Shopify Rollouts Now Schedule and Test Discounts

Rollouts arrived in June as a way to schedule, gradually publish and A/B test themes and checkout configurations from Markets > Rollouts in the Shopify admin. We covered that launch in our guide to scheduling and A/B testing themes with Shopify Rollouts. Until this week, discounts sat outside it. A merchant who wanted to know whether a percentage discount beat free delivery had to run one offer, then the other, and compare two different weeks of trading.

The timing matters. Black Friday falls on 27 November this year and Cyber Monday on 30 November, so most UK stores are settling their offer plans now. Being able to test an offer on a slice of traffic in October, then launch the winner with the Black Friday theme in one scheduled change, is a practical gain for that planning. The catch is in Shopify's developer notes. A discount in a rollout behaves differently on other sales channels and in older apps, and that affects any store running email, loyalty or reporting tools on top of Shopify.

How it works

A discount can now be one of the changes inside a rollout, in the same way as a theme or a checkout configuration. Shopify's changelog lists three uses. You can schedule a discount change, launching a new offer to a portion of online store traffic and expanding it when you are ready. You can compare offers, testing a discount change against your existing offers and reviewing conversion results before deciding which to keep. And you can launch changes together, scheduling a discount to go live alongside theme and checkout changes in a single rollout.

The rollout runs on its own schedule. According to Shopify's developer guide to discount Rollouts, while a rollout is serving, a treatment can add the discount for the buyers it reaches or remove it for them. It does not change the discount's own start and end dates or its configuration. The discount's customer eligibility rules still apply, and native start and end dates keep working as before for standalone promotions that are not in a rollout.

Reach is where the detail sits. The developer changelog entry says a discount activated at 100% effective traffic is available across all sales channels, subject to the usual eligibility rules. A partial-traffic activation, such as a test on half of visitors, applies on the Online Store only. Shopify also separates the traffic allocation you configure from the reach the rollout actually achieves, and says the configured figure alone does not determine how many buyers get the offer.

Examples / Use cases

A soft launch for a new offer. A homeware store wants to try a bundle discount but is unsure about the margin. It adds the discount to a rollout at a share of online store traffic for a few weeks, reviews the conversion results, then expands the offer to everyone or lets it lapse. Customers outside the test never saw the offer, so dropping it does not mean taking a visible promotion away from the whole store.

Offer against offer. A fashion brand has run an automatic first-order discount for a year and wants to know whether a free delivery offer would convert better. Instead of switching one off and comparing two different months, it tests the new discount against the existing offer over the same period, on the same traffic, and keeps the one that converts. Running both at once strips out the seasonal noise that makes a before-and-after comparison unreliable.

Black Friday in one change. A store builds its Black Friday theme, any checkout changes and the sitewide discount as one scheduled rollout, set to go live together and to conclude when the sale ends. That removes the late-night manual switch where the banner says 25% off but the discount has not been turned on yet, or the discount is live under the old homepage.

What this means for UK merchants

The first job is an app audit. Shopify's developer notes say that apps on Admin GraphQL API versions earlier than 2026-10 have some rollout discounts filtered out of what they can read: any discount that is active but may not reach all buyers, or whose status differs from the status its own dates imply. The developer guide adds that nothing in the response says why. For a merchant, that means a discount app, a loyalty programme, an email platform that pulls offers into campaigns, or a reporting tool may simply not see a discount that is under test. It comes back into view when the rollout stops affecting it. Apps also need a new read_rollouts access scope, and Shopify says existing installations may need the merchant to reauthorise them, so expect some approval prompts as app developers catch up.

The second is promotion. Shopify's guidance to app developers is blunt: only promote a discount that reaches every buyer, because one in a serving rollout might not. The same rule should apply to your marketing. If an email, a paid social ad or a Shopping campaign announces 20% off, and only half the store's visitors are in the treatment that gets it, the other half land on full price. That is a poor experience, and it can also be a misleading practice under UK consumer law. The CMA's guidance on unfair commercial practices under the Digital Markets, Competition and Consumers Act 2024 notes that breaches of the banned practices can bring civil penalties of up to the higher of £300,000 or 10% of worldwide turnover. Test quietly on the storefront, and advertise only what everyone can get.

The third is reporting. Shopify tells developers that a discount in a rollout can reach only part of the store, so crediting it with store-wide results overstates it. The same applies to your own analysis. If a test discount ran on a third of traffic, judge it against the matching group in the rollout, not against total sales for the week. For stores that want help planning offer tests and reading the results, our conversion rate optimisation service can run that work alongside the Shopify set-up.

Practical considerations

Partial tests run on the Online Store only. If you also sell through point of sale or other sales channels, customers there will not get a discount that is still on partial traffic. It only reaches every channel once it is activated at 100% effective traffic. Omnichannel retailers that promise the same offer online and in store should time the switch to 100% before they announce anything.

Expanding a test needs care with deadlines. A rollout lets you start an offer small and widen it later, which is the point of the feature. But if the test copy said the offer ends on a given date, extending it can turn a true deadline into a false one. Falsely stating that an offer is only available for a limited time is one of the banned practices in the CMA guidance above, and the CMA's separate advice on urgency and price reduction claims online is worth rereading before Black Friday. Keep countdowns and end dates out of test copy unless the end date is real.

Two set-up points. Themes built in Shopify Canvas do not support Rollouts at launch, as we noted in our breakdown of Shopify Canvas and its launch limits, so a Black Friday rollout that bundles a theme change needs a theme from the current editor. If you are commissioning a new theme for peak season, make Rollouts compatibility part of the Shopify theme build brief. And give tests enough time. A discount test on a small share of a small store's traffic can take weeks to show a reliable difference, so tests started in the next fortnight have the best chance of producing an answer before late November. Shopify's Help Center documentation on Rollouts covers discount scheduling and testing in the admin.

Black Friday lines often need different returns terms as well as different offers. Shopify now lets a store give a collection, product or variant its own return window, and we cover how that works, and the UK 14-day cancellation rules it has to respect, in our guide to Shopify return window overrides.

FAQ