Checkout Payment Statistics: Methods, Declines and Abandonment

By Rene Siller, Zobi. Sources checked 2026-11-11.

See what shoppers report about payment abandonment, how online payment methods vary by market, and what a randomized checkout test found.

## Quick answer

A missing payment option and a declined card can both cost a checkout, but they happen at different points. Baymard’s US research found that 10% reported a declined card and 9% reported too few payment methods in its published breakdown of reasons for cart abandonment. The ECB’s diary shows what people actually paid with. Stripe’s holdback experiment tests what happens when a relevant option is shown. Start with whichever of those questions you’re trying to answer.

Baymard Institute, reasons for abandoning a cart (https://baymard.com/blog/cart-abandonment-rate)

[See the source and calculation.](index.html#payment-abandonment-reasons) · [Download the statistics and sources.](downloads/statistics-with-sources.csv)

## What shoppers say when a payment sends them away

Trust, a missing payment option and a declined card point to different fixes. Baymard asked shoppers about abandoning carts. Its separate checkout benchmark looked at how sites present payment choices and respond to errors. The survey tells you what shoppers remember. The site benchmark tells you what teams can inspect.

### What payment problems do shoppers give for abandoning a cart?

In Baymard's US research, 19% reported distrusting the site with their card details, 10% reported a declined credit card and 9% reported too few payment methods. Trust was 10 percentage points ahead of limited choice in the same published breakdown.

US online shoppers in Baymard's abandonment research. Public analysis updated October 2026. Zobi calculation: 19% − 9% = 10 percentage points. The reason percentages can overlap. They are self-reports, without a denominator of payment attempts or a processor log.

Baymard Institute, reasons for abandoning a cart (https://baymard.com/blog/cart-abandonment-rate) · Baymard Institute, card trust (https://baymard.com/blog/cart-abandonment-rate) · Baymard Institute, declined credit card (https://baymard.com/blog/cart-abandonment-rate) · Baymard Institute, too few methods (https://baymard.com/blog/cart-abandonment-rate) · [Link to this statistic](index.html#payment-abandonment-reasons)

### How common are problems with the way sites offer payment methods?

Baymard found issues with offering payment options on 21% of sites in its ecommerce UX benchmark.

Sites in Baymard's ecommerce UX benchmark. Public analysis published June 2026. This is a site observation. It cannot establish what share of that site's shoppers abandoned checkout.

Baymard Institute, payment option presentation (https://baymard.com/blog/payment-ux) · [Link to this statistic](index.html#payment-choice-ux)

### How often does a validation error make shoppers enter details again?

Baymard reported that 34% of sites failed to retain entered data after a validation error.

Sites in Baymard's ecommerce UX benchmark. Public analysis published June 2026. The figure describes how sites handled validation errors. It gives no rate of errors or shopper abandonment.

Baymard Institute, validation errors (https://baymard.com/blog/payment-ux) · [Link to this statistic](index.html#payment-error-retention)

## What people pay with online

Figure description: Remote debit payments account for a bigger share of dollars. Share of transactions: 34.37%, Share of transaction value: 47.52%

Remote payments made up 13.15 percentage points more by value than by count. The two bars show shares of separate totals. Card-not-present includes remote channels beyond web checkout. Source: Federal Reserve, Table 2.

Payment diaries show methods people actually used. A euro-area average gives a starting point, then the country figures show where it can mislead. The US debit network data answers a narrower question about where debit spending happens, including payments away from physical terminals.

### What share of online payments are cards and e-payment solutions in the euro area?

In the ECB's 2024 euro-area diary, cards accounted for 48% of online payments and e-payment solutions accounted for 29%, a 19-point gap by transaction count.

Euro-area consumers in the ECB SPACE 2024 study. 2024. Zobi calculation: 48% − 29% = 19 percentage points. The ECB counts debit, credit and prepaid cards as cards. E-payment solutions include wallets, PayPal, other mobile apps and iDEAL in the Netherlands. Recurring bills and orders paid in person sit outside this online mix.

