Online retail has stabilised at 27.3% of total UK retail (ONS, April 2026), ending years of post-pandemic volatility. Cards still settle most checkouts, but digital wallets now command 40% of e-commerce value, and AI-assisted “agentic” shopping is emerging as the next disruption.
Two lead figures plus four supporting metrics that show how big online retail is, who pays on a phone, and how cross-border and AI flows fit in.
The ONS Retail Sales release publishes weekly and monthly internet-sales-as-percentage-of-all-retail. The 2020–2021 peak above 36% has unwound; since mid-2024 the share has held remarkably steady around 27%, reaching 27.3% in April 2026.
The 2021 peak was a one-off pull-forward. The market has settled at a new baseline meaningfully above 2019, and has now been stable for two years.
| Category | Online share of retail sales |
|---|---|
| 2007 | 3.4% |
| 2014 | 11.3% |
| 2019 | 19.2% |
| Jan 2021 (peak) | 37.1% |
| 2025 | 27.5% |
| Mar 2026 | 28.7% |
Annual averages; 2026 is the April monthly reading.
| Year | Online % | YoY (pts) |
|---|---|---|
| 2020 | 27.9 | ▲ 9.0 |
| 2021 | 31.4 | ▲ 3.5 |
| 2022 | 26.6 | ▼ 4.8 |
| 2023 | 26.1 | ▼ 0.5 |
| 2024 | 27.1 | ▲ 1.0 |
| Apr 2026 | 27.3 | ▲ stable |
The 2020–2021 surge was a one-off pull-forward of demand, not a permanent step-change. The market has settled at a new baseline meaningfully above 2019, and has now held it for two years. Online retail growth is mature: the story has shifted from “how much” to “how people pay”.
UK Finance splits online card transactions by entry method. Wallets win because they remove friction and reduce SCA challenges. The underlying rail is still the card.
Digital wallets and direct card entry account for the vast majority of UK online spending; BNPL and pay-by-bank together are still under one in seven checkouts. Card-funded wallets count as wallet rather than card, so cards underpin more than the direct-entry share alone suggests.
| Category | Share of UK online spending by value |
|---|---|
| Card, direct entry | 46.0% |
| Digital wallet | 40.0% |
| BNPL | 8.0% |
| Pay by bank / other | 6.0% |
Wallets have grown by capturing share from manual card entry, not by displacing it as a payment instrument. The card is still the underlying rail for the 40% wallet share. BNPL (8%) and pay-by-bank are growing but together still account for around one in eight online checkouts.
UK Finance and the ONS both report that the majority of online retail transactions originate on a phone. The share has risen steadily and is now well over 60% by transaction count.
Mobile and wallets reinforce each other: a phone shopper is far more likely to use a stored mobile wallet than to manually type a card number. No UK primary body publishes a device split, so the figures alongside are estimates (see the note below).
Share of transactions, %.
| Year | Mobile | Desktop | Tablet |
|---|---|---|---|
| 2020 | ~52 | ~42 | ~6 |
| 2022 | ~58 | ~37 | ~5 |
| 2024 | ~62 | ~33 | ~5 |
| 2026 | ~64 | ~32 | ~4 |
Mobile and wallets reinforce each other. By transaction count, desktop is now a minority channel for UK online retail, especially for under-35s. Checkout optimisation in 2026 is mobile-first by default.
Cart abandonment is reported by acquirers and platform vendors, not by a UK regulator. The figures here are industry estimates rather than official statistics; we name the source for each.
Long-running benchmark from the Baymard Institute, 2026 update. Read as a directional figure, not an official statistic.
Baymard Institute · 2026Mobile baskets are abandoned far more often than desktop, despite carrying most of the traffic, the central conversion problem of mobile-first retail.
Industry estimate · 2026When an issuer mandates an active 3DS challenge, 15–25% of UK carts are lost. Frictionless approval (rich-data 3DS 2.2/2.3) clears 60–80% without a challenge.
Cybersource / Paytia · 2026The most important friction lever for UK e-commerce in 2026 is not payment-method choice but authentication design. Transaction Risk Analysis exemptions and rich-data 3DS routing determine whether a transaction clears frictionlessly or hits a challenge that loses up to a quarter of carts.
A “cross-border” transaction is one where the cardholder is UK-based and the merchant is incorporated outside the UK. The value share is higher than the volume share because cross-border averages a larger basket.
Steady 1-point-per-year drift upward post-Brexit. UK-to-EEA card transactions are now treated as cross-border by the schemes and attract higher interchange.
Chart pending sourced data. No UK primary source publishes the cross-border share of online card spend as a time series; the ONS Visa dataset reports the inverse (how much home and overseas spend is online). The underlying table has been requested from the ONS.
Post-Brexit, UK-to-EEA card transactions are treated as cross-border by the schemes and attract materially higher interchange. This is one reason marketplace acquirers route flows carefully, and a major focus of the PSR scheme-fee market review.
Open Banking is profiled in detail on the Payments page. For e-commerce specifically, retail checkout remains a minority of volume but the rail is maturing fast, and commercial Variable Recurring Payments launched in Q1 2026.
Open Banking reached 351 million payments in 2025 (+57% year on year), running at roughly 33 million a month by late 2025. Retail checkout is around 9% of that, up from under 5% in 2022. The UK Payments Initiative launched commercial Variable Recurring Payments ‘Wave 1’ in Q1 2026, the first time consumers can set up app-authorised recurring merchant payments outside pilots. Mainstream checkout adoption now depends on an Apple Pay-style pay-by-bank flow reaching the major platforms.
See the Payments page, Section 4 →Focused datasets in this series, each tracing its figures to named primary sources.
Every figure on this page maps to one of the primary sources below. Vendor surveys and platform-vendor “state of e-commerce” reports are excluded except where flagged as industry estimates.
| Source | Publisher | Period covered | Type | Last checked |
|---|---|---|---|---|
| Retail Sales Index · series J4MC | ONS | Monthly, to Apr 2026 | Primary government | 3 Jun 2026 |
| Global Payments Report 2026 | Worldpay (FIS) | 2025 (wallet/BNPL shares) | Industry survey | 3 Jun 2026 |
| Card Spending Update · e-commerce | UK Finance | To Feb 2026 | Industry body | 3 Jun 2026 |
| Annual Fraud Report 2025 · CNP | UK Finance | 2024 calendar year | Industry body | 3 Jun 2026 |
| Open Banking Impact Report | Open Banking Ltd | To Jan 2026 | Industry body | 3 Jun 2026 |
| AI commerce & cart benchmarks | Adobe Analytics / Baymard | 2025–26 | Industry estimate · flagged | 3 Jun 2026 |
The ONS publishes the authoritative online retail share; UK Finance the authoritative payment-method splits. Where industry vendors publish competing figures, we prefer ONS and UK Finance.
Where no official figure exists (cart abandonment, SCA drop-off), we cite the industry source by name and treat the number as directional, not authoritative.
“Mobile wallet” means an Apple Pay / Google Pay / Samsung Pay card-funded transaction. PayPal is reported separately even though some sources combine it.
Cardholder UK / merchant non-UK, by card-issuer perspective. Excludes UK cardholders abroad in person.
Headline figures map to named primary sources (ONS, UK Finance) or, where stated, the Worldpay Global Payments Report 2026 as a corroborating industry source. Supporting figures on mobile device share, wallet mix, agentic commerce and cart abandonment are trade-body or vendor estimates (UK Finance, Worldpay, Adobe Analytics, Baymard) and are sourced and labelled. Last full review: 3 Jun 2026.