Download PDF
Cite this reportIntroduction
This report examines the measured performance of online retail in 2026 across five areas: market size and growth, marketplace concentration, checkout conversion, AI-driven discovery, and seller economics. It draws on government statistics, named institutional research and company disclosures, with every figure traced to its original publisher. The central finding is a reacceleration. United States eCommerce growth reached 9.8% year over year in the first quarter of 2026, close to double the rate recorded across the four preceding quarters, and measured by government survey rather than modeled forecast.
A second finding runs alongside it. Many of the sector's most-repeated statistics do not survive contact with the publishers they are attributed to. Figures circulate with the wrong publisher, the wrong reference year, or no traceable origin at all, and they accumulate authority through repetition. This report removes them. Where two credible publishers disagree, both appear with the methodological difference stated. Where no current figure exists for a claim, the report says so rather than substituting an unsourced estimate. The page is reviewed quarterly.
How big is the eCommerce market in 2026, and how fast is it growing?
$326.7 billion. United States retail eCommerce sales reached that figure in the first quarter of 2026, according to the Census Bureau release of May 18. The total represents growth of 9.8% year over year and accounts for 16.9% of all retail spending in the country. Growth across the four preceding quarters had run consistently near 5%. The rate nearly doubled.
This reacceleration is the most consequential figure in the report and among the least widely cited, having been released after most annual statistics compilations were already published.
Global market sizing is less settled, and the reason is definitional rather than factual. Statista Market Insights projects worldwide eCommerce at $3.88 trillion (f) for 2026, growing 6.5%, on a definition covering business-to-consumer physical goods and excluding B2B, digital media, travel services and resale. eMarketer's widely quoted $6.88 trillion (f) forecast for the same year applies a broader definition of retail eCommerce, which is the definition most trade press and vendor research adopt. The two figures measure different quantities rather than disagreeing about one. eMarketer's live forecast pages now carry a June 2026 vintage stamp on a revision held behind subscription, which means the published $6.88 trillion predates the current forecast run.
The scale of the wider market exceeds what either retail figure suggests. UNCTAD's first global e-commerce value database, released May 21, 2026, records business e-commerce sales of $28 trillion across 45 economies representing roughly three-quarters of global GDP, inclusive of B2B activity. Retail eCommerce is a slice of a considerably larger whole.
For operators, the practical implication is that United States demand strengthened through the first half of 2026 rather than softening, and planning assumptions built on mid-single-digit growth now sit a full quarter behind the measured data.
Which marketplaces control online sales, and how concentrated has the market become?
51.0%. Amazon and Walmart will together account for just over half of United States retail eCommerce in 2026, according to eMarketer's May forecast. This marks the first year the pair has exceeded 50%, against a combined 41.3% in 2019.
Positions across the rest of the marketplace sector diverge sharply. eBay reported first-quarter 2026 gross merchandise volume of $22.2 billion, an 18% increase, alongside 136 million active buyers. Etsy's active buyer count returned to sequential growth for the first time in two years, reaching 86.6 million in the same quarter, and the company agreed to sell Depop to eBay for $1.2 billion. Amazon Business disclosed $60 billion in annualized gross sales on July 21, against $35 billion at its previous confirmed disclosure.
TikTok Shop now operates under restructured ownership. The United States divestiture closed on January 22, 2026, with Oracle, Silver Lake and MGX each holding 15% and ByteDance retaining 19.9%, deliberately below the foreign-adversary threshold. eMarketer forecasts TikTok Shop United States sales of $23.41 billion (f) for 2026, a 48% increase, making it the fastest-scaling challenger to the two incumbents.
Country-level share data did not survive verification at the precision commonly published. Claims placing the United States at 38.2% of global eCommerce fail basic arithmetic: measured United States eCommerce sits near a fifth of the global total on any credible pairing of published figures, not two fifths. Country tables are largely subscription-held at eMarketer and estimated elsewhere with methodology variance the published precision does not support. The one country claim with solid primary backing is that China is the only major market where online sales have passed 50% of retail.
The concentration figure carries the clearest operational implication. With half of United States online retail transacting through two companies, marketplace presence functions as a channel-concentration decision before it functions as a growth strategy.
What is a good eCommerce conversion rate in 2026, and where does the remaining revenue go?
2.03%. The current benchmark eCommerce conversion rate, measured by IRP Commerce for June 2026, an increase of 9.74% against the same month a year earlier. Average order value stands at £127.06. Mobile accounts for 63.5% of sales.
Against that benchmark, 70.22% of carts are abandoned. Baymard Institute's figure, averaged across 50 separate studies, has held remarkably steady over time. The composition of the reasons beneath it has not.
Website errors and crashes nearly doubled as a stated reason for abandonment, rising from 9% to 17% between published waves. Slow delivery moved into second position at 20%. Distrust of the site with card details now ranks third at 19%, ahead of forced account creation, which fell to fourth. Checkout friction has migrated from form-filling toward trust and technical reliability, which is a materially different problem requiring different remediation.
Baymard also quantifies the opportunity. The average large eCommerce site stands to gain 35.26% in conversion rate from checkout design improvements alone.
Payment behavior shifted in parallel. Worldpay's Global Payments Report 2026, published April 20 across more than 40 markets, records digital wallets at 56% of global online spending and 33% of in-person spending. Within the United States, online payment value divides approximately 40% digital wallets, 32% credit and 16% debit. Buy now, pay later represents around 6% of United States eCommerce value and is forecast to reach $500 billion globally by 2030. PayPal reported 439 million active accounts and $486.4 billion in total payment volume on July 28, one day before this report published.
