Omnichannel statistics at a glance
| Category | Statistic | Source |
|---|---|---|
| Channel use | 77% of consumers shop across three to four channels; more than 20% use five or more | Klaviyo, 2025 |
| Channel use | 73% of 46,000 shoppers used more than one channel; 20% used only a store, 7% only online | Harvard Business Review, 2017 |
| Channel use | Social media users each visit 6.75 platforms a month, out of 5.66 billion users | DataReportal, Digital 2026 |
| Store vs online | E-commerce is 17.1% of US retail sales, up from 16.3% a year earlier | US Census Bureau, Q2 2026 |
| Store vs online | Internet sales are 27.3% of UK retail sales | ONS, August 2026 |
| Store vs online | 78% of EU internet users bought something online in the last 12 months | Eurostat, 2025 |
| Store vs online | 72% of 18,000 consumers across 23 countries still shop in physical stores | IBM Institute for Business Value, January 2026 |
| Devices | Mobile took 56.4% of US online holiday transactions, up from 54.5% | Adobe, January 2026 |
| Fulfilment | Curbside pickup was 17.1% of online orders at retailers offering it, peaking at 39.0% on 23 December | Adobe, January 2026 |
| AI as a channel | 45% of consumers use AI somewhere in the buying journey; 41% use it to research products | IBM Institute for Business Value, January 2026 |
| Marketing | Automated messages are 2% of email sends and 30% of email revenue | Omnisend, 2026, vendor data |
| B2B | A typical B2B buying decision now involves 13 internal stakeholders and 9 external influencers | Forrester, January 2026 |
| Experience | 35% of customers would use an AI agent rather than a person to avoid repeating themselves | Salesforce, 2024 |
How many channels do shoppers use?
Most shoppers use three or more channels before they buy: Klaviyo's 2025 survey of 2,000 consumers in the US, UK, Australia and New Zealand found 77% shopping across three to four channels and more than 20% using five or more.
| Statistic | Source |
|---|---|
| 77% of consumers shop across three to four channels and more than 20% use five or more, from 2,000 consumers in the US, UK, Australia and New Zealand. | Klaviyo, 2025 |
| 61% of those consumers are actively subscribed to between three and eight brands. | Klaviyo, 2025 |
| Boomers pick email (44%) and the store (27%) as their preferred way to hear from a brand. Gen Z picks SMS (14%) and branded apps (12%). | Klaviyo, 2025 |
| The average social media user visits 6.75 platforms a month, across 5.66 billion social users and 6.04 billion internet users. | DataReportal, Digital 2026 |
| 91% of US adults own a smartphone and 16% are smartphone-only internet users, from 5,022 adults surveyed February to June 2025. | Pew Research Center, 2025 |
| Mobile took 49.36% of worldwide web page views in August 2026, desktop 49.11% and tablets 1.54%. | StatCounter, August 2026 |
Channel-count numbers are the softest data on this page and the most quoted. They are self-reported, and people forget the Instagram ad they glanced at three weeks ago. What I trust from Klaviyo is the shape, not the 77%: almost nobody buys on one channel any more, and one default channel misses half your list.
What is the difference between multichannel and omnichannel?
Multichannel means a brand runs several channels that each keep their own record of the customer; omnichannel means those channels share one record, so a shopper can start in one and finish in another without repeating themselves. The channel count is the same in both. What changes is whether the customer's context travels with them.
The practical test: if a customer emails support, then calls, and has to explain the problem twice, that brand is multichannel no matter how many channels it runs. Salesforce's 2024 survey of 15,015 consumers and 1,570 business buyers in 18 countries found 35% would rather deal with an AI agent than a person if it meant not repeating themselves, rising to 47% among business buyers.
| Statistic | Source |
|---|---|
| 35% would work with an AI agent instead of a person to avoid repeating themselves: 34% of consumers, 47% of business buyers. | Salesforce, 2024 |
| 73% say most companies treat them as a unique individual rather than a number, up from 39% in 2023. | Salesforce, 2024 |
| 43% stopped buying from a brand in the past year after poor customer service, the second most common reason after high prices. | Salesforce, 2024 |
I have never seen a brand fail at omnichannel because it was short a channel. They fail because the data model is per channel, so the loyalty app, the till and the help desk each hold a different version of the same person. Salesforce's 47% is the number to take to your engineering lead, and the marketing attribution statistics page covers what the same fragmentation does to measurement.
How much retail still happens in a store?
