Search "cart abandonment" and almost every result tells you to send a better email. Write three of them. Add a discount to the second. Time the first one for an hour after the shopper left. The advice is not wrong, exactly. It is just aimed at the wrong moment.
Here is the moment that actually matters. A shopper has chosen the product, added it to the cart, and started checkout. They are as close to buying as they will ever be. Then something on the checkout page stops them, and they leave. The recovery email arrives later and tries to drag them back to the exact page that already failed once. If the page failed because shipping was higher than expected, or the form asked for an account, or their preferred way to pay was not there, the email is asking them to return and hit the same wall.
The average online store loses roughly 70 percent of the shoppers who reach this point. Recovery emails, sent well, claw back a single-digit slice. The larger lever, by a wide margin, is fixing the checkout so fewer people leave at all. This post is about what actually drives abandonment, why so much of it is a payments and friction problem, and what to fix before you write another email.
Where the cart abandonment number came from
The shopping cart metaphor dates to mid-1990s web stores, and the "abandonment rate," the share of carts filled but never checked out, followed quickly as analysts saw how many shoppers walked away mid-purchase.
The figure most people quote traces to one organization. The Baymard Institute, an independent ecommerce usability research firm founded in Denmark in 2009, has tracked cart abandonment for over a decade by aggregating dozens of separate studies. Its running average, drawn from 50 studies, sits at about 70.2 percent. That number has barely moved in years, which tells you something. A decade of recovery-email tooling, popup apps, and exit-intent widgets has not meaningfully lowered the rate. The friction is structural, and the industry has mostly been treating the symptom.
Baymard's more useful contribution is not the headline rate but the breakdown of why people leave, drawn from its checkout testing and surveys of US shoppers. That breakdown is where the payments story lives.
What actually makes people abandon a cart
Strip out the shoppers who were just browsing and never intended to buy, which Baymard puts at around 43 percent of all abandoners, and you are left with the people who wanted the product and still left. Among those shoppers, Baymard's survey ranks the reasons:
- 39 percent: extra costs (shipping, tax, fees) were too high
- 21 percent: delivery was too slow
- 19 percent: did not trust the site with credit card information
- 19 percent: the site required them to create an account
- 18 percent: the checkout was too long or complicated
- 15 percent: the returns policy was unsatisfactory
- 15 percent: the site had errors or crashed
- 14 percent: could not see or calculate the total order cost upfront
- 10 percent: not enough payment methods
- 8 percent: the credit card was declined
Read that list as a whole and a pattern is hard to miss. Almost none of it is "changed my mind about the product." It is the checkout itself. Unexpected costs, a forced account, a slow form, security doubt, wrong or missing payment methods, a card that would not go through. The shopper had already decided to buy. The checkout talked them out of it.
Notice how much of the list is a money-and-payments problem specifically. Extra costs and an unclear total are about the price the payment screen reveals. Security doubt is about handing over card details. Insufficient payment methods and declined cards are payment problems by definition. Group those together and they dwarf every other category. An abandoned cart is, more often than not, a checkout that mishandled the moment of paying.
Why the recovery email is treating a symptom
Recovery emails do work, within limits, and you should run them. They are cheap, and a shopper who got distracted, lost their connection, or wanted to think it over is exactly who an email can win back.
The limits are the point. Open rates on these emails are high for marketing email, often around 40 percent, because the shopper recognizes what the email is about. But the conversion that matters, carts actually recovered, is modest. Most stores recover a few percent of abandoned carts through email, and even strong programs land at roughly 10 to 14 percent. On a generous reading, a good program recovers maybe one in seven lost carts. The other six are gone.
Now line that up against the causes. An email cannot lower your shipping cost, remove the account-creation step, add iDEAL for your Dutch shoppers, or fix a card form that rejects valid cards. It can offer a discount code, which does address the "costs too high" group, but a blanket discount is an expensive, indiscriminate fix. It hands money to every recovered shopper, including the many who would have completed at full price if the checkout had simply been clearer. You are paying to paper over a problem the checkout created.
