Nic Dunn, CEO, Charle Agency
What Is Shopify Rollouts?
Shopify Rollouts is a native tool in the Shopify admin (Markets > Rollouts) that lets you schedule changes to your Shopify store, show them to a share of visitors and measure the result before you commit. It runs server-side, so there is no extra apps, no injected script and no page flicker. Rollouts is available from the Basic plan, but running a true split test needs the Grow plan or higher.
Every rollout is one of three types. A Launch merges your changes into the store permanently at a date and time you choose. An Event publishes changes for a fixed period, such as a Black Friday theme, then rolls them back automatically. An Experiment compares a treatment against a control so you can measure the impact before a full Launch. Rollouts arrived with the Winter '26 Edition and has since widened from theme changes to checkout, customer accounts, discounts and catalogue prices.
What You Can Roll Out
Shopify's help documentation lists four areas a rollout can change: the online store theme (including a replacement theme, so you can compare two themes head to head), checkout and customer accounts configuration, discounts, and catalogue changes such as prices and availability for products. The Experiment analytics Shopify documents cover theme changes and checkout and accounts changes, so treat those as your testable areas.
For theme changes you work in the normal theme editor, which covers most of what a store tests day to day: the homepage hero content and blocks, navigation structure, collection page grids and filters, product page layouts, and call-to-action buttons. If you sell internationally with Shopify Markets, you can scope a rollout to one market. That lets you trial a UK layout while Europe or Australia stays unchanged.
Events are the underrated use. Schedule a seasonal theme weeks ahead, set the end date, and Shopify restores the previous theme without anyone working at midnight on launch day.
How to Set Up a Shopify Rollout
Create the rollout
In your Shopify admin, go to Markets > Rollouts > Create rollout. Name it so the result is findable later, such as "Homepage hero test, October". Choose the change type, then set a start date and an optional end date. You can deploy straight away or schedule the change for a specific time.
Set the traffic split
On eligible plans, a slider sets what percentage of traffic and visitors see the treatment rather than the control, which is your current live theme. Shopify shows estimated visitor numbers so you can judge the sample before committing. Start at 10 to 25 per cent of traffic to catch faults, then scale up. Allocation and split are separate settings: a 50 per cent allocation with a 50/50 split exposes 25 per cent of eligible visitors to the treatment. Visitors outside the test keep seeing the current version of the Shopify store.
Edit the treatment
Open the theme editor from the rollout and a banner confirms you are editing the treatment, not the live theme. Change sections, copy and content as normal, then save that version. Editing theme code inside a rollout needs the Edit code permission, so staff without it will see a restricted editor.
Shopify Plan Requirements for Rollouts
According to Shopify's rollouts requirements page, Rollouts is available from the Basic plan and Experiments need Grow or higher. Shopify Plus is not required. Many guides still say Advanced for split testing, as an earlier version of this article did. Shopify's own documentation says Grow, so check your admin before you pay for an upgrade.
Store owners and administrators can manage rollouts. Other staff need Markets > Rollouts access plus a permission that matches the change, for example Discounts for discount rollouts. If you are weighing an upgrade, our breakdown of Shopify pricing tiers shows what else changes between plans.
What Rollouts Can't Do
Liquid template changes cannot be part of a rollout, and vintage themes are not supported. Headless and custom storefront checkouts cannot be tested, and analytics cover only the online store, cart and checkout, so POS and headless visitors are excluded. A catalogue or discount can sit in only one unended rollout at a time, and B2B catalogues assigned directly to company locations cannot be added. Google, YouTube and app channels may show different prices from your Shopify store while an experiment or partial rollout runs.
Testing options are narrower than a dedicated tool. Hands-on reviews report that targeting is limited to Markets, with no built-in segmentation by new or returning visitors or by device. Checkout experiments report one number, checkout conversion rate, with no revenue per visitor. Shoplift also points out that an active test blocks theme updates and hotfixes, and that only one theme test runs per storefront at a time. Check the behaviour in your own admin before you plan a testing calendar.
Reading Rollouts Analytics
The metrics depend on the rollout type. A theme Experiment reports conversion rate over time, bounce rate, reached checkout rate and add to cart rate. A checkout and accounts Experiment reports checkout conversion rate. Launches add gross sales, average order value and orders. Shopify's documentation does not describe how significance is shown, so treat early gaps cautiously.
Sample size is where most Shopify stores go wrong. For example, a store that calls a winner after a few hundred orders is usually reading noise. As a rule of thumb, detecting a 10 per cent relative lift on a 2.5 per cent conversion rate at 95 per cent confidence and 80 per cent power takes roughly 62,000 visitors per variant. At 1,000 daily sessions split 50/50, that is about four months. Detecting a 30 per cent lift needs roughly 7,000 per variant, or about two weeks. Lower-traffic stores should therefore test bold changes, not button colours.
Run tests in whole weeks, because Monday visitors behave differently from Saturday shoppers, and aim for at least 14 days. Stopping after three promising days is the most common testing mistake.
Rollouts vs Third-Party A/B Testing Apps
Use Rollouts for Shopify theme and checkout changes, seasonal scheduling, and any store running its first experiment, because it adds no monthly cost or page-speed overhead. Keep or adopt an app when you need to test prices or offers, segment audiences, track custom goals, change Liquid, or read revenue per visitor with proper statistics. Popular options include Intelligems, Shoplift, Convert and Optimizely, typically from around £50 a month and rising with traffic. Our comparison of the best A/B testing tools for Shopify covers each in detail.
The two combine well. Run layout tests natively and reserve the app for pricing or checkout experiments. If you want a structured programme, see our CRO agency services or split testing services.
Why we treat Rollouts as a release tool first
Shopify Rollouts is excellent at deployment and only decent at experimentation, and merchants get into trouble when they confuse the two. The scheduled launch, the 10 per cent ramp and the automatic rollback are the real prize, because they remove risk from every theme release. The Experiment reports are useful evidence for decisions, not proof, until the sample size supports them.
Our rule on client builds is simple. Every theme release and seasonal swap goes through Rollouts. Price, offer and segment tests go through an app. No test starts without a written hypothesis, one primary metric and a minimum of two full weeks. We run split tests on charle.co.uk itself, using cookie-based variants tagged in Microsoft Clarity and reported in PostHog, and the lesson is consistent: traffic volume and discipline decide the outcome far more than the tool. The mistake to avoid is cancelling your testing app the week Rollouts lands without checking what you used it for.
What to Test First
New to experimentation? Start with high-impact, low-risk theme changes, and read our guide to A/B split testing for method. Good first tests include the homepage hero content (lifestyle imagery against product photography), navigation structure (dropdown against mega menu), collection grid density and filter placement, product page layout (reviews above or below the fold), and call-to-action copy such as "Add to Cart" against "Add to Bag".
Ramp gradually rather than starting at 50/50. Run at 10 per cent for the first three days to catch faults, move to 25 per cent for days four to seven, then 50 per cent until the planned end date. Finish by rolling the winner out to 100 per cent or rolling back. For wider ideas, see how to improve your ecommerce conversion rate.