Physical stores still take the large majority of retail spending: e-commerce was 17.1% of US retail sales in Q2 2026 and internet sales were 27.3% of UK retail sales in August 2026.
| Statistic | Source |
|---|---|
| E-commerce was 17.1% of total US retail sales in Q2 2026, up from 16.3% in Q2 2025, seasonally adjusted. | US Census Bureau, August 2026 |
| Internet sales were 27.3% of UK retail sales in August 2026, against 26.1% a year earlier, peaking at 32.4% in November 2025. | Office for National Statistics, series J4MC |
| 78% of EU internet users aged 16 to 74 bought online in the previous 12 months in 2025, out of the 95% who used the internet at all. | Eurostat, extracted February 2026 |
| Among EU online buyers, 34% made one or two purchases in three months and 33% made three to five. | Eurostat, 2025 |
| 72% of consumers still shop in physical stores, from 18,000 consumers across 23 countries surveyed in Q3 2025. | IBM Institute for Business Value and NRF, January 2026 |
| Indoor mall visits rose 1.8% and visit durations 3.3% in the first half of 2025 against the first half of 2024. | NRF, November 2025 |
The most useful number here is the boring ratio. At 17.1% online, US shoppers still spend $4.85 offline for every $1 they order online; the UK, ten points ahead, still spends $2.66. Every omnichannel programme I have seen budget as though the store were a legacy cost centre is arguing with that ratio and losing.
Do omnichannel customers really spend more?
Omnichannel customers do spend and return more in the published studies, but the evidence is correlational and the most cited source is nine years old. Harvard Business Review's January 2017 study of 46,000 shoppers, covering purchases from June 2015 to August 2016, found that 73% used more than one channel, 20% used only a store and 7% only online, and that shoppers who had an omnichannel experience made 23% more repeat trips to the retailer's stores in the following six months.
| Statistic | Source |
|---|---|
| 73% of 46,000 shoppers used more than one channel, in purchases made between June 2015 and August 2016. Dated. | Harvard Business Review, January 2017 |
| 20% of the same shoppers used only a physical store and 7% used only online channels. Dated. | Harvard Business Review, January 2017 |
| Shoppers who had an omnichannel experience logged 23% more repeat shopping trips to the retailer's stores within six months. Dated. | Harvard Business Review, January 2017 |
| Omnichannel shoppers spent 4% more per trip in the store and 10% more online than single-channel shoppers, after controlling for shopping experience. Dated. | Harvard Business Review, January 2017 |
That study sits in roughly every omnichannel deck ever built, usually without its date or its sample, and here is what nobody puts on the slide: omnichannel shoppers are self-selected. People who browse your app, visit your store and open your email already like you and already buy more. The 2017 study compared the 73% who used several channels with a 27% residual of light buyers. Adding a channel to a casual shopper does not turn them into a heavy one; heavy shoppers collect channels on the way. Treat every "omnichannel customers spend X% more" claim as a description of your best customers, not a recipe for making more.
Which channels drive orders, and what do multi-channel campaigns deliver?
Adobe's holiday data gives the cleanest public read on channel mix: US shoppers spent $257.8 billion online between 1 November and 31 December 2025, up 6.8%, with mobile taking 56.4% of online transactions.
| Statistic | Source |
|---|---|
| US online holiday spending reached $257.8 billion from 1 November to 31 December 2025, up 6.8%, measured across more than 1 trillion visits to US retail sites and 100 million SKUs. | Adobe Analytics, January 2026 |
| Mobile took 56.4% of online transactions, up from 54.5% in 2024. | Adobe Analytics, January 2026 |
| Curbside pickup was 17.1% of online orders at retailers offering it, down from 17.5% in 2024, and peaked at 39.0% on 23 December. | Adobe Analytics, January 2026 |
| Social media drove 4.6% of holiday online revenue, up from 3.3%. Affiliates and partners drove 20.4%, up from 17.6%. | Adobe Analytics, January 2026 |
| Traffic to US retail sites from AI tools grew 693.4% year over year over the holiday season. | Adobe Analytics, January 2026 |
| Buy now pay later took $20 billion of holiday spend, up 9.8%, and 82.2% of it was placed on a smartphone. | Adobe Analytics, January 2026 |
| US households planned to spend $1,595 on the holidays, down 10% from 2024, in the 40th Deloitte Holiday Retail Survey. | Deloitte, October 2025 |
The best public marketing numbers come from vendors, so read them as vendor data, not industry benchmarks. Omnisend's 2026 report covers 150,000 brands, 27 billion emails, 321 million SMS and 458 million push notifications sent in 2025.
