Executive summary
- What: Rollouts, the scheduled-change system under Markets, now accepts discount changes alongside theme, checkout, accounts, and catalog changes.
- Why: You can expose a new offer to a share of online store traffic, measure conversion against your current offer, and expand only once the data supports it.
- Who: Merchandising and growth teams running promotional calendars, plus anyone shipping theme and discount changes as separate releases.
What changed
A rollout is a scheduled bundle of store changes with a start time and a defined fate: a Launch merges permanently, an Event publishes for a window then rolls back, and an Experiment runs a treatment against a control. Discounts are now a valid change type inside that bundle.
Three capabilities follow. You can schedule a discount to activate as part of a wider release. You can publish it to a fraction of online store traffic and widen the split later. And you can run it as an Experiment against your existing offer, then read the conversion comparison in rollout analytics before deciding which offer survives.
Existing customer eligibility rules still apply, and standalone discount start and end dates work as before. Rollouts is available on Basic and above, while Experiments require Grow or higher. One real constraint: traffic-split rollouts and Experiments apply to the online store only, so headless checkouts cannot be tested this way.
Why it matters
Most enterprise promotional calendars fail at the seams. The theme update lands Tuesday, the discount gets activated Wednesday by whoever is awake, and the checkout change waits for the next deploy window. Every gap is a chance for a landing page to advertise an offer the cart does not honour yet. Putting discounts into the same scheduled bundle removes that class of incident.
The measurement side is the larger gain. Discount depth is usually argued from intuition and last year's revenue, because the honest test is expensive: you need a traffic split, consistent attribution, and a clean control. Running the offer as an Experiment gives you that comparison inside the platform, with the control being your current promotion rather than nothing at all. If you are formalising that practice, our conversion rate optimisation work covers how to size and sequence these tests.
Role-specific impact
- Marketers: Promotional calendars become schedulable artefacts rather than a list of manual tasks. Offer depth can be tested rather than asserted, which changes how discount budgets get defended internally.
- Developers: Fewer out-of-band requests to toggle a discount at an awkward hour. Confirm which storefronts are in scope before promising coverage, since split traffic and Experiments do not reach headless checkouts.
- Store admins: Rollout permissions are granular. Adding a discount change requires the Discounts permission on top of Markets and Rollouts, so audit staff roles before the first campaign rather than during it.
Use-case example
Real-world scenario
A home goods brand runs a sitewide 15 percent promotion every autumn and assumes the depth is necessary. This year the team builds one rollout: the seasonal theme update, the campaign checkout banner, and an Experiment comparing 15 percent sitewide against 10 percent with free shipping over 75 dollars. Across ten days at an even split, the 10 percent variant converts 0.4 points lower but lifts average order value by 11 percent, producing higher contribution margin per session. The team ships it for the remaining six weeks and keeps roughly 5 points of margin it would otherwise have given away on faith.
Implementation checklist
- Confirm your plan tier. Rollouts needs Basic or higher, Experiments need Grow or higher.
- Audit staff permissions for Markets, Rollouts, Discounts, and Themes, then assign them to whoever will build the campaign.
- Inventory which storefronts carry your traffic. If a material share sits on a headless storefront, scope the test to the online store and say so in the results.
- Define the success metric before launch. Contribution margin per session is more honest than conversion rate alone.
- Build the rollout as a single bundle so theme, checkout, and discount changes share one activation time.
- Set a run length that clears your weekly seasonality, typically two full weeks, then review analytics before expanding.
FAQ
Q: Does a rollout override the eligibility rules already configured on a discount?
A: No. Customer eligibility conditions on the discount continue to apply. The rollout governs when the change is live and which share of online store traffic sees it, not who qualifies.
Q: Can we test a discount on a Hydrogen storefront?
A: Not with a traffic split or an Experiment. Those apply to the online store only, and headless checkouts are excluded from theme and checkout configuration changes. A full Launch at 100 percent is the workable path there.
Resources
Shopify's documentation on rollouts, including requirements and analytics, covers plan availability and the permission matrix in detail.
Need guidance? Talk to Makro.