Baymard's own estimate of the upside makes the comparison stark. The firm calculates that a large ecommerce site can lift its conversion rate by about 35 percent through checkout design alone, and puts the lost orders recoverable from poor checkout UX, across US and EU ecommerce, at around 260 billion dollars. No email program comes close, because the email operates after the damage and the checkout fix prevents it.
Put plainly: recovery email is the ambulance, checkout work is the guardrail. You want both, but a store with a polished three-email flow and an unaudited checkout has spent its effort on the ambulance.
The payment-method problem most stores underrate
Of all the checkout fixes, payment-method coverage is most often left until last, and it has some of the best evidence behind it.
Start with the simplest version: offering more than one way to pay. Stripe ran a holdback experiment across its checkout, comparing flows that surfaced a relevant additional payment method against flows that did not. Adding even one relevant method beyond cards lifted conversion by about 7 percent and revenue by about 12 percent on average. That is a meaningful gain from a configuration change, not a redesign.
The word doing the work there is "relevant." This is where most merchants get payment methods wrong. They treat it as a checklist, more logos equals better, and bolt on whatever their processor offers. The real lever is matching the method to the market, because payment preference is intensely local and the gap between a shopper's preferred method and a generic card form is where carts die.
The numbers on local methods are not subtle. In Stripe's test, surfacing the dominant local method produced country-level conversion lifts far beyond the global average: about 91 percent in China with Alipay, about 46 percent in Poland with BLIK, about 39 percent in the Netherlands with iDEAL. Those reflect markets where a card-first checkout is the wrong tool. A Dutch shopper expects iDEAL, the bank-transfer scheme that handles most online payments in the Netherlands. A Brazilian shopper increasingly expects Pix, the instant bank-payment network that took a large share of Brazilian ecommerce in a few years. Show those shoppers only a Visa and Mastercard field and a good portion leave, not because they cannot pay but because paying suddenly feels foreign and slow.
Worldpay's Global Payments Report 2026 puts the structural shift in one figure: digital wallets accounted for roughly 56 percent of global ecommerce transaction value in 2025, up from 53 percent a year earlier. Cards have not disappeared, and many wallet users fund those wallets with a card underneath, but the front-end experience the shopper expects has moved. A checkout built around a raw card form is, in much of the world, already behind the shopper's habit.
Express wallets: the fastest fix for the slow-form problem
Two of Baymard's top causes, a checkout that runs too long and security doubt about entering card details, share one strong remedy: express wallets that skip the form entirely.
When a shopper pays with Apple Pay, Google Pay, PayPal, or Shop Pay, they are not typing a card number and addresses into your form. They authenticate with a fingerprint or face scan, and the payment, shipping, and contact details flow from a vault they already trust. The long form, the typo risk, the doubt about whether your site is safe, all of it collapses into one tap.
The conversion evidence is consistent. In Stripe's payment-methods test, Apple Pay lifted conversion by roughly 22 percent. Stripe has separately found that surfacing Apple Pay early, as an express button rather than an option buried inside checkout, can roughly double its conversion contribution. Shopify's Shop Pay, citing a study by a major management consulting firm, reports that Shop Pay can lift conversion by up to 50 percent against a standard guest checkout, and that its mere presence as an option, even unused, lifts lower-funnel conversion by about 5 percent.
Treat the vendor percentages with caution, since vendors pick flattering comparisons and shoppers who reach for a wallet may already be more committed. But the direction is not in dispute: removing the form removes a large, well-documented source of abandonment. Placement matters as much as availability. An express wallet button at the top of the cart, before the shopper has invested effort in a form, prevents abandonment. The same button shown only at the end just speeds up people who would have converted anyway.
A fix-the-checkout playbook
If recovery emails are the ambulance, here is the guardrail, in rough order of return on effort.
