Cart Abandonment Statistics (2026)

A shopper adds a product, reaches shipping or payment, and disappears. Your analytics records an abandoned cart, but that label can cover very different moments: someone saving an item for later, someone reacting to delivery costs, or someone whose payment attempt failed.

The short answer: cart abandonment remains roughly 70% to 77% in the most widely cited public benchmarks, but there is no universal 2026 rate. The figures below combine a 50-study compilation, vendor-observed datasets, and shopper surveys. Use them to frame your own performance—not as interchangeable targets or proof that every cart exit is checkout friction.

Cart abandonment statistics at a glance

• 70.22% average documented cart abandonment — Baymard’s compilation of 50 ecommerce studies, last updated September 22, 2025. This is a multi-study average, not a single 2026 market sample.

• 77.38% global cart abandonment — Dynamic Yield’s displayed vendor benchmark from its customer base, which it describes as including 200 million monthly unique users across more than 300 million total sessions. The display refers to the past 12 months but does not publish a clear calendar period.

• 72.5% global cart abandonment — Vibetrace’s 2024 vendor-observed benchmark, based on 120 million monthly sessions across Vibetrace instances plus about 25 million monthly sessions from an anonymous partner-data source.

• 76.98% mobile abandonment in 2026 — Stripe’s published vendor benchmark, 12.2 percentage points higher than desktop in its data.

• 87.08% travel and airlines abandonment in 2026 — Stripe’s category benchmark, compared with 71.34% for apparel and 70.83% for electronics.

Do not average those percentages. They use different populations, event definitions, source compositions, and time windows. A precise-looking blended number would describe no published dataset.

Get the Mobile Testing Playbook Used by 800+ QA Teams

Discover 50+ battle-tested strategies to catch critical bugs before production and ship 5-star apps faster.

Email addressGet Instant Access

100% Free. No spam. Unsubscribe anytime.

What does cart abandonment measure?

Cart abandonment is the share of carts that do not become completed orders under a particular measurement definition. That definition can begin at cart creation, checkout initiation, payment attempt, or another funnel event. It can also include shoppers who were browsing, comparing prices, or deliberately saving a cart.

That is why an aggregate rate is a useful alert, not a diagnosis. A vendor’s observed session data answers a different question from a survey asking shoppers why they left. Neither is automatically a count of lost, high-intent orders.

Baymard’s reasons distribution illustrates the distinction. It excludes the “just browsing / not ready to buy” segment, then reports reasons among the remaining presented responses. In that distribution, 40% cite extra costs that were too high, 20% slow delivery, 19% distrust about credit-card information, 18% forced account creation, 17% a too-long or complicated checkout, and 17% website errors or crashes.

Those percentages are reported reasons, not mutually exclusive causal classifications of every abandoned cart. A shopper can experience more than one problem, and an observed cart exit does not reveal which one mattered.

Cart abandonment by device and industry

Mobile deserves its own baseline because the input method, screen size, connection quality, and checkout context differ from desktop. Stripe reports 76.98% mobile abandonment for 2026 and says that result is 12.2 percentage points above desktop in its benchmark. Stripe recommends segmenting performance by device, traffic source, and checkout stage rather than managing to one blended rate.

Vibetrace’s 2024 dataset shows a similar direction in a separate vendor population: 75% on mobile, 71% on tablet, and 68% on desktop. The agreement in direction is useful; it does not make either vendor’s level a universal target.

Barilliance’s 77.13% global average, 80.79% phone rate, and 73.93% desktop rate are explicitly historical 2019 context. They can show that mobile-desktop gaps are not new, but they should not become a 2026 forecast.

Industry benchmarks also vary substantially within Stripe’s 2026 vendor data:

A travel checkout can involve dates, inventory changes, passenger information, and fare rules that do not exist in a simple apparel order. Treat category benchmarks as context for investigation, not a license to accept avoidable friction.

Why shoppers say they abandon carts

Digital Commerce 360’s 2025 Ecommerce Conversion Report survey asked shoppers who said they had abandoned carts about their reasons. 30.1% said shipping costs made the total higher than expected, 30.1% wanted to save the cart for later, 26.6% said the order did not qualify for free shipping, and 9.3% said the website loaded too slowly.

The “save for later” response matters operationally. It means an abandoned-cart count includes at least some shoppers who did not intend to complete an order in that session. Recovery messaging may still be appropriate, but the event is not evidence that a checkout malfunctioned.

Cost and delivery are distinct from technical performance. A cart can be abandoned after a correct, fast page reveals an unattractive delivery price. Conversely, a shopper might tolerate a price but leave after an error. Your reporting should preserve those separate paths instead of assigning every exit to one generic abandonment bucket.