ECB SPACE, Chart 10 payment instrument shares (https://www.ecb.europa.eu/stats/ecb_surveys/space/html/ecb.space2024~19d46f0f17.en.html#toc10) · ECB SPACE, Chart 10 payment category definitions (https://www.ecb.europa.eu/stats/ecb_surveys/space/html/ecb.space2024~19d46f0f17.en.html#toc10) · [Link to this statistic](index.html#europe-payment-mix)

### Has the share of online payments made by card fallen in the euro area?

Yes. Cards accounted for 54% of euro-area online payments in 2019, 51% in 2022 and 48% in 2024. The 2024 share was 6 percentage points below 2019.

Euro-area consumers in ECB SPACE. 2019, 2022 and 2024. Zobi calculation: 54% − 48% = 6 percentage points. The finding tracks share. The number of online payments also changed, so the lower share cannot establish a drop in card transaction volume.

ECB SPACE, Chart 10 payment instrument shares (https://www.ecb.europa.eu/stats/ecb_surveys/space/html/ecb.space2024~19d46f0f17.en.html#toc10) · [Link to this statistic](index.html#europe-card-share-over-time)

### How much does e-payment use vary across the euro area?

The ECB recorded e-payment solutions for 76% of online payments in the Netherlands, 46% in Germany and 33% in Portugal in 2024. The Dutch share was 30 points higher than Germany's and 43 points higher than Portugal's.

Euro-area consumers in the ECB SPACE 2024 study. 2024. Zobi calculation: 76 − 46 = 30; 76 − 33 = 43 percentage points. The ECB uses a broad category, including iDEAL in the Netherlands. The country shares cannot tell a merchant which individual wallet will convert best.

ECB SPACE, Chart 11 shares by country (https://www.ecb.europa.eu/stats/ecb_surveys/space/html/ecb.space2024~19d46f0f17.en.html#toc10) · ECB SPACE, Chart 10 payment category definitions (https://www.ecb.europa.eu/stats/ecb_surveys/space/html/ecb.space2024~19d46f0f17.en.html#toc10) · [Link to this statistic](index.html#europe-online-country-gap)

### How often are online payments over €50 compared with in-person payments?

In the ECB's 2024 diary, 30% of euro-area online payments exceeded €50, compared with 12% of in-person payments. The online share was 2.5 times the in-person share.

Euro-area consumers in the ECB SPACE 2024 study. 2024. Zobi calculation: 30 ÷ 12 = 2.5. The chart includes peer-to-peer payments in both channels. Treat it as a payment-size comparison, without reading it as an average ecommerce order value.

ECB SPACE, Chart 12 online and point-of-sale value (https://www.ecb.europa.eu/stats/ecb_surveys/space/html/ecb.space2024~19d46f0f17.en.html#toc11) · ECB SPACE, Chart 12 peer-to-peer scope (https://www.ecb.europa.eu/stats/ecb_surveys/space/html/ecb.space2024~19d46f0f17.en.html#toc11) · [Link to this statistic](index.html#europe-online-payment-size)

### What share of US debit payments are card-not-present?

Card-not-present transactions made up 34.37% of the count and 47.52% of the value in the Fed's 2023 debit and prepaid network table. The share by value was 13.15 points higher.

US debit and general-use prepaid transactions reported by payment card networks to the Federal Reserve. 2023. Zobi calculation: 100 × (2213.79 ÷ 4658.42 − 34.61 ÷ 100.70) = 13.15 percentage points after rounding. Card-not-present includes remote channels beyond web checkout. The table cannot measure attempted checkouts or declines.

Federal Reserve, Table 2 all transactions (https://www.federalreserve.gov/paymentsystems/2023-interchange-fee.htm) · Federal Reserve, Table 2 card-not-present transactions (https://www.federalreserve.gov/paymentsystems/2023-interchange-fee.htm) · [Link to this statistic](index.html#us-debit-count-versus-value)

### Were card-not-present debit payments larger within each network type?

In the Fed's 2023 data, card-not-present debit and prepaid payments averaged $32.36 more than card-present payments on dual-message networks and $4.67 more on single-message networks.

US debit and general-use prepaid transactions reported by payment card networks to the Federal Reserve. 2023. Zobi calculation: $65.09 − $32.73 = $32.36; $48.01 − $43.34 = $4.67. These are transaction averages. They cannot establish that paying remotely causes a larger order.

