🇹🇷 Türkçe: Bu yazının Türkçesini oku →

Every year around this week I get the same question from a board or a founder: we are going into the busiest five days of the calendar, our checkout is fine, so why do we still have an abandonment problem. The honest answer is that the checkout is not fine. It is as good as the last redesign made it, and in the intervening two years the market has moved the finish line.

The move that most mid-market retailers have missed is not another A/B test on the button colour. It is that three of the largest commerce ecosystems have built accelerated checkout wallets that keep a shopper’s address, card and preferences in a network token rather than a form, and they now publish what happens when a named retailer plugs that in. The pattern is consistent. The methodology is not.

I went and read the three primary sources this quarter, not the aggregator summaries that round every figure up by five points. Shopify’s Everlane case. Stripe’s Link product page and the Kickstarter line on it. PayPal’s Fastlane case on Adorama. Three different vendors, three different retailers, three different ways of expressing the same mechanic: a returning buyer who does not fill in a form converts more often than one who does. The question for a buyer is what any of that actually means on their own stack.

Here is what the three cases say, and what I would tell a board asking which of them to care about going into 2027.

The lifts are real; the numbers are not comparable

Every figure below is a vendor’s own publication, from its own case study, under its own methodology. None of them is an independent benchmark. That is not a reason to ignore them, it is a reason to read them as directional. The underlying mechanism is the same across all three: a tokenised network remembers who the buyer is between sites, so the slowest part of a mobile checkout (typing an address into a form while holding a phone in one hand) disappears. Where the vendors differ is in what they choose to measure and against what baseline.

The three cases I want to work through: Everlane on Shop Pay, which Shopify published with named adoption and conversion figures. Kickstarter on Stripe Link, which is the clearest public share-of-checkout number in the category. And Adorama on PayPal Fastlane, which is the only one of the three that separates guest and returning users and reports each side.

Everlane and the 15% day-30 adoption on Shop Pay

Everlane is a California DTC clothing brand that re-platformed to Shopify and turned Shop Pay on at launch. The Shopify case study reports that 15% of Everlane’s U.S. transactions were processed through Shop Pay within 30 days of the launch, attributed on the page to Anna M. Peterson, product lead at Everlane. Shopify adds that when Shop Pay is available in a checkout it is selected about two-thirds of the time, that Shop Pay orders carry a 15% higher average order value than other checkout methods on the same store, and that Shop Pay reached peak conversion rates of up to 70% on Everlane specifically.

That 70% is a conversion rate at the top of the funnel the vendor chose, not a lift against a baseline. Two baselines to hold next to it: Shopify’s own cross-brand claim on the Shopify Checkout page is that Shop Pay can lift conversion by as much as 50% versus plain guest checkout, from a commissioned consulting study, and that it outperforms other accelerated checkouts by at least 10%. Both figures are vendor-reported. Both are plausibly directionally right. Neither tells you what any particular Everlane shopper would have done without Shop Pay in the way, which is the number the CFO will ask for and nobody actually has.

The one transferable lesson here is not the 70%. It is the two-thirds. When shoppers are presented with an accelerated wallet they already have an account on, they use it. Removing the choice is not required; offering the choice is enough.

Kickstarter and the one-in-three checkout on Stripe Link

Kickstarter’s own figure on the Stripe Link product page is that over 35% of transactions are processed using Link. The page describes Link as powering one in three payments across Kickstarter. Around 70% of those Link transactions come from returning customers, which matters because it tells you what Link is actually optimising: a network that remembers between sites, not a wallet a buyer sets up site by site.

Stripe’s broader figures on the same page: Link reaches over 300 million consumers worldwide, buyers check out 3x faster than non-Link customers, and across Stripe’s merchant base Link delivers an average 14% conversion increase for businesses with a large returning customer base. The 14% is the only cross-merchant figure that is honest about who it applies to. On a store where most orders come from first-time buyers the number is lower by definition, because they have not yet made the first purchase that puts them in the Link network.

Stripe’s global checkout analysis covers nearly 20,000 B2C businesses on the Stripe network between August 2023 and February 2026, so it is at least a large sample across many stores. The Kickstarter line is a single-brand case; the 14% is the one I would quote to a board. Headline conversion lifts attributed to specific retailers (Italic at 34% in its first month on Link, reported in Stripe’s own newsroom) are correct as published but they describe a launch window, not a steady state.

The transferable lesson: before a store commits to Link, it should know what proportion of its revenue comes from repeat customers. If it is 70% like Kickstarter the case is almost obvious. If it is 20% the mechanism is not the lever to pull first.

Adorama and the split between guest and returning users on PayPal Fastlane

Adorama is a US photography and electronics retailer that sits in the considered-purchase category: the average order is high, the specifications matter, the customer researches. PayPal’s Fastlane case study on Adorama reports that guest checkout conversions rose by up to 8% and that repeat Fastlane users saw a lift of close to 20%. The figures are from Adorama’s internal data as of Q1 2025 and are published as directional, not guaranteed.

