Shopify Rollouts lets merchants schedule, gradually release, temporarily publish or A/B test a new online store theme and checkout and customer accounts configuration from Shopify Admin. A launch rollout is available on Basic or higher. A true experiment that compares a treatment with a control requires Grow or higher. The safest workflow is to define one business hypothesis, prepare a clean replacement or edit, choose eligible markets and traffic, test critical purchasing paths, monitor the native rollout analytics and decide in advance whether the change should be applied permanently or rolled back.
Rollouts is useful because it separates preparing a change from exposing every shopper to it. It does not make an untested theme safe automatically, and it does not replace functional quality assurance. Product selection, pricing, cart behaviour, discount logic, checkout extensions, analytics and integrations still need to be checked before live traffic is allocated.
Last reviewed: August 28, 2026. Shopify is actively developing Rollouts. Confirm the current options and plan requirements in your own Shopify Admin before scheduling a production change.
What Is Shopify Rollouts?
Shopify Rollouts is a native system for controlled storefront releases. It is managed from Shopify Admin > Markets > Rollouts. Instead of replacing the live experience for everyone in one manual action, a merchant can define the change, the affected markets, the reach, the timing and the result that should happen when the rollout ends.
The feature currently works with two customer-facing resources:
- the online store theme;
- the checkout and customer accounts configuration.
A rollout can include a theme change, a checkout and accounts change, or both. Shopify documents two rollout types: a launch and an experiment. These are not interchangeable. A launch is primarily a controlled publishing mechanism. An experiment is a comparative test with a control and a treatment.
Why use a rollout instead of publishing directly?
Direct publishing is appropriate only when the change is already accepted, the exact release time does not matter and the risk of exposing every visitor immediately is low. Rollouts adds operational controls that direct publishing does not provide in one workflow:
- schedule a theme or checkout configuration for a specific date and time;
- publish a temporary seasonal experience and automatically revert it;
- release a change to part of eligible traffic;
- target selected markets;
- compare a treatment with the existing experience;
- pause or end an active rollout;
- review rollout-specific analytics when the configuration supports collection.
The practical benefit is controlled exposure. If the treatment has a hidden cart, pricing or tracking defect, fewer customers can be affected while the team investigates.
Launch Rollout Versus A/B Experiment
Choose the rollout type from the decision you need to make. Do not label a gradual launch as an A/B test if there is no valid control comparison.
| Question | Launch rollout | Experiment |
|---|---|---|
| Primary purpose | Publish, schedule, phase or temporarily apply an approved change | Compare a treatment with the existing control |
| Minimum plan | Basic or higher | Grow or higher |
| Change-level split | No treatment-versus-control split | Yes, for eligible visitors |
| Typical use | Scheduled redesign, phased release, temporary campaign theme | Theme A versus current theme, new checkout configuration versus control |
| Decision at the end | Apply permanently or roll back | Use the evidence to apply or reject the treatment |
Use a launch when the decision is already made
A launch is the right model when the business has approved a change and needs safer delivery. Examples include switching from an older theme to a new one at a planned time, releasing a redesign gradually, or running a holiday theme for a defined week.
A launch has no change-level traffic split. Shopify applies the launch change to eligible visitors according to the markets and launch reach configured for the rollout. That means a 25% launch is a phased deployment, not automatically a statistically valid A/B comparison.
Use an experiment when you need comparative evidence
An experiment pairs an unchanged control with a modified treatment. It is appropriate when the question is measurable: does the new theme improve add-to-cart rate, reached-checkout rate or conversion without creating unacceptable side effects?
The treatment should be based on one primary hypothesis. Testing a new navigation, different product-page structure, revised cart drawer, new checkout upsell and new promotional message simultaneously can produce a result, but it will not tell you which change caused it.
Immediate publishing is a separate choice
Shopify warns that a rollout launched immediately does not collect rollout analytics, can no longer be edited and applies the changes permanently. If measurement or reversible exposure is important, do not use the immediate option as a shortcut.
Shopify Rollouts Plans, Permissions and Limitations
Plan requirements
- Basic or higher: launch rollouts.
