Shopify checkout customization now runs entirely through checkout extensibility: app-based UI extensions, Shopify Functions, web pixels and the branding API, which replaced the old checkout.liquid file.
Shopify Plus stores can add blocks to the information, shipping and payment steps, while every plan can customize the Thank you and Order status pages.
The changes that tested best in our database are small trust additions such as reviews and USPs; urgency timers in checkout have both won and lost.
This page covers what you can change, which plan you need, and what real A/B tests say about each kind of change.
For the wider picture, see our checkout flow guide and the Shopify CRO hub.
Shopify checkout extensibility: what it is and what it replaced
For years, Plus merchants edited a theme file called checkout.liquid and dropped in any HTML, CSS or JavaScript they liked.
That flexibility broke during Shopify upgrades and gave scripts access to the payment page.
Shopify's checkout developer docs now say checkout.liquid customizations for the information, shipping and payment steps are no longer supported.
The Thank you and Order status pages followed.
According to the same docs, script tags on those pages were sunset on August 28, 2025 for Plus stores and on August 26, 2026 for non-Plus stores.
Shopify Scripts, the Ruby code Plus stores used for discounts, shipping and payment rules, stopped running on June 30, 2026; editing was locked on April 15, 2026.
The four building blocks
| Building block | What it does | Who can use it (as of November 2026) |
|---|---|---|
| Checkout UI extensions | Add content blocks, banners, fields, upsells and reviews at fixed spots in checkout | Information, shipping and payment steps: Plus only. Thank you and Order status pages: all plans |
| Shopify Functions | Server-side logic for discounts, delivery and payment options, checkout validation, bundles | Any plan through public App Store apps; custom apps with Functions need Plus |
| Checkout Branding API | Colors, typography, corner radius, buttons, form controls, sections | Plus only |
| Web pixel extensions | Analytics and conversion tracking events in a sandbox | All plans |
The plan limits come from Shopify's checkout UI extensions reference, the Shopify Functions docs and the checkout styling docs, which state that "checkout styling customizations are available only to Shopify Plus merchants."
What you can no longer do
UI extensions run in an isolated sandbox. Shopify's docs say they "don't have access to sensitive payment information or the checkout page itself (HTML or other assets)," and they can only render Shopify's own components.
You can't inject custom CSS, move the payment form, or rewrite the page layout.
For CRO work, this matters in two ways.
Your ideas must fit the slots Shopify exposes, such as above or below the order summary, under the Pay now button, or inside a step.
And a client-side testing script can't reach into checkout to change it the way it changes a product page.
No-code options
You don't have to build an app. Shopify's free Checkout Blocks app lets any merchant add images, text, banners and accordions to the Thank you and Order status pages.
On Plus, it also adds custom fields, hides, renames or reorders delivery and payment methods, and creates Functions-based rules.
Many third-party apps in the App Store add reviews, trust badges and upsells as checkout blocks too.
Checkout extensibility in practice: what to change and what the tests show
Because the slots are fixed, almost every checkout customization falls into one of a few types: trust content, form changes, urgency, upsells and payment or delivery rules.
Here is how each has performed in tests from our database.
Results come from different stores and traffic levels, so treat them as ideas to test, not lifts to expect.
Trust blocks: reviews and USPs near the Pay button
This is the strongest pattern in the data.
Dave Diederen showed customer reviews on the checkout page and measured +17.96% conversion rate, +23.13% revenue per visitor and +4.39% AOV.
In another store, Dave Diederen added three USPs below the Pay now button (guarantee, orders shipped, delivery speed) and recorded +9.90% conversion rate and +10.07% revenue per visitor, with no change in AOV.
Both tests put specific answers to last-minute doubts (Will it arrive in time?
Can I return it? Do other people like it?) right where the shopper commits.
That is the job a checkout block does best.
Trust blocks don't always move the number. Clean Commit added three trust icons to a Shopify checkout and got +0.9% conversion rate and +3.6% RPV across 213,715 visitors over 40 days, a well-powered flat result.
Generic icons that repeat what shoppers already believe seem to do less than specific reviews or guarantees.
Form and field changes
You can't redesign Shopify's form, but you can add helper content and remove optional fields.
Surefoot tested this at BioLite: the variant grouped email and phone into a labelled Contact Information block with text explaining how each would be used.
The BioLite test reported +18% transactions, 14% less drop-off from step 1 to step 2, and +20% email capture.
On Plus, the same idea extends to custom fields.
Every field you add through a UI extension is another thing to fill in, so make optional fields clearly optional and ask for them after the shipping address, not before.
Countdown timers: the result depends on the audience
Timers are a common checkout block, and the evidence is split.
Dave Diederen added a checkout countdown timer and saw +3.16% revenue per visitor and +2.99% conversion rate.
A later Dave Diederen test of the same idea in another store lost: -7.1% revenue per visitor and -5.4% conversion rate.
Kasper Dolk went the other way in a Plus checkout.
A post-purchase survey found Dutch shoppers felt the timer was too much pressure, and hiding the timer lifted conversion rate 25.61% and revenue per visitor 26.18%.
If you add a timer, test it, and check results by market.
Checkout upsells
Upsell blocks are popular because they raise order value without touching the form.
Kasper Dolk reports that a checkbox upsell in the Shopify Plus checkout was taken by 50% of orders for fashion clients.
Take rate isn't the same as profit, though, so track revenue per visitor too.
Measured results are smaller, and some are negative.
Dash of CX offered an engraved case for €6 at Loop Earplugs' checkout and lost 3.04% conversion rate and 2.96% revenue per visitor at 99% confidence.
The lesson is that an extra choice at the final step can distract as easily as it adds value.
Low-priced, easy add-ons with an obvious fit to the cart are the safest bet, and they still need a test.
Payment and delivery rules with Functions
Functions don't add visible content, but they change what shoppers see.
A delivery customization can hide a slow rate for some regions or rename an option so its speed is clear.
A payment customization can reorder methods or hide one for certain carts.
These changes affect conversion as much as any banner, so test them the same way.
Be careful with before/after claims
Some checkout results are period comparisons, not A/B tests.
A Kasper Dolk client with custom checkout blocks for USPs, reviews and an upsell reported a 3.8% conversion rate (+24%) month to date versus the previous period.
Sessions also rose 58% over that time, so other changes could explain part of the gain.
Use results like this to choose what to test, not as proof.
How to A/B test checkout customizations
Since testing scripts can't edit checkout directly, you have three routes:
- Shopify Rollouts. Since June 2026, Rollouts can A/B test checkout configurations, with "A/B testing a new upsell block for your checkout page" as Shopify's own example. Shopify's requirements page says Experiments need the Grow plan or higher and don't support headless or custom storefront checkouts.
- Testing apps with checkout support. Some tools assign the variant on the storefront, store it in a cart attribute, and have a checkout UI extension read it. Kameleoon documents this method for Plus checkouts.
- Test upstream. On a non-Plus plan, test the same message in the cart or cart drawer, where any tool can reach it.
Use revenue per visitor as the main metric, not checkout conversion alone, since upsells and timers can trade one for the other.
Our guide to A/B testing on Shopify covers tool choice and setup.
Checkout tests only count shoppers who reach checkout, so they gather data more slowly than homepage or product page tests.
Plan the test length before you start, and don't stop early on a good-looking day.
Want more ideas? Browse every result in the test database or return to the CRO guides.