Four widely circulated checkout statistics did not survive verification and are absent from this page: a claimed $155 desktop to $112 mobile order-value split, which neither Adobe nor IRP Commerce publishes; a 2.1% mobile to 3.5% desktop conversion gap originating in data now five years old; an 85% mobile cart abandonment rate exceeding by roughly twelve points the highest figure recorded by any firm actually measuring device-level abandonment; and a pair of payment-preference figures attributed to J.P. Morgan Payments that the firm does not publish.
The operational implication follows directly from the reason list. Load testing and delivery-promise auditing now address a larger share of abandonment than checkout form length does.
Is AI sending buyers to stores yet, or does it remain largely prospective?
$20.57 billion. eMarketer projects United States consumers will spend that amount through AI platforms including ChatGPT, Perplexity and Gemini during 2026, representing 1.5% of United States eCommerce and close to quadruple the preceding year's total. The firm's longer-range forecast places the figure at $144 billion (f) by 2029, or 8.8% of all United States retail eCommerce.
A 1.5% share reads as marginal. The quality of that traffic is what distinguishes it.
Adobe Analytics, working from more than one trillion visits to United States retail sites, recorded AI-referred traffic converting 54% better than traffic from all other sources in May 2026. Visitors arriving from AI referrals spend 53% more time on site and view 23% more pages per visit. Twelve months earlier the same channel converted 38% worse than non-AI traffic. The performance gap did not narrow. It reversed.
Volume is expanding rapidly from a low base. AI referrals to United States retail sites rose 138% year over year in May 2026, and 1,324% since Adobe began tracking the channel. Shopify reported orders originating from AI-powered searches growing approximately thirteen-fold year over year in the first quarter of 2026, the fastest-growing channel the company has measured. Amazon attributed roughly $12 billion in incremental annualized sales to its Rufus assistant, on the company's own modeled attribution.
The countervailing evidence is equally documented. OpenAI deprioritized in-chat Instant Checkout in March 2026, relocating the capability into Apps. Forrester identified approximately 30 Shopify merchants live on the feature in February, against the more than one million anticipated at launch. Forrester's mid-2026 assessment publishes no adoption figures at all and concludes that very few consumers permit agents to complete purchases without direct oversight. Agents are influencing purchase decisions at scale. They are not yet executing many of them.
Social commerce remains the larger and slower-moving channel, reaching $2.11 trillion (f) globally in 2026 according to Mordor Intelligence, with North America the fastest-growing region at a 32.11% compound annual growth rate through 2031.
The resulting position for operators is an unusual one. AI referral volume remains small enough to exclude from a revenue forecast, while the traffic quality is strong enough that excluding it from site architecture decisions would be a measurable error.
Is eCommerce still profitable for a small seller in 2026?
$849.9 billion. The most recent National Retail Federation and Happy Returns measurement values United States retail merchandise returns at that figure annually. The online return rate stands at 19.3%, against 15.8% across all retail. Approximately 9% of returns are fraudulent, representing around $76.5 billion. Appriss Retail's independent measurement, applying a different methodology, values total returned merchandise at $706 billion with $100 billion attributed to fraud and abuse. The two measurements differ in magnitude and agree in direction. This report anchors on NRF, the benchmark most retail analysts cite, and shows both.
Returns are the line item that determines whether a small store's margin survives operational reality.
The dropshipping market has no agreed size, and the disagreement is more informative than any single estimate. Global Market Insights places the 2026 market at $343 billion. Precedence Research places the same market in the same year at $537.81 billion, a gap of nearly $200 billion. Any single dropshipping market figure published without naming the research firm and the edition is functionally a coin flip.
Fulfillment economics rest on firmer ground. Grand View Research values the eCommerce fulfillment services market at $140.07 billion on its most recent update, projecting $272.14 billion by 2030 at a 14.2% compound annual growth rate. The frequently repeated claim that 60% of online retailers use third-party logistics does not withstand verification: it traces through a 3PL vendor to an aggregator naming no source. The only primary survey located points the opposite direction, finding 70% of brands still fulfilling in-house, with outsourcing crossing 50% only above $50 million in revenue.
Platform share figures attributed to BuiltWith could not be independently verified, with reported WooCommerce share varying between 18.2% and 39% depending on the citing source. This report substitutes the HTTP Archive crawl, which is reproducible: WooCommerce at 6.64% of the measurable web against Shopify at 4.76%. Among the top one million sites the ordering reverses, Shopify at 6.08% against WooCommerce at 3.12%.
SaleHoo first-party seller community data, labeled as such: approximately 80% of small stores generate under $1,000 per month. This reflects the structure of a low-barrier channel rather than a market failure, and it explains why returns and fulfillment cost exert more influence on outcomes at small scale than market-size headlines do.
Conclusion
Three findings emerge clearly from the verified data.
United States eCommerce growth reaccelerated in early 2026. First-quarter growth reached 9.8% year over year, close to double the rate of the four preceding quarters, and it is measured by government survey rather than modeled forecast.
Checkout friction changed composition without changing scale. Site errors nearly doubled as a stated abandonment reason, and delivery speed and payment trust now outrank forced account creation. The 70.22% headline held steady while its components moved underneath it.
AI-referred shopping is small and unusually high quality. It represents 1.5% of United States eCommerce in 2026 while converting 54% better than all other traffic, from a base low enough that percentage growth rates carry limited information on their own.
"If I could go back in time and give myself one piece of advice in the early days of my companies, I'd tell myself to gather more data."
— Simon Slade, CEO and co-founder, SaleHoo
Next update: November 2026, following the Census Bureau's Q2 2026 release on August 18 and eMarketer's midyear forecast revision.