- Grow or higher: experiments with control and treatment.
Plan eligibility does not mean every store architecture is compatible. Review the storefront, theme and checkout limits before planning the project.
Required staff permissions
The store owner, organization owner, administrators and authorized staff can manage rollouts. A custom staff role needs the Markets > Rollouts permission plus permissions for each resource it changes. Theme work requires Online store > Themes access. Checkout replacement requires permission to view and edit checkout and customer accounts. Translation or compatible app-managed content changes can require additional app permissions.
Use least-privilege access. The person analysing results does not necessarily need permission to edit a checkout configuration, and a theme contractor does not automatically need full store administration.
Current technical limitations
Shopify’s current documentation identifies three particularly important restrictions:
- Rollouts supports online store checkouts, not headless or custom storefront checkouts;
- vintage themes cannot be used with Rollouts;
- Liquid templates cannot be changed as part of a rollout.
A merchant using Hydrogen, a custom frontend or a heavily customized legacy theme needs a different release and experimentation plan. Confirm compatibility before promising that Rollouts will replace an existing testing platform.
Rollouts is not a staging environment
A rollout controls live exposure. It does not automatically validate the build. Preview the theme and checkout configuration first, use test transactions where appropriate, and document the production state before any visitors enter the rollout.
What Can You Change With Shopify Rollouts?
A rollout can contain only one change of each supported type. Shopify currently supports:
- editing the main theme;
- replacing the main theme;
- replacing the checkout and customer accounts configuration.
A single rollout can combine a theme change and a checkout configuration change, but combining resources also makes diagnosis harder. If a result changes, you must consider whether it came from the storefront, the checkout or their interaction.
Theme replacement
A theme replacement is suitable for a redesign, a switch to a new theme family or a materially different navigation and product-page experience. Shopify creates copies of the relevant published resources for the rollout workflow, allowing the live resource and rollout version to be managed separately. Keep a precise record of apps, theme app extensions and custom code that must be present in the treatment.
Main-theme edits
A theme edit is more focused, but the same testing discipline applies. Check templates, sections, blocks, menus, metafields, translations and app embeds. A change that looks correct on the home page can still fail on a product template with unusual variants.
Variant logic deserves specific attention. If a treatment changes product selectors, compare it against the failure modes in our guide to fixing the wrong Shopify variant being added to cart.
Checkout and customer accounts configuration
A checkout rollout replaces a saved checkout and accounts configuration. Possible differences can include layout, branding and compatible checkout extensions or blocks. The available configuration still depends on the merchant’s plan and the capabilities of each extension.
Rollouts does not restore retired platform features. If the store previously depended on legacy checkout customizations, review the Shopify Checkout Scripts migration checklist separately.
What to Prepare Before Creating a Rollout
1. Write the release objective
Use one sentence that describes the business decision. For example: “We want to know whether the simplified product template increases reached-checkout rate without increasing variant-selection errors.” This is better than “test the new theme” because it identifies a measurable outcome and a guardrail.
2. Select one primary metric and guardrails
A theme experiment might use conversion rate, add-to-cart rate or reached-checkout rate as its primary metric. A checkout experiment might focus on checkout conversion rate. Guardrails can include order value, error rate, refunds, support contacts, page performance or analytics completeness.
Do not choose a winner from revenue alone after a short period. A few high-value orders can distort a low-volume test. Define the metric, evaluation window and minimum evidence before seeing the result.
3. Freeze the treatment scope
List every deliberate difference between control and treatment. If an active experiment is edited, the measured treatment changes during the test and the result becomes harder to interpret. Reserve mid-test edits for correcting an urgent defect, and record the time and reason.
4. Record the current production state
Before starting, save:
- the current theme name and version;
- the active checkout and accounts configuration;
- enabled app embeds and extensions;
- analytics and pixel configuration;
- navigation and market-specific content;
- known incidents, promotions and inventory constraints.
This baseline helps distinguish rollout effects from unrelated changes.
5. Check for overlapping campaigns and deployments
A flash sale, email campaign, major paid-media launch, price import or inventory migration can change visitor behaviour during the same period. If the overlap cannot be avoided, document it and apply it consistently to control and treatment.
