More sales from the traffic you already have
The usual pitch treats Shop Pay as a conversion lever on its own, but the published retail and checkout-form data only supports a narrower claim about field count and quarter over quarter share, and this piece settles what that narrower claim actually is.
Most merchants who search for a Shop Pay conversion rate are looking for one clean number that says a button raises sales by a fixed percentage, and no such number appears in the published data covered here. What the record actually contains is checkout form-field research from Baymard and quarterly retail sales figures from the US Census Bureau, and both point away from a single button explaining anything by themselves. The story worth telling is about form length and market share, not about one payment option carrying a checkout on its own.
Baymard’s testing found 42 percent of checkouts failing on one measure during testing and 30 percent failing on a separate measure during the same testing process. Those two figures describe different problems inside the same set of tested flows, and together they say that friction in checkout is common rather than rare. A UX benchmark from the same research showed a pattern present in 70 percent of tested checkouts, which is a wide majority, not a minority case that a single feature could patch over. Whatever a merchant adds at checkout, whether that is an express button or a saved-card option, it sits on top of a form that fails some measurable share of the time before any button is even reached.
The more uncomfortable number in Baymard’s data is the 35 percent figure, described as the share of checkouts where the same issue persisted despite changes already made. That number matters more than the failure rates above it, because it says that a merchant who shortens a form or removes a field cannot assume the fix holds without retesting. A checkout that passed a test once can still carry the same problem into its next version, and 35 percent is not a small tail case. Anyone reading a case study that claims a permanent fix from a single change should ask whether that claim was retested the way Baymard retested its own sample.
US Census Bureau quarterly data shows e-commerce sales moving 0.4 percent from the first quarter of 2026 in one reported comparison, and 0.2 percent from the first quarter of 2026 in a separate comparison. Neither number describes fast growth. A third figure shows e-commerce sales moving 0.9 percent from the second quarter of 2025, while total retail sales in the same comparison grew 6.7 percent, and a fourth figure repeats the 0.9 percent e-commerce move against 6.6 percent total retail growth from the same quarter. Read together, these four figures describe a period where total retail outpaced e-commerce by a wide margin rather than one where online checkout innovations were driving outsized growth. Any argument that says a checkout button explains rising online sales this period runs against the market numbers themselves.
| Measure | Figure | What it describes |
|---|---|---|
| Checkout testing failure rate, first measure | 42 percent | Share of tested checkouts failing during testing |
| Checkout testing failure rate, second measure | 30 percent | Share of tested checkouts failing during testing, separate issue |
| UX benchmark pattern | 70 percent | Share of tested checkouts showing the benchmarked pattern |
| Issue persistence after changes | 35 percent | Share of checkouts where the issue remained despite fixes |
| E-commerce sales change, comparison one | 0.4 percent | Change from the first quarter of 2026 |
| E-commerce sales change, comparison two | 0.2 percent | Change from the first quarter of 2026, separate reporting |
| E-commerce sales change, comparison three | 0.9 percent | Change from the second quarter of 2025, against 6.7 percent total retail growth |
| E-commerce sales change, comparison four | 0.9 percent | Change from the second quarter of 2025, against 6.6 percent total retail growth |
The Census Bureau figures come from quarterly retail sales reporting published by the agency and read from its e-commerce retail sales page as loaded on September 9, 2026, covering comparisons against both the first quarter of 2026 and the second quarter of 2025. The Baymard figures come from checkout flow testing described in the same study on average checkout form fields, also read as loaded on September 9, 2026. Neither source measures a Shop Pay specific conversion rate, and neither claims to. What they measure instead is form-level friction on one side and aggregate retail movement on the other, and a merchant trying to answer a question about express checkout has to fill the gap between those two datasets with a test of their own, because the published record stops short of naming the button by name.
A merchant chasing a Shop Pay conversion number should treat the 42 percent, 30 percent, 70 percent, and 35 percent figures from Baymard as the actual checkout variables worth testing, since those are the only checkout-specific numbers on record here. Anyone reporting store growth this quarter should compare it against the 0.4 percent, 0.2 percent, and 0.9 percent e-commerce movements from the Census Bureau rather than assuming a single feature explains a result the market data does not otherwise support.
The facts available here do not include a Shop Pay specific conversion number, only Baymard's checkout form-field data and Census Bureau retail figures. What the form-field data shows is that 42 percent and 30 percent of tested checkouts had measurable friction, and that a pattern held in 70 percent of tested flows. Any claim that a single express-checkout button changes conversion by a specific percentage is not supported by the facts in this piece, and should be treated as unverified until a source publishes a Shop Pay specific rate.
Baymard's testing put failure rates at 42 percent and 30 percent for issues tied to form design, and a related UX benchmark showed a pattern in 70 percent of flows tested. Even after fixes, 35 percent of checkouts kept the same issue, which suggests the form itself carries more weight than which express-payment button sits above it. A shorter form with a slow button will often outperform a long form with a fast one, though no number here proves that ordering absolutely.
US Census Bureau quarterly figures show e-commerce moving 0.4 percent and separately 0.2 percent from the first quarter of 2026, and a 0.9 percent e-commerce change against 6.7 percent and 6.6 percent total retail growth from the second quarter of 2025. Those are small, mixed movements rather than a clean upward trend. Anyone citing a bigger single number for e-commerce growth this period is not using the same source as this article.
Baymard's 70 percent figure describes a UX benchmark pattern found during testing, not a target number to hit by deleting fields blindly. The 35 percent persistence rate for issues after changes is the more useful number here, because it shows that removing fields without retesting still leaves a real chance the same problem resurfaces. Any field reduction should be tested the same way Baymard tested, not assumed to work from the percentage alone.
The facts point to checkout form structure as the measured variable, with 42 percent and 30 percent failure rates during testing and a 35 percent recurrence rate after changes. Express-payment buttons are not covered by a conversion-rate fact in this set, so no claim about their independent effect can be made here. Merchants should treat the form-field data as the primary lever until a source publishes button-level numbers.