“Merchants can no longer view the checkout button as the end of the shopping experience, but instead must treat it as the start of the payment process.”
This quote reframes everything.
When a user hits “checkout,” the emotional purchase decision is already made — they want what you’re selling. Your only job now is to not get in their way.
And yet, 70% of users abandon their carts before completing purchase.
That’s not a UX problem. That’s a $260 billion problem (the estimated annual cost of checkout abandonment in e-commerce).
So why — after two decades of online shopping — are we still losing customers at the moment they’re trying to give us money? (source.) Why is there still friction when someone is trying to give you their money?
I investigated this through five lenses:
- Why familiarity wins — The psychology behind standardized checkout patterns
- Where sites still fail — Baymard’s research on the most common missteps (with failure rates)
- Who gets it right — Allbirds, Peloton designs analyzed
- The global complexity — Why your checkout may be failing international users
- Subscription checkout — Different rules for recurring revenue
Why do checkout flows feel the same?
When you hand cash to a barista, the transaction is instantaneous. Online checkout stretches this into a multi-step process that requires users to manually input information to a backend system that needs to process your payment.
Over 25 years, consistent checkout patterns have created deeply embedded mental models. Users don’t just prefer familiar checkout flows — they expect them.
This is Jakob’s Law in action: Users spend most of their time on other sites, so they expect your site to work like all the others.
Familiarity isn’t a design constraint. It’s a trust signal. Innovation that disrupts checkout patterns creates cognitive friction — and friction converts directly to abandonment.
Despite these patterns being well-established, mistakes can and still do happen more often than you think.
If they feel the same, what are the design missteps?
In 2021, the Baymard Institute released “The Current State of Checkout UX” where they said:
“During Baymard’s 12 years of large-scale checkout usability testing we have also consistently found the checkout design and flow to frequently be the sole cause for users abandoning their cart during the checkout flow.”
Baymard’s research reveals a troubling pattern: most e-commerce sites fail basic checkout guidelines — not edge cases, but fundamentals.
Here are the highest-impact failures, grouped by what they cost users:
Friction at Account Creation
(Guideline > Failure Rate > User Cost)
Guideline: Guest checkout not prominent
- Failure rate: 65%
- User cost: Forced account creation when user just wants to buy
Guideline: Complex password requirements
- Failure rate: 80%
- User cost: Signup abandonment
Guideline: Required vs. optional fields unclear
- Failure rate: 85%
- User cost: Form completion anxiety
Friction at Payment
(Guideline > Failure Rate > User Cost)
Guideline: Credit card field doesn’t auto-format
- Failure rate: 40% failure rate
- User cost: Manual formatting errors
Guideline: Card sequence doesn’t match physical card
- Failure rate: 38% failure rate
- User cost: Cognitive load during high-anxiety moment
Guideline: Expiration date field formatted incorrectly
- Failure rate: 50%
- User cost: Input errors
Friction at Shipping
(Guideline > Failure Rate > User Cost)
Guideline: “Delivery speed” instead of “Delivery date”
- Failure rate: 36% failure rate
- User cost: User must calculate arrival date
Guideline: Store pickup not in shipping selector
- Failure rate: 63%
- User cost: Missed option for faster fulfillment
Pattern insight: Each failure forces users to do mental work that doesn’t serve their goal. The checkout flow should reduce cognitive load, not add to it.
Services like Stripe and Affirm have emerged over the years to make some of these mistakes go away.
However, as I dived deeper into what’s behind these issues…the ways we pay nowadays has become so varied that it puts further stress on the need for checkout flows to adapt to user expectations.
For instance, here’s how we pay for things across the world:
- Credit cards: especially popular in the U.S.
- Debit cards: also popular but are only good in “domestic” transactions
- Bank transfers: can be done via PayPal, but problematic if it’s an international transaction
- eWallets: think Alipay, Wechat Pay, Venmo; more options are emerging and growing fast across the world
- Cash: can still happen based on user preference, and/or limited access to the options mentioned above
Between usability issues and the evolving ways we pay for the things online, it highlights why checkout flows are in need of constant optimization so they can best meet your customer needs.
Who is doing it well?
This blog contains multiple examples of companies that have optimized their checkout flows to meet their customer expectations though I will highlight 2 of my favorites:
Allbirds: Reducing Decisions at the Moment of Commitment
Image: Allbirds checkout. View on Medium ↗
What they got right:
- Express checkout prominence — Payment options (Shop Pay, Amazon Pay, PayPal) appear before the traditional form. This isn’t just convenience — it’s recognition that returning users have already stored payment info elsewhere. Allbirds is borrowing trust from established platforms.
- Split-view architecture — Payment method (left) and order summary (right) create a clear information hierarchy. Users can verify their cart without leaving the payment context.
- Color restraint serves function — By limiting color to the primary CTA (“Continue to Shipping”), Allbirds eliminates visual competition. The user’s eye goes exactly where it needs to go.