Federal Reserve, Table 2 dual-message networks (https://www.federalreserve.gov/paymentsystems/2023-interchange-fee.htm) · Federal Reserve, Table 2 single-message networks (https://www.federalreserve.gov/paymentsystems/2023-interchange-fee.htm) · [Link to this statistic](index.html#us-debit-size-by-network)

## When an extra method helps

Stripe tested the same practical choice many merchants face: show a relevant extra method or leave it out. A holdback compares what happened when eligible shoppers saw different options. The market examples are a reason to test the methods your shoppers know, starting with a consistent checkout experience for returning visitors.

### Does showing another payment method improve checkout conversion?

Stripe reported 7.4% higher conversion and 12% higher revenue on average when its experiment dynamically showed at least one additional relevant payment method beyond cards.

Eligible businesses and checkouts in Stripe's Optimized Checkout Suite experiment. Experiment results published April 2025. Stripe measured an average across its eligible experiment. Results for a specific store depend on the shopper, method and mix of orders.

Stripe checkout experiment, aggregate results (https://stripe.com/blog/testing-the-conversion-impact-of-50-plus-global-payment-methods) · Stripe checkout experiment, randomization method (https://stripe.com/blog/how-we-tested-the-conversion-impact-of-global-payment-methods) · [Link to this statistic](index.html#stripe-extra-method-conversion)

### How much did conversion lift vary across Stripe's market examples?

Across six examples Stripe chose to report, relative conversion lift ranged from 5% for Cash App in US digital games to 91% for Alipay in China. The examples span 86 percentage points of relative lift.

Six payment method, country and industry examples selected by Stripe. Experiment results published April 2025. Zobi calculation: Maximum 91% − minimum 5% = 86 points across reported relative lifts. These examples have different eligibility and shopper mixes. The range describes the six reported results and should not be used to rank providers or predict a new rollout.

Stripe checkout experiment, six market examples (https://stripe.com/blog/testing-the-conversion-impact-of-50-plus-global-payment-methods) · Stripe checkout experiment, randomization method (https://stripe.com/blog/how-we-tested-the-conversion-impact-of-global-payment-methods) · [Link to this statistic](index.html#stripe-selected-market-range)

### Can a payment option change revenue and conversion by different amounts?

Yes. Stripe reported 13% relative revenue lift and 3% relative conversion lift when Revolut Pay was offered to eligible checkouts. That's a 10-point difference between the reported lifts.

Eligible businesses and checkouts in Stripe's Optimized Checkout Suite experiment. Experiment results published April 2025. Zobi calculation: 13% − 3% = 10 points between relative lifts. The two estimates measure different outcomes. Their gap cannot establish an average order value effect or explain why revenue moved.

Stripe checkout experiment, digital wallet results (https://stripe.com/blog/testing-the-conversion-impact-of-50-plus-global-payment-methods) · Stripe checkout experiment, randomization method (https://stripe.com/blog/how-we-tested-the-conversion-impact-of-global-payment-methods) · [Link to this statistic](index.html#stripe-wallet-metrics)

## Where a payment actually fails

Start with the last step the shopper reached. If they never submitted payment, check the visible options and form. If the attempt stopped during authentication, look at that step before you look for an issuer response. Record issuer declines and completed orders separately. If a shopper replies to a recovery message, ask which step stopped them and point them back to the secure checkout. Keep payment credentials out of the conversation.

### What online payment difficulty do European consumers report most often?

Across the euro area, 79% of respondents reported no online payment issues in ECB SPACE 2024. Entering payment information was the most commonly reported type of difficulty, cited by 7% of respondents.

Respondents answering the ECB's online payment difficulties questions. 2024. Respondents could name multiple difficulties. The source reports shopper experiences, without a transaction-level failure rate.

ECB SPACE, Box 2 technical difficulties (https://www.ecb.europa.eu/stats/ecb_surveys/space/html/ecb.space2024~19d46f0f17.en.html#toc15) · [Link to this statistic](index.html#europe-online-technical-difficulty)

### How do European consumers say they prefer to authenticate online payments?

In ECB SPACE 2024, 50% preferred credentials with an additional authentication step, compared with 24% who preferred credentials alone and 22% who chose biometrics. The first two options were 26 points apart.