Show the full price early. Unexpected cost is the number one cause of abandonment, and most of the time the cost was unexpected only to the shopper, because you hid shipping and tax until the final step. Surface shipping cost, or a clear estimate, on the cart page. If you offer free shipping above a threshold, state the threshold loudly. The goal is that the number on the payment screen is the number the shopper already expected.
Make account creation optional. Roughly a fifth of would-be buyers abandon when forced to register. Offer guest checkout as the default and, if you want the account, invite the shopper to create one after the order is placed, pre-filled from what they just entered. You lose nothing and stop punishing first-time buyers.
Cut the form down. Baymard finds the average checkout asks for far more fields than it needs, often around 24 form elements against an ideal closer to 12 to 14. Drop fields you do not need, collapse billing into shipping by default, and use address autocomplete. Every field removed is a smaller chance of a typo, a stall, or a second thought.
Match payment methods to your markets. Look at where your traffic and orders actually come from, then make the dominant local method present and prominent for each meaningful market: iDEAL for the Netherlands, BLIK for Poland, Pix for Brazil, the wallets that lead in your top countries. This is configuration, not engineering, and the conversion data behind it is among the strongest on this list.
Put express wallets up top. Add Apple Pay, Google Pay, and your platform's accelerated checkout, and surface them as express buttons at the start of checkout, not the end. This is the fastest structural fix for the slow-form and security-doubt causes at once.
Reduce avoidable declines and earn trust. A real share of abandonment is cards declined, sometimes wrongly, by overcautious fraud rules or weak retry logic. Review your decline and authorization rates with your processor. And reduce security doubt with the plain signals shoppers look for: visible card-brand marks, a clean checkout, no surprise redirects to an unfamiliar domain.
Then, and only then, run your recovery emails. With the checkout fixed, the emails are doing their proper job, recovering the genuinely distracted shopper, rather than repeatedly driving traffic back into a wall.
Where this is heading
Two shifts make the checkout-first view more correct over time, not less.
The first is that the checkout is leaving your website. Through 2025 and into 2026, AI shopping assistants began completing purchases inside the chat, with no redirect to the store at all. When the buying flow happens inside an assistant, there is no abandoned-cart email to send and no exit-intent popup to fire. What the assistant evaluates is your price, your delivery promise, and whether your payment and product data are clean enough to transact against. The recovery email as a safety net quietly disappears, and the quality of the underlying checkout is all that is left.
The second is that payment methods keep fragmenting and localizing. Account-to-account schemes like Pix and UPI keep taking share, real-time payment rails keep launching, and buy-now-pay-later remains an expectation in markets like Germany. A checkout well matched to its markets two years ago drifts out of alignment as habits move. Payment-method coverage is becoming an ongoing operational task, reviewed per market on a schedule, not a one-time setup.
Both shifts point the same way. As the checkout moves off your site and paying gets more local and varied, the recovery email gets thinner as a patch and the underlying checkout matters more. Send the emails. They are cheap and they help. But if you only have time to fix one thing, fix the page where the 70 percent are actually leaving.
Council summary
This post argues that cart abandonment is mostly a checkout and payments failure, not a missing email, and that fixing the page where shoppers leave beats recovering them after. Every figure was fact-checked against primary sources: Baymard's 70.2 percent average across 50 studies, its ten-reason abandonment breakdown, the 43 percent who were just browsing, the 35 percent checkout conversion upside, Stripe's 7 and 12 percent lifts with country-level numbers, and Shopify's Shop Pay claims. The review corrected one misattribution: the 56 percent digital-wallet share is from Worldpay's Global Payments Report 2026, so the citation and link were repointed there, and the Stripe payment-volume wording was tightened. The reader takeaway: run recovery emails because they are cheap, but spend real effort showing full price early, allowing guest checkout, cutting form fields, and matching payment methods to each market.
Comments