How much abandonment is technical friction?

There is no credible public percentage for the share of all abandoned carts caused by bugs, slow pages, or payment failures. The available figures are narrower, and their denominators are not compatible.

The WooCommerce and Stripe 2026 survey covered more than 1,400 shoppers across North America, Europe, and Asia-Pacific. It reports that three in four surveyed shoppers had abandoned a cart. Among abandoners, 63% cited checkout or payment issues, including 28% failed transactions, 21% too many steps, and 21% inability to use a preferred payment method. The article identifies network timeouts, routing errors, and missed retries as examples of soft failures behind transactions that do not complete.

That 28% is highly actionable, but its scope is essential: it is a survey response within checkout-related abandonment in that study. It is not a finding that failed transactions account for 28% of every cart created across ecommerce.

Spreedly’s State of Checkout 2025 offers an operational perspective rather than shopper behavior. Its survey of 500 senior US-based executives found that 17% cited network or service downtime among checkout failures and 12% cited outdated card credentials. It also says one in four companies reported losing more than $1 million annually at online checkout. These are executive-reported checkout-failure and business-impact figures, not observed cart events.

Baymard’s checkout-usability research provides a related UX signal: 17% of US online shoppers said they had abandoned an order during the past quarter because checkout was too long or complicated. Baymard’s testing further estimates that an average large ecommerce site could achieve a 35.26% conversion-rate increase through checkout-design improvements. That is a research estimate for large ecommerce sites, not a guaranteed reduction in abandonment or a market-wide conversion result.

How should you measure cart abandonment in your store?

Start by declaring the denominator. If your rate is completed orders divided by carts created, label it that way. If it begins at checkout initiation, call it checkout abandonment instead. Changing the start event can change the rate before the shopper sees a payment field.

Instrument a sequence that can distinguish intent, price friction, UX friction, and technical failure:

1. Cart creation

2. Checkout initiation

3. Shipping and delivery-cost display

4. Payment-method selection

5. Payment attempt and authorization result

6. Hard decline, soft failure, timeout, routing failure, and retry outcome

7. Validation and server errors

8. Page-load and interaction performance by device

9. Completed order

Then slice the same funnel by device, traffic source, new versus returning shopper, product category, and checkout stage. That lets you test a concrete hypothesis—for example, whether a mobile payment timeout increases exits after payment selection—rather than treating every unfinished cart as a generic conversion issue.

For mobile flows, test payment and confirmation paths on the devices and network conditions your buyers actually use. Quash’s mobile ecommerce statistics guide provides broader context on mobile commerce and checkout behavior; it does not supply a cart-abandonment benchmark for this report.

What public cart abandonment statistics cannot tell you

Public statistics establish that abandonment is high and that price, delivery, trust, checkout complexity, page speed, errors, and payments can all be associated with exits. They cannot identify the causal share of cart abandonment attributable to a particular bug in your store.

No first-party Quash telemetry, recurring bug-pattern dataset, customer-language dataset, or completed Quash experiment covers ecommerce cart abandonment, checkout defects, or page-speed abandonment. That absence is important: it would be misleading to present a Quash-derived percentage where none exists.

The missing evidence is an event-level dataset that connects a defined cart or checkout population to page performance, error states, payment outcomes, retries, and eventual completion under one denominator. Until a source publishes that kind of dataset across retailers, treat public technical-friction figures as scoped signals and investigate them against your own instrumentation.

Methodology and source notes

This report separates four kinds of evidence rather than presenting them as one market census:

1. Multi-study compilation: Baymard’s 70.22% average across 50 studies.

2. Vendor-observed benchmarks: Dynamic Yield, Vibetrace, Stripe, and historical Barilliance data.

3. Self-reported surveys: Digital Commerce 360, WooCommerce and Stripe, and Spreedly.

4. Research estimate: Baymard’s estimated conversion improvement from checkout-design changes.

Current, historical, observed, and stated figures remain separate throughout. Stripe’s 2026 category and device benchmarks should not be blended with Vibetrace’s 2024 results or Barilliance’s 2019 data. Likewise, a response such as “the website loaded too slowly” is useful shopper evidence, but it is not a controlled page-speed measurement.

Conclusion

The useful benchmark is not a single industry percentage. It is a clear view of where your own funnel leaks: before checkout, after delivery costs appear, at payment selection, during authorization, or after an error. Use the public cart abandonment statistics to set questions, then use consistently defined events to decide which problem is worth fixing first.