The thing worth saying about this case, which neither of the first two offers, is the split. Shop Pay’s cross-brand number collapses guest and returning into one headline. Link’s 14% is honestly scoped to returning-heavy businesses. Fastlane reports both sides and the gap between them is 2.5x. That gap is the clearest signal in the category: the first conversion a shopper does on any one of these wallets is where most of the lift will come from on their second.

Fastlane is also the only one of the three positioned to lift guest checkout as such, because it recognises a returning shopper by email alone without requiring them to have created a PayPal account. Shop Pay and Link lean on an identity the shopper has signed up for; Fastlane assumes they have not. On a store whose repeat rate is low, that framing is worth testing before anything else.

The transferable lesson: the number that tells you whether an accelerated wallet will work on your store is not the vendor’s headline figure, it is your own repeat-purchase rate, broken out by channel and by category.

One mechanism, three different claim styles

What the three cases share is not the lift percentage. It is the shape. Each wallet keeps a buyer’s details in a tokenised network, so the second checkout on any participating site does not ask them to fill a form again. The lift comes from removing the mobile form, not from the branding of the button above it. A buyer who has never bought anywhere in that network gets a small or zero lift. A buyer who has gets a large one.

Where the three vendors differ is what they publish. Shopify emphasises in-funnel conversion and adoption on a specific Shopify-hosted store. Stripe emphasises network size and share of checkout. PayPal emphasises the guest/returning split because it is the only one of the three whose guest baseline is also PayPal’s own. Read any one of the three in isolation and you will overweight whichever metric that vendor leads with. Read all three together and the mechanism is clearer than any single figure.

What this looks like inside a WooCommerce store

We build on WooCommerce, so the question I get asked every time one of these cases comes up is what any of it means outside Shopify’s own walls. The honest answer is that two of the three work on WooCommerce today and one of them is optional. Shop Pay is Shopify-only and is not coming off that platform, so the Everlane figures are a Shopify-only argument. Fastlane and Link both have WooCommerce paths.

PayPal Fastlane ships on WooCommerce through the official PayPal Payments extension on WooCommerce.com, which added Fastlane support during 2025. The install is the same as any other PayPal gateway, a connect flow against a PayPal business account and a toggle in the WooCommerce payments settings. The practical cost is small on a typical mid-market store: a couple of hours to install, test on a sandbox, and go live. The non-obvious cost is that Fastlane renders inside the WooCommerce checkout block, which means the store needs to be on the block-based checkout rather than the legacy shortcode. If the store is on the shortcode checkout, that migration is the real work, and it is a conversation about theme compatibility, custom field handling at checkout, and whatever third-party plugins hang off the checkout hooks. A week of effort on a site with a non-trivial checkout is a realistic estimate.

Stripe Link is on WooCommerce via the official Stripe extension, and it is on by default on new installs. The decision point there is not whether to ship it, it is whether to surface it prominently or leave it as one wallet among others. On a store with a strong returning customer base the honest call is to raise Link’s priority in the checkout; on a store where most buyers are first time, leave it where it is and spend the attention elsewhere.

What a buyer should actually measure, inside WooCommerce, is not the wallet’s own reported lift. It is the baseline they are comparing it to. On a store with GA4 and the enhanced ecommerce events set up correctly, the comparison is straightforward: conversion rate for sessions that complete checkout with Fastlane or Link versus sessions that complete it without, segmented by new versus returning, over at least four full weeks. On a store where GA4 is half-configured and the checkout events are inconsistent, the measurement work is the first thing to do; without it the vendor’s headline will be the only number anyone cites, and nobody in the business will trust it.

One further operational point that does not appear in any of the three case studies: an accelerated wallet moves customer data out of your store’s forms and into a third party’s network. On a WooCommerce store that integrates with an ESP, a CRM or an ERP, the address and contact capture that used to happen in the form now has to be pulled from the wallet’s webhook payload and mapped to the downstream integration. That mapping is the piece that tends to break in the first week after launch. It is cheap to fix and expensive to miss.

What I would do this quarter

If you run a mid-market WooCommerce store and you are reading vendor case studies for ammunition going into the Christmas review, I would do three things. Pull your own repeat-purchase rate broken out by channel and category for the last twelve months. Decide whether Stripe Link or PayPal Fastlane is the better fit based on that rate, not on which vendor had a bigger press release. And commit the measurement work to tell the difference six months from now, because without it you are quoting someone else’s number about your own store, which is how a decision gets made that nobody can defend in the next budget cycle.

We handle this every quarter for brands on WooCommerce and on WordPress VIP, and the framework above is what we start with. If any of it maps onto a decision you are making this month, write to us and we will look at your own numbers against the three cases above.

Leave a Reply

Your email address will not be published. Required fields are marked *

Close Search Window
↑