UK online-shopping statistics can look contradictory even when they are measuring the same month. In September 2025, ONS data showed internet sales at 27.2% of retail sales, card-based online spending at retail merchant categories at 40.7%, and online spending across all UK consumer card spend at 50.5%.
Those numbers are not interchangeable. They come from different measurement systems, scopes and definitions. Understanding that difference matters whenever online-shopping data is used to explain consumer behaviour or technology-buying trends.
The 27.2% figure measures retail internet sales
The ONS Retail Sales Inquiry measures internet sales as a percentage of total retail sales. The September 2025 figure was 27.2% on the non-seasonally adjusted series.
This is a retail measure, not a measure of every card transaction made online. It focuses on goods retail and uses the ONS retail-sales framework.
The 40.7% figure uses retail merchant card categories
ONS also publishes Visa-based card data mapped to retail merchant category codes. In September 2025, 40.7% of spending in those retail merchant categories was online.
The mapping is provisional, and card-not-present or online card spending is not identical to the ONS retail internet-sales definition. This is why the result sits well above the Retail Sales Inquiry figure.
The 50.5% figure covers all consumer card spending
The broadest figure is the share of all UK consumer card spending that was online. In September 2025 this was 50.5%.
That includes categories beyond retail goods. Travel, services and other types of card spending can contribute to the online share, so it should not be labelled as the percentage of retail shopping done online.
The 23.3-point gap can be decomposed, but not casually explained
Simple subtraction gives a 23.3 percentage-point gap between 50.5% and 27.2%. The intermediate retail-card measure means 9.8 points separate all-card online spend from retail-card online spend, while 13.5 points separate retail-card online spend from the ONS retail internet-sales ratio.
That decomposition is arithmetic, not an ONS statement of cause. The series use different bases and one mapping is provisional.
Keep the measurement label attached to the number
The safest rule is simple: never quote an online-shopping percentage without saying what it measures.
“27.2% of retail sales were internet sales” is different from “50.5% of card spend was online”. Both can be useful, but only when the scope remains visible.
For 4KM-Tech readers comparing online consumer behaviour, this distinction is more valuable than chasing one universal number that does not exist.
Sources: Office for National Statistics Retail Sales Index internet-sales data and ONS consumer-card spending analysis, September 2025. The card-to-retail mapping is provisional and the series should not be merged as though they use one methodology.