Online shopping keeps growing and most of the fight is now at checkout, not at the front door. In the US, e-commerce was 17.1% of total retail sales in the second quarter of 2026, and about seven in ten carts are abandoned. The biggest documented reasons are costs that appear late, slow delivery, doubts about trust, forced account creation and a long checkout. All of these are fixable, and speed pays back: in a Vodafone test, a page with a 31% better Largest Contentful Paint produced 8% more sales. For most stores the best mobile investment is a fast, simple mobile site, and an app comes second.
Key takeaways
- The US Census Bureau reports Q2 2026 e-commerce sales of $340.2 billion, 17.1% of total retail sales and 12.2% higher than a year earlier (released 18 August 2026).
- Baymard's average documented cart abandonment rate is 70.22%, based on 50 studies (last updated 22 September 2025).
- Among the reasons shoppers gave Baymard: 40% extra costs too high, 20% delivery too slow, 19% did not trust the site with card details, 18% account creation was required, 17% checkout was too long or complicated.
- In a Vodafone A/B test, the version with a 31% better LCP gave 8% more sales, a 15% better lead-to-visit rate and an 11% better cart-to-visit rate.
- The European Accessibility Act has covered e-commerce services since 28 June 2025, which affects stores that sell to EU consumers.
How big is online shopping now?
The US Census Bureau's quarterly report for the second quarter of 2026 (released 18 August 2026) estimates e-commerce sales at $340.2 billion, an increase of 3.8% on the first quarter and 12.2% on the second quarter of 2025. E-commerce made up 17.1% of total retail sales. This is US data and is revised over time. It does not split phone from desktop, so the number that matters most for you is in your own analytics: look at the device breakdown for sessions, orders and revenue before deciding where to invest.
Where do mobile shoppers drop out?
Baymard Institute, which studies checkout usability, puts the average documented cart abandonment rate at 70.22%, from 50 studies. Setting aside people who were "just browsing", it lists these reasons from US shoppers:
| Reason for abandoning | Share of shoppers |
|---|---|
| Extra costs (shipping, tax, fees) too high | 40% |
| Delivery was too slow | 20% |
| Did not trust the site with their credit card information | 19% |
| The site wanted them to create an account | 18% |
| Checkout was too long or complicated | 17% |
| The site had errors or crashed | 17% |
| Returns policy was unsatisfactory | 13% |
| Could not see or calculate the total order cost upfront | 12% |
| Credit card was declined | 10% |
| Not enough payment methods | 9% |
Respondents could give more than one reason, so the figures do not add up to 100%. Each reason points to a fix, covered in the checklist below.
Does speed really change sales?
Yes, measurably. In a Vodafone test described on web.dev, paid traffic was split 50/50 between two landing pages that were visually and functionally identical, except that one was optimised for Web Vitals. The optimised version had a 31% better Largest Contentful Paint in field data and delivered 8% more sales, a 15% better lead-to-visit rate and an 11% better cart-to-visit rate. It is one company's test, but a controlled one. For how speed is measured in 2026, see Web development trends in 2026.
Should you build an app or improve your mobile site?
Start with the mobile site. It reaches every visitor, including those arriving from search and ads, and does not ask anyone to install anything. An app earns its place when people shop often, when you want loyalty features or push notifications, or when you need device features the browser lacks. If you do build one, Cross-platform or native in 2026 explains the options and the yearly store requirements. The 2017 question, "eCommerce sites or apps", has the same answer today: decide from how your customers behave, using your own data.
What should a mobile checkout do?
Use Baymard's reasons as a to-do list:
- Show the full cost early, including shipping and tax, so there are no surprises at the last step (40% and 12%).
- State delivery times honestly and offer a fast option if you can (20%).
- Build trust: clear business details, a visible returns policy and secure payment (19% and 13%).
- Offer guest checkout (18%).
- Cut steps and fields, and keep the form usable one-handed (17%).
- Test for errors on real phones and watch your error logs (17%).
- Offer the payment methods your customers use, and handle declined cards gracefully (10% and 9%).
More detail is in 11 hacks to increase the conversion rate of your eCommerce website.
What else is changing?
Accessibility. The European Accessibility Act has applied since 28 June 2025 and covers e-commerce services, with an exemption for microenterprises that provide services. Details differ by country and this is not legal advice. Button size, focus visibility and sign-in rules also matter on phones; see UX and UI design trends in 2026.
AI in shopping. People increasingly ask AI assistants what to buy, and some stores are preparing for agents that can shop on a customer's behalf. We cover how to get ready in Agentic commerce readiness and how assistants find products in AI product search compared for Shopify, Magento and WooCommerce.
A quick mobile commerce review
- Open your store on a phone you do not normally use, and buy something.
- Check your analytics: sessions, orders and revenue by device.
- List every cost that appears after the product page, and move it earlier.
- Count the steps and fields from cart to payment.
- Measure LCP and responsiveness for real visitors, not just in a lab test.
- Compare your abandonment reasons with the Baymard list.
We build and improve stores on platforms such as Shopify, WooCommerce, Magento, BigCommerce and OpenCart; see our eCommerce solutions page, or start with a Free Website Audit.
FAQ
What is a good cart abandonment rate?
Baymard's documented average is 70.22%, so a rate near that is typical. Compare yourself with your own past rate and your own device split, and look at the reasons, not only the number.
Is a mobile app better than a mobile website?
Not by default. A fast mobile site is the foundation, and an app suits frequent shoppers, loyalty features and notifications. Decide from your customers' behaviour.
Does page speed affect sales?
In Vodafone's controlled test, a 31% better LCP gave 8% more sales. Results vary by site, but speed is among the cheapest improvements to try.
Should we force customers to create an account?
Baymard found that 18% of shoppers abandoned because the site required account creation. Offer guest checkout, and invite an account after the purchase.
Do the EU accessibility rules apply to our store?
They can, if you sell to EU consumers and are not exempt as a microenterprise providing services. Rules are set country by country, so check with a lawyer in your market.
Sources
- US Census Bureau, "Quarterly Retail E-Commerce Sales, 2nd Quarter 2026" (released 18 August 2026): https://www.census.gov/retail/ecommerce.html
- Baymard Institute, "50 Cart Abandonment Rate Statistics 2025" (updated 22 September 2025): https://baymard.com/lists/cart-abandonment-rate
- web.dev, "Vodafone: A 31% improvement in LCP increased sales by 8%": https://web.dev/case-studies/vodafone
- Council of the EU, "Accessibility of goods and services": https://www.consilium.europa.eu/en/policies/accessibility-goods-services/
- EUR-Lex, "Accessibility of products and services" (summary of Directive (EU) 2019/882): https://eur-lex.europa.eu/EN/legal-content/summary/accessibility-of-products-and-services.html