Euro-area consumers in the ECB SPACE 2024 study. 2024. Zobi calculation: 50% − 24% = 26 percentage points. These are stated preferences. The result cannot establish the conversion effect of adding or removing an authentication step.

ECB SPACE, Box 3 authentication preferences (https://www.ecb.europa.eu/stats/ecb_surveys/space/html/ecb.space2024~19d46f0f17.en.html#toc16) · ECB SPACE, Box 3 alternate authentication options (https://www.ecb.europa.eu/stats/ecb_surveys/space/html/ecb.space2024~19d46f0f17.en.html#toc16) · [Link to this statistic](index.html#europe-authentication-preferences)

### What should you count before reporting a payment failure rate?

Stripe's payment-success measure uses submitted payment attempts. It can include failed 3D Secure, processor blocks and issuer declines. Its raw reporting also counts retries of the same purchase. A shopper who left before submitting payment is outside that payment-attempt denominator.

Payment attempts in Stripe Acceptance. Current Stripe documentation. Record whether the denominator is raw or deduplicated and keep it separate from checkout starts. Stripe's definitions do not supply a typical merchant decline rate.

Stripe, stages counted in payment success (https://docs.stripe.com/payments/analytics/acceptance) · Stripe, how raw rates count retries (https://docs.stripe.com/payments/analytics/acceptance) · [Link to this statistic](index.html#payment-success-denominator)

## How to use the comparisons

If shoppers leave before submitting payment, first check whether they can see a method they can use. Once they submit, separate validation, authentication, provider blocks and issuer responses. Track completed orders on their own. If you test a new option, choose a market and compare eligible sessions with a stable holdback.

## Sources, definitions and downloads

Baymard supplies US shopper self-reports and a separate ecommerce checkout benchmark. The ECB records payment methods in a diary and asks separate questions about difficulty and authentication. Federal Reserve data covers US debit and general-use prepaid transactions. Stripe reports a randomized checkout experiment and provides definitions for payment reporting. Calculated comparisons use the published figures and preserve each source’s denominator.

[Download the statistics and sources](downloads/statistics-with-sources.csv).

## Sources and measurement context

### What payment problems do shoppers give for abandoning a cart

Reported value: 10 percentage points.

Population: US online shoppers in Baymard's abandonment research

Measurement period: Public analysis updated October 2026

Denominator: Baymard's published breakdown after setting aside the browsing reason

Limitations: The reason percentages can overlap. They are self-reports, without a denominator of payment attempts or a processor log.

Read the original source

### How common are problems with the way sites offer payment methods

Reported value: 21 %.

Population: Sites in Baymard's ecommerce UX benchmark

Measurement period: Public analysis published June 2026

Denominator: Sites evaluated in the benchmark

Limitations: This is a site observation. It cannot establish what share of that site's shoppers abandoned checkout.

Read the original source

### How often does a validation error make shoppers enter details again

Reported value: 34 %.

Population: Sites in Baymard's ecommerce UX benchmark

Measurement period: Public analysis published June 2026

Denominator: Sites evaluated for payment-step UX

Limitations: The figure describes how sites handled validation errors. It gives no rate of errors or shopper abandonment.

Read the original source

### What share of online payments are cards and e-payment solutions in the euro area

Reported value: 19 percentage points.

Population: Euro-area consumers in the ECB SPACE 2024 study

Measurement period: 2024

Denominator: Number of non-recurring payments made online, under the ECB's diary definition

Limitations: The ECB counts debit, credit and prepaid cards as cards. E-payment solutions include wallets, PayPal, other mobile apps and iDEAL in the Netherlands. Recurring bills and orders paid in person sit outside this online mix.

Read the original source

### Has the share of online payments made by card fallen in the euro area

Reported value: 6 percentage point decline.

Population: Euro-area consumers in ECB SPACE

Measurement period: 2019, 2022 and 2024

Denominator: Online payment count in each survey wave

Limitations: The finding tracks share. The number of online payments also changed, so the lower share cannot establish a drop in card transaction volume.

