG data is the BRC-KPMG Retail Sales Monitor, a monthly retail sales monitor produced from a panel of retailers, and it is published separately from the ONS release. It is not a replacement for the ONS retail sales series. It is a cross-check.

When the two sources point the same way, you can treat the national direction as settled and focus on the size of your own gap. When they diverge, treat the month as uncertain and avoid drawing conclusions from a single data point.

Divergence often reflects different panel composition, different timing within the month, or different treatment of categories such as food and fuel.

The BRC-KPMG Retail Sales Monitor is also useful because it is published monthly, ahead of the ONS release for the same month. That gives you an earlier view of the month, which matters if your review reporting cycle closes before the ONS figures land.

You can publish an internal review dashboard with a provisional national line, then revise it when the ONS release arrives.

Both sources are backward looking. Neither tells you what your customers will do next month. Their value is in removing the excuse that a weak review month was simply the market. Sometimes it was. Often it was not.

For a fuller treatment of how to choose between competing benchmark sources, see our guide to commerce reviews benchmark research. The same logic applies to BRC-KPMG data: it is one input, and it should not be the only one.

Lining ONS monthly data up with your review volume and rating

Monthly alignment is the core discipline. Your review volume and your rating should be plotted on the same time axis as the ONS retail sales series and the ONS internet sales series, with no gaps and no mixed periods.

Do this in four steps.

Rating needs different handling. Average rating is bounded between one and five, so it cannot be indexed the same way. Instead, track the share of reviews at four and five stars, and the share at one and two stars, month by month. Those two shares move more than the average and they are easier to explain.

A simple table helps keep the monthly alignment honest.

The pattern in that example is the common one: review volume outrunning national online sales, with the one and two star share falling. That usually means your review collection is working, not that your products suddenly improved.

Seasonality does most of the work in UK retail. November and December lift both sales and review volume, January lifts return-related reviews, and the summer months are flatter. If you compare December with July without adjusting, you will mistake seasonality for performance.

Comparing response rates and review velocity by month

Response metrics are where most UK retail review programmes are weakest. Response rate is the share of reviews that receive a reply, and it should be reported separately for positive and negative reviews. A single blended rate hides the fact that a team replies to most five star reviews and few one star reviews.

Velocity is the second metric. Review velocity is the number of new reviews per hundred orders, or per thousand pounds of revenue, in a given month. It is the cleanest way to compare review performance across categories with different order values.

Both metrics need a seasonal baseline. If you only started measuring velocity in a strong month, your baseline is wrong. Use at least twelve months, ideally twenty four, before you set a target.

Neither the ONS nor the BRC-KPMG release publishes a review-rate or response-rate benchmark, so the comparison here is about shape and direction: the national series tells you what the market did, and your own twelve-month history tells you what normal looks like for your shop.

A useful checklist for the monthly review meeting:

Response time matters in the UK because the Consumer Rights Act 2015 and, for some practices, the Digital Markets, Competition and Consumers Act 2024 set expectations about how firms treat consumers. A slow response to a genuine complaint is a commercial risk, not just a service metric.

The Competition and Markets Authority has taken an interest in how review and ranking practices are presented, and the Advertising Standards Authority has rules on testimonials and endorsements. The Information Commissioner's Office enforces UK GDPR and the Data Protection Act 2018, which govern how you store and use reviewer data.

None of that changes your monthly numbers, but it changes how you should present them. A review programme that cannot show a response rate is harder to defend than one that can.

Building a customer review software dashboard that tracks ONS cycles

Your customer review software is the source of the review data. The dashboard is the layer that puts it next to national data. Keep the two separate, because review platforms change and national series do not.

Build the dashboard around the ONS release calendar. The ONS retail sales series lands on a known monthly schedule, and the BRC-KPMG data lands on a different one. Your dashboard should show a provisional national line and a confirmed national line, so nobody has to remember which release they were looking at.

Include a small set of views. One view for review volume against the internet sales series. One for rating distribution. One for response metrics by rating band. One for velocity. One for the gap analysis. Fewer views means your team opens the dashboard; more than five and it sits unused.

If you are building this for a board audience, the structure in commerce reviews reporting dashboard is a sensible template. The main addition here is the national comparison line and a note on which ONS release it came from.

Data governance matters as much as the chart. Review data often contains personal data, so retention periods, access controls and lawful basis sit under UK GDPR and the Data Protection Act 2018. If you are aggregating review text into themes, keep the audit trail.

If you are exporting review data to a warehouse, check your VAT and record keeping obligations with HM Revenue & Customs where the data supports ecommerce reporting.

Finally, decide who owns the monthly commentary. A dashboard with no named owner drifts. One short written note per month, tying the review numbers to the ONS release, is enough.

Interpreting gaps between national retail sales and your own review performance

A gap between national retail sales and your review performance is not automatically a problem. It is a question. There are four common explanations, and each has a different response.

The first is market share movement. If national sales are flat and your review volume is rising, you may be taking share. Check your own sales before you celebrate. If your sales are also rising, the gap is real and worth explaining in your reporting.

The second is collection change. A new post-purchase email, a change in the review request timing, or a switch in customer review software will move review volume without any change in demand. Mark those dates on the chart. A gap that starts on the day you changed your request flow is not a market signal.

The third is category mix. UK retail is not one market. Grocery, fashion, electricals and homeware have different seasonal shapes and different review rates. If your category is a small part of the national aggregate, the ONS retail sales series will not track it closely.

Use the ONS business statistics for category context and accept that the national line is a rough guide.

The fourth is genuine operational change. A delivery problem, a returns backlog or a pricing error will show up in review volume and rating before it shows up in national data, because national data is an average. In that case your reviews are the leading indicator and the ONS release is the lagging one.

For a wider discussion of which numbers deserve attention, see our guide to commerce reviews measurement. The principle is the same: a metric earns its place only if someone would change a decision because of it.

Common questions