Automated checkout testing for