Read the original source

### How much does e-payment use vary across the euro area

Reported value: 30 / 43 percentage points.

Population: Euro-area consumers in the ECB SPACE 2024 study

Measurement period: 2024

Denominator: Number of online payments within each country

Limitations: The ECB uses a broad category, including iDEAL in the Netherlands. The country shares cannot tell a merchant which individual wallet will convert best.

Read the original source

### How often are online payments over €50 compared with in-person payments

Reported value: 2.5 times.

Population: Euro-area consumers in the ECB SPACE 2024 study

Measurement period: 2024

Denominator: Payment count in each channel, including peer-to-peer transactions

Limitations: The chart includes peer-to-peer payments in both channels. Treat it as a payment-size comparison, without reading it as an average ecommerce order value.

Read the original source

### What share of US debit payments are card-not-present

Reported value: 13.15 percentage points.

Population: US debit and general-use prepaid transactions reported by payment card networks to the Federal Reserve

Measurement period: 2023

Denominator: All 100.70 billion transactions and $4,658.42 billion of transaction value in Table 2

Limitations: Card-not-present includes remote channels beyond web checkout. The table cannot measure attempted checkouts or declines.

Read the original source

### Were card-not-present debit payments larger within each network type

Reported value: 32.36 / 4.67 USD per transaction.

Population: US debit and general-use prepaid transactions reported by payment card networks to the Federal Reserve

Measurement period: 2023

Denominator: Transactions in each channel within each network type

Limitations: These are transaction averages. They cannot establish that paying remotely causes a larger order.

Read the original source

### Does showing another payment method improve checkout conversion

Reported value: 7.4 / 12 % relative lift in conversion and revenue.

Population: Eligible businesses and checkouts in Stripe's Optimized Checkout Suite experiment

Measurement period: Experiment results published April 2025

Denominator: Randomized treatment versus holdback in Stripe's published experiment

Limitations: Stripe measured an average across its eligible experiment. Results for a specific store depend on the shopper, method and mix of orders.

Read the original source

### How much did conversion lift vary across Stripe's market examples

Reported value: 86 percentage points across relative lifts.

Population: Six payment method, country and industry examples selected by Stripe

Measurement period: Experiment results published April 2025

Denominator: Each example compares the eligible treatment with its own holdback

Limitations: These examples have different eligibility and shopper mixes. The range describes the six reported results and should not be used to rank providers or predict a new rollout.

Read the original source

### Can a payment option change revenue and conversion by different amounts

Reported value: 10 percentage points between relative lifts.

Population: Eligible businesses and checkouts in Stripe's Optimized Checkout Suite experiment

Measurement period: Experiment results published April 2025

Denominator: Eligible Revolut Pay treatment and holdback, with each metric measured against its own baseline

Limitations: The two estimates measure different outcomes. Their gap cannot establish an average order value effect or explain why revenue moved.

Read the original source

### What online payment difficulty do European consumers report most often

Reported value: 79 / 7 % of respondents, as reported.

Population: Respondents answering the ECB's online payment difficulties questions

Measurement period: 2024

Denominator: Respondents in the ECB’s two Box 2 survey charts

Limitations: Respondents could name multiple difficulties. The source reports shopper experiences, without a transaction-level failure rate.

Read the original source

### How do European consumers say they prefer to authenticate online payments

Reported value: 26 percentage points.

Population: Euro-area consumers in the ECB SPACE 2024 study

Measurement period: 2024

Denominator: German respondents who generally pay online and respondents elsewhere who recorded an online payment in their diary

Limitations: These are stated preferences. The result cannot establish the conversion effect of adding or removing an authentication step.

Read the original source

### What should you count before reporting a payment failure rate

Reported value: Submitted attempts measurement base.

Population: Payment attempts in Stripe Acceptance

Measurement period: Current Stripe documentation

Denominator: Submitted payment attempts in the chosen reporting view, excluding invalid API calls

Limitations: Record whether the denominator is raw or deduplicated and keep it separate from checkout starts. Stripe's definitions do not supply a typical merchant decline rate.

Read the original source

## Publisher disclosure

Zobi publishes this research and provides shopper conversation services for Shopify stores.

Zobi · zobi.com