| Statistic | Source |
|---|---|
| Ecommerce email open rates rose to 30.7% in 2025 from 26.6% in 2024. Vendor data, 150,000 brands. | Omnisend, 2026 |
| Automated messages were 2% of all email sends and produced 30% of email revenue. Vendor data. | Omnisend, 2026 |
| An automated email earned $2.87 per send against $0.18 for a scheduled campaign, a 16x gap. Vendor data. | Omnisend, 2026 |
| Push notification click-to-conversion rose to 22.9% in 2025 from 13.9% in 2024. Vendor data. | Omnisend, 2026 |
| Campaigns using three or more channels recorded a 494% higher order rate than single-channel campaigns, across 135,000 campaigns and 610 million messages sent in 2021. Dated vendor data. | Omnisend, 2022 |
The 494% figure is everywhere, and it is five years old, from one vendor's own customers. It carries the same selection problem as the retail studies: brands running three-channel campaigns have bigger lists to begin with. The Omnisend numbers I would use are the automation ones, because 2% of sends producing 30% of revenue is a claim about timing, and timing you can copy.
How is B2B omnichannel different from retail omnichannel?
B2B omnichannel is about a buying group, not a shopper: Forrester's State of Business Buying 2026 reports that a typical business buying decision involves 13 internal stakeholders and 9 external influencers. Retail statistics describe one person moving between channels; B2B statistics describe twenty-two people moving between channels and sending each other screenshots.
| Statistic | Source |
|---|---|
| A typical B2B buying decision includes 13 internal stakeholders and 9 external influencers, and 94% of buyers in groups of six or more report clear benefits. | Forrester, January 2026 |
| More than 60% of business buyers now use a trial before purchase, rising to 78% of buyers spending $10 million or more. | Forrester, January 2026 |
| Procurement professionals are decision-makers in 53% of business buying cycles, engaging from the start. | Forrester, January 2026 |
| 47% of business buyers would use an AI agent rather than a person to avoid repeating themselves, against 34% of consumers. | Salesforce, 2024 |
Consumer and B2B omnichannel numbers are not interchangeable, and mixing them is the most common error in this category: a statistic about 46,000 retail shoppers says nothing about a procurement committee. The measurement problem that creates is covered on the dark funnel statistics page.
The B2B lesson in Forrester's numbers is not "add channels". A trial, used by more than 60% of buyers, has quietly become the main channel, and procurement sits in more than half of cycles from day one. If your plan is a nurture sequence and a chat widget, it is aimed at the wrong twenty-two people, and lead response time matters more than channel count.
How is AI becoming another channel?
AI assistants are now a measurable shopping channel: 45% of consumers use AI somewhere in the buying journey, in IBM's survey of 18,000 consumers across 23 countries, fielded Q3 2025 and published with NRF in January 2026.
| Statistic | Source |
|---|---|
| 45% of consumers turn to AI for help during their buying journeys, from 18,000 consumers in 23 countries. | IBM Institute for Business Value and NRF, January 2026 |
| Consumers use AI to research products (41%), interpret reviews (33%) and hunt for deals (31%). | IBM Institute for Business Value and NRF, January 2026 |
| 29% of consumers want effortless purchasing through social platforms. | NRF and IBM, January 2026 |
Adobe's 693.4% looks enormous because the base was tiny; AI referrals are still a rounding error next to paid search or email. IBM's 41% is the figure that matters, because product research is where preference gets formed, and a channel that shapes the shortlist without sending a click gets recorded by most stacks as direct traffic.
How we calculated the original numbers
Four figures on this page do not appear anywhere else. Here is the arithmetic, so you can check it or change the inputs.
- At least 97% of consumers shop across three or more channels. Klaviyo's 2025 survey of 2,000 consumers reports 77% across three to four channels and more than 20% across five or more. The second band is open-ended, so the two add to 97% or higher and under 3% are left on one or two channels. Self-reported, so read 97% as a floor.
- $4.85 of offline retail spending for every $1 ordered online in the US. E-commerce was 17.1% of US retail sales in Q2 2026, so the remaining 82.9% divided by 17.1% is 4.85. The same sum on the UK's 27.3% internet sales share gives $2.66. The two agencies define online sales differently, so compare ratios, not decimals.
- 16x, not 21x: what an automated message is worth against a scheduled one. Omnisend reports $2.87 per automated email send against $0.18 for a scheduled campaign, which is 15.9x. Its share figures, 2% of sends and 30% of revenue, imply 21x, but only because that 2% is rounded: the per-send dollars put automated sends nearer 2.6%. Use the lower number.
- 27%: the size of the comparison group in the 2017 study. Harvard Business Review's 46,000-shopper study split into 73% multi-channel, 20% store-only and 7% online-only, so the single-channel group it measured against was 27% of the sample. When the omnichannel group is the majority and the control is a quarter of light buyers, a spending gap describes customer value, not causation.