- Trust signals without clutter — Refund policy link in footer is available but doesn’t add visual noise. It’s there for users who need reassurance; invisible to those who don’t.
Design principle: Allbirds optimizes for the confident buyer while providing escape routes for the hesitant buyer.
Peloton: High-Consideration Purchase, High-Transparency Checkout
Image: Peloton checkout. View on Medium ↗
Peloton faces a unique challenge: their checkout is the final step in a $2,500+ decision. Their flow demonstrates how checkout UX must adapt to purchase psychology.
What they got right:
- Cart summary prominence — For high-ticket items, buyers need constant reassurance they’re getting what they expect. The detailed summary (Bike+ Basics Package, membership terms) isn’t redundant — it’s anxiety reduction.
- “Estimated delivery by” vs. shipping speed — This directly addresses Baymard’s guideline. Users care about when it arrives, not how fast you ship it.
- Gift checkbox — Small detail, massive insight. Peloton knows their buyer isn’t always their user. This single checkbox acknowledges a entire use case (gift-giving) without adding friction for self-purchasers.
- Commitment anxiety mitigation — Two features work together…100-day home trial which reduces perceived risk (“I can return it if I hate it”)…and…Affirm financing which reduces immediate financial pain (“I don’t have to pay $2,500 today”)
Design principle: For high-consideration purchases, checkout UX must address commitment anxiety, not just transaction completion.
The Global Checkout Problem: Familiarity is Relative
Everything I’ve discussed so far assumes a US-centric mental model. But “familiar” varies dramatically by region — and a checkout flow that works in San Francisco may fail completely in São Paulo, Berlin, or Mumbai.
Two non-negotiables for international checkout based on this blog post:
- Familiar payment methods — Users won’t complete purchase if their preferred payment isn’t available
- Localized address flows — Address formats vary by country; forcing US conventions creates errors
However, consumer behaviors and expectations for checkout flows vary across the world which makes addressing these factors complicated, but not impossible. For instance, here are some regional patterns:
Germany:
- Address expectation: Street name first, then building number (opposite of US)
- Payment preference: Direct bank payments preferred by 31% over credit cards
India
- Address expectation: PIN code first (triggers address autofill)
- Payment preference: Cash on delivery dominant; UPI essential
Various examples from around the world:
- “For other countries during the checkout stage, we recommend having a single search field which asks the user to enter as complete an address as possible, ensuring that users aren’t prevented from entering their own address manually if they wish.”
- “As for payment options, again, local knowledge and research pays off. If the option is available to consult with a local company that specialises in ecommerce in your target location, having that knowledge under your belt will go miles towards making your site’s experience as familiar as possible to local users in your target country.”
- “Our final recommendation is to take a look at resources provided by the Department for International Trade. They provide a number of guides for selling to different countries and if you are looking for more general advice on selling abroad, they have advice for businesses of all kinds, depending on how much you currently export overseas.”
Key insight: Localization isn’t optional polish applied after launch. It’s structural — international checkout that ignores regional mental models creates friction equivalent to a broken form field.
Tactical recommendation: If expanding internationally, partner with local e-commerce consultants or use localized checkout providers (like Adyen or Stripe’s localized flows) rather than attempting to build regional expertise in-house.
Subscription Checkout: Selling Commitment, Not Products
Subscription checkout differs fundamentally from e-commerce because you’re not selling a product — you’re selling a recurring commitment.
The checkout flow must address different anxieties:
E-commerce checkout
- “Will I get what I ordered?”
- “When will it arrive?”
- “Can I return it?”
Subscription Checkout
- “What exactly am I signing up for?”
- “When will I be charged?”
- “How hard is it to cancel?”
The Subscription Trust Stack:
- Value clarity — What do I get? (Features, content, access)
- Billing transparency — When and how much? (Show exact charge date and amount)
- Exit visibility — How do I leave? (Cancellation policy before payment, not buried in ToS)
Why this matters for conversion: Users who feel trapped don’t subscribe. Making cancellation easy increases signups because it reduces perceived commitment risk.
Practical application: Display “Cancel anytime” prominently in checkout. It’s counterintuitive, but reducing perceived lock-in increases conversion by reducing commitment anxiety.
The Checkout Paradox
Checkout design is simultaneously one of the easiest yet hardest UX problems to solve:
- Easiest: The patterns are well-established. Baymard has published the guidelines. The mistakes are documented. Following best practices requires discipline, not innovation.
- Hardest: Every deviation from patterns — every “creative” solution, every “on-brand” choice that sacrifices clarity — costs real revenue. The business pressure to differentiate can fight directly against the UX imperative to be invisible.
The best checkout flows share one quality: they disappear. Users don’t notice them. They don’t remember them. They just… buy.
Originally published in Bootcamp on Medium.