How to Create a Shopify Theme Rollout
The exact labels can evolve, but the operational sequence should remain disciplined.
Step 1: Open Markets > Rollouts
In Shopify Admin, go to Markets > Rollouts and create a rollout. Use a descriptive name containing the resource, hypothesis and planned window, such as “Theme navigation test – EU – September”.
Step 2: Choose launch or experiment
Select a launch for controlled deployment of an approved change. Select an experiment only when the store is eligible and you need a control-versus-treatment comparison.
Step 3: Add the theme change
Choose whether to edit the main theme or replace it. For a replacement, select the prepared theme. Verify that it is the correct copy and not an unrelated development version.
Step 4: Configure markets and reach
Select the affected markets and the launch reach. A market-specific rollout can be useful when translations, policies, currencies or merchandising differ, but market targeting should follow the hypothesis. Do not exclude difficult markets only to improve the result.
Step 5: Configure timing and end behaviour
Set the start and, when appropriate, end date. Decide whether the rollout should apply permanently or roll back. A seasonal theme normally needs a defined end and automatic reversion. A phased permanent redesign needs a decision gate before full application.
Step 6: Preview the affected experience
Review the rollout from the relevant Markets view and preview the theme itself. Test more than the home page. Include collection, search, product, cart, account and policy templates, plus a product with multiple variants and a product with a sale price.
If the preview displays an old or inconsistent price, use the checks in our Shopify stale-price troubleshooting guide before assigning real traffic.
Step 7: Activate and monitor
Confirm the owner, monitoring window and rollback contact before activation. Watch both Shopify rollout analytics and operational signals such as failed payments, customer contacts and app errors.
How to Test a Shopify Checkout Configuration
A checkout experiment is high impact because every exposed visitor has already shown purchase intent. The treatment should solve a defined problem, such as unclear delivery information, distracting layout or ineffective compatible checkout content.
Build a separate configuration
Create and review the replacement checkout and accounts configuration before adding it to the rollout. Confirm branding, supported extensions, validation rules and customer-account pages. Do not assume that a checkout app block behaves identically after being copied to a new configuration.
Test realistic order paths
At minimum, cover:
- guest checkout and signed-in checkout where supported;
- desktop and mobile devices;
- single-item and multi-item carts;
- discount codes and automatic discounts;
- shipping and pickup options relevant to the store;
- different markets, currencies and languages;
- tax-inclusive and tax-exclusive scenarios;
- supported payment methods;
- post-purchase and order-status behaviour;
- analytics, pixels and consent behaviour.
Protect measurement
A checkout can convert normally while reporting incorrectly. Verify that completed test purchases reach Shopify Analytics and the approved external platforms. If purchase events are missing or duplicated, follow the diagnostic path in Shopify purchases missing from GA4 and Google Ads before interpreting the experiment.
Define an emergency threshold
Set an operational stop rule before activation. Examples include a confirmed inability to pay, broken shipping selection, an incorrect total, a material increase in checkout errors or a loss of required compliance content. A statistical test should never continue through a critical purchasing defect.
How to Choose Traffic, Markets and Test Duration
Launch reach
Launch reach determines how much eligible traffic can experience the rollout. Start with a smaller percentage when the change is risky and increase it only after functional and operational checks pass. A small reach reduces exposure but also slows evidence collection.
Experiment split
For an experiment, Shopify applies the experiment split within the eligible traffic defined by launch reach and markets. Treat these as two separate filters. A 50% experiment allocation inside a 50% eligible reach does not expose half of all store visitors to the treatment.
Market targeting
Target a specific market when the treatment is designed for that market. If the purpose is a global redesign, limiting the first release to one representative market can reduce risk, but the result may not generalize to countries with different language, device mix, payment methods or delivery expectations.
Duration
Do not end an experiment as soon as one variation moves ahead. Allow time for weekday and weekend patterns, campaign cycles and conversion delay. Low-traffic stores may need a longer test, a larger treatment effect or a more focused question. If the store cannot produce enough observations for a meaningful comparison, use a controlled launch plus qualitative and operational evidence instead of presenting an underpowered result as proof.
Overlapping rollouts
Shopify can run multiple rollouts and experiments on the same resource. It calculates effective traffic and gives experiments mutually exclusive visitor segments so a visitor is assigned to one view for that resource. However, the intended percentage can differ from effective traffic when rollouts overlap.
Operational simplicity is usually better. Before overlapping tests, confirm that the team understands the effective allocations, control copies and interaction with launch traffic. Otherwise a clean test plan can become difficult to explain.
How to Read Shopify Rollout Analytics
Rollout analytics become available after an eligible rollout begins. Available metrics depend on the rollout type and whether the change affects the storefront, checkout or both. Shopify does not let merchants customize the native metric set for a rollout.
| Rollout resource | Useful native metrics | Operational guardrails |
|---|---|---|
| Theme launch | Gross sales, average order value, orders, sessions, conversion rate, bounce rate | Performance, navigation errors, product-selection defects |
| Theme experiment | Conversion rate, bounce rate, reached-checkout rate, add-to-cart rate | Wrong variants, stale prices, app and tracking errors |
| Checkout launch | Sales, orders, conversion and checkout conversion metrics | Payment, shipping, discount and validation failures |
| Checkout experiment | Checkout conversion rate | Payment failures, support contacts, tracking completeness |
Read exposure before outcome
Confirm which markets and how much effective traffic actually received the rollout. If overlapping rollouts changed the allocation, the observed sample may differ from the original plan.
Separate statistical movement from business value
A higher add-to-cart rate is not automatically a win if reached-checkout rate falls or customers add the wrong variant. A higher checkout conversion rate might be valuable, but the business should also review average order value, refunds, support and margin where available.
Check data quality
Compare orders, Shopify Analytics and approved downstream reporting. Bot traffic, consent configuration and pixel changes can alter measurements. If the store already has noisy analytics, resolve the data problem before using a small difference to approve a permanent redesign.
Document the decision
At the end, record:
- the hypothesis and treatment;
- dates, markets and effective allocation;
- primary result and guardrails;
- incidents and overlapping campaigns;
- the apply or rollback decision;
- follow-up changes that were deliberately excluded.
Pause, End, Roll Back or Apply a Shopify Rollout
Pause
Pause an active rollout when the team needs time to investigate a non-critical concern without closing the rollout decision permanently. A paused rollout can be reviewed and resumed.
End and roll back
End the rollout and choose rollback when the change is unsuccessful, unsafe or no longer required. A predetermined end can also return a temporary theme or checkout configuration to the previous state.
End and apply permanently
Apply the change when the rollout has passed functional checks and the evidence supports permanent use. Shopify warns that after a rollout is applied, the change cannot be reverted through the rollout. Preserve backups and release notes even when the result looks conclusive.
Archive or delete
Archiving and deleting are administrative cleanup actions, not rollback substitutes. Shopify states that an archived rollout cannot be restarted or unarchived and that archiving permanently deletes its theme changes. A deleted rollout cannot be recovered. Confirm that the change history and any necessary theme copy are preserved before cleanup.
Shopify Rollout Pre-Launch QA Checklist
Storefront and content
- Home, collection, search, product, cart, account and policy templates checked.
- Navigation, filters, predictive search and breadcrumbs work.
- Product media, prices, sale prices, availability and variants are correct.
- Market-specific language, currency and content are correct.
- Mobile, tablet and desktop layouts are usable.
- Accessibility basics such as keyboard navigation, labels, focus and contrast are reviewed.
Cart and checkout
- Add, update and remove actions work for normal and unusual products.
- Discounts, gift cards and bundles behave as expected.
- Shipping, pickup, tax and duties are correct for target markets.
- Payment methods and accelerated checkouts are tested where available.
- Checkout extensions display and fail safely.
- Order confirmation and customer-account pages are correct.
Analytics and operations
- Primary metric and guardrails are written before activation.
- Shopify Analytics and approved external purchase tracking are checked.
- Support and fulfillment teams know the test window.
- Emergency owner and stop threshold are documented.
- Promotions, deployments and data imports that overlap the test are recorded.
- Rollback and permanent-apply decisions have named owners.
Applications remain a common source of inconsistent behaviour between theme copies and checkout configurations. If the treatment depends on several overlapping tools, perform a Shopify app audit before the rollout.
Common Shopify Rollout and A/B Testing Mistakes
Calling every phased launch an A/B test
A launch is controlled deployment. An experiment is a control-versus-treatment comparison. Use the correct design and avoid claiming causal proof from ordinary before-and-after traffic.
Testing too many ideas at once
A complete redesign can answer whether the total experience performed differently, but it cannot isolate the responsible component. Use focused treatments when learning is the goal.
Changing the treatment during the experiment
Editing a treatment after traffic begins creates multiple versions inside one result. If an emergency fix is necessary, document it and consider restarting the measurement.
Ignoring market and device differences
A global average can hide a serious failure in one language, market or device type. Functional QA should cover each targeted experience even if the native report is aggregated.
Using immediate launch when analytics are required
Shopify does not collect rollout analytics for immediately launched permanent changes. Choose the rollout configuration that matches the measurement objective.
Forgetting operational guardrails
A conversion increase does not excuse incorrect prices, inaccessible controls, missing consent, broken payment methods or a surge in fulfillment mistakes. Protect the customer experience first.
Running overlapping experiments without understanding effective traffic
Shopify can allocate mutually exclusive traffic automatically, but effective allocation can differ from intended percentages. Review each active rollout before interpreting its result.
Treating rollback as a substitute for backups
Rollback protects the active rollout path, but applied, archived or deleted states have important limits. Keep independent theme and configuration records appropriate to the risk.
Shopify Rollouts FAQ
What is Shopify Rollouts?
Shopify Rollouts is a native Admin feature for scheduling, gradually releasing, temporarily applying or testing online store theme and checkout and customer accounts configuration changes.
Which Shopify plan includes Rollouts?
Launch rollouts are available on Basic or higher. Experiments that compare a treatment with a control require Grow or higher.
Can I A/B test two Shopify themes?
Yes, an eligible store can use an experiment rollout to compare a replacement theme treatment with the existing control. The store must be on Grow or higher.
Can I test a Shopify checkout configuration?
Yes. Rollouts can replace a checkout and customer accounts configuration for a launch or eligible experiment. Only online store checkouts are supported; headless and custom storefront checkouts are not.
Does Shopify Rollouts work with headless stores?
No. Shopify currently states that headless and custom storefront checkouts are not supported by Rollouts.
Can Rollouts test Liquid template changes?
No. Shopify currently lists Liquid template changes as unsupported within Rollouts.
Does an immediate rollout collect analytics?
No. Shopify states that a rollout launched immediately does not collect rollout analytics, cannot be edited afterward and applies the changes permanently.
Can I target a rollout to one Shopify market?
Yes. Eligible visitors are determined partly by the markets selected in the rollout configuration, so a rollout can be scoped to relevant market experiences.
Can I run several Shopify experiments at the same time?
Yes. Shopify can run multiple experiments on the same resource and assigns mutually exclusive traffic segments. Effective traffic is recalculated when rollouts overlap, so review actual allocation before analysing results.
Can I pause a Shopify rollout?
Yes. An active rollout can be paused and later resumed. It can also be ended early, at which point the merchant chooses to roll back or apply the changes.
Can I undo a permanently applied rollout?
Not through the rollout. Shopify warns that after a rollout is applied permanently, the change cannot be reverted using that rollout. Preserve appropriate backups and release records.
How long should a Shopify A/B test run?
There is no universal duration. The test needs enough eligible traffic and time to cover normal business cycles without being dominated by promotions or anomalies. Define the evaluation rule before the result is visible.
Plan the Release Before You Allocate Live Traffic
Shopify Rollouts can reduce the risk of a theme or checkout change, but only when the treatment, QA scope, measurement and rollback decision are clear. If your store needs a theme release, checkout configuration review, analytics repair or a structured Shopify troubleshooting plan, explore more Shopify development and store support guides from SmallDeskApps.