What these omnichannel numbers add up to
Omnichannel is the default, so it is not a strategy. When at least 97% of shoppers cross three or more channels and 73% did so back in 2015, adding a channel is table stakes. XX of the customer, which is an engineering decision more than a marketing one.
The store is still the majority channel. 17.1% of US retail and 27.3% of UK retail happens online, so any plan that treats the store as a declining side channel is arguing with two national statistics agencies. The work worth doing is instrumenting the store, not replacing it.
The spending gap is mostly selection. Every headline claim that omnichannel customers spend more compares self-selected heavy buyers with a residual of light ones. I would still build for omnichannel, because 43% of customers walk after one bad service experience. Build it for the experience, then prove the revenue with a holdout on your own data.
FAQ
What percentage of shoppers use more than one channel?
At least 97%, computed from Klaviyo's 2025 survey of 2,000 consumers, which found 77% using three to four channels and more than 20% using five or more. Harvard Business Review's 2017 study of 46,000 shoppers put multi-channel use at 73% for purchases made in 2015 and 2016.
What is the difference between multichannel and omnichannel?
Multichannel means a brand runs several channels that each hold their own record of the customer; omnichannel means those channels share one record, so a customer can move between them without repeating themselves. Salesforce's 2024 survey of 15,015 consumers and 1,570 business buyers found 35% would rather deal with an AI agent than a person to avoid retelling their story.
Do omnichannel customers spend more than single-channel customers?
They spend and return more in the published studies, but the relationship is correlational. Harvard Business Review's 2017 study of 46,000 shoppers found 23% more repeat store trips within six months after an omnichannel experience, comparing a 73% multi-channel majority with a 27% residual of lighter buyers. Heavy buyers collect channels; channels do not create heavy buyers.
What share of retail sales happens online in 2026?
E-commerce was 17.1% of US retail sales in Q2 2026, up from 16.3% a year earlier (US Census Bureau), and internet sales were 27.3% of UK retail sales in August 2026 (Office for National Statistics). Eight in ten US retail dollars and seven in ten UK retail dollars are still spent offline.
How many people are involved in a B2B buying decision?
A typical B2B buying decision involves 13 internal stakeholders and 9 external influencers, in Forrester's State of Business Buying 2026 (January 2026). More than 60% of business buyers use a trial before purchase, and procurement professionals are decision-makers in 53% of cycles.
Is AI a real shopping channel yet?
Yes, for research rather than checkout. 45% of consumers use AI somewhere in the buying journey and 41% use it to research products, in IBM's survey of 18,000 consumers across 23 countries published with NRF in January 2026. Adobe measured AI-tool referral traffic up 693.4% year over year in the 2025 holiday season, from a small base.
Sources
- Harvard Business Review, A Study of 46,000 Shoppers Shows That Omnichannel Retailing Works (January 2017)
- Klaviyo, Online Shopping Report (2025)
- US Census Bureau, Quarterly Retail E-Commerce Sales (Q2 2026, released August 2026)
- Office for National Statistics, Internet sales as a percentage of total retail sales, series J4MC (August 2026)
- Eurostat, E-commerce statistics for individuals (2025 data, extracted February 2026)
- Adobe, Holiday Shopping Season Drove a Record $257.8 Billion Online (January 2026)
- IBM Institute for Business Value and NRF, Brands and Retailers Navigate a New Reality as AI Shapes Consumer Decisions (January 2026)
- NRF and IBM, Own the agentic commerce experience (January 2026)
- NRF, 10 trends and predictions for retail in 2026 (November 2025)
- Forrester, The State of Business Buying 2026 (January 2026)
- Salesforce, State of the AI Connected Customer, 7th edition (2024)
- Omnisend, 2026 Ecommerce Marketing Report (2026, covering 2025)
- Omnisend, Omnichannel Statistics (2021 campaign data, published 2022)
- Deloitte, 2025 Holiday Retail Survey (October 2025)
- DataReportal, Digital 2026 Global Overview Report (Digital 2026 edition)
- Pew Research Center, Mobile Fact Sheet (2025)
- StatCounter, Desktop vs Mobile vs Tablet Market Share Worldwide (August 2026)
Methodology. Numbers were collected in September 2026 and checked against the original publisher's page, never a statistics roundup. Vendor data from Omnisend and Klaviyo is labelled as such, with its stated sample and period. Figures from the 2017 Harvard Business Review study are marked dated and always shown with their 46,000-shopper sample and their June 2015 to August 2016 window. The US Census Bureau e-commerce page blocks automated requests, so its Q2 2026 figure was cross-checked against the FRED copy of the same series. Retail and B2B statistics sit in separate sections because their samples describe different populations. More sourced number sets are in the statistics hub. Last updated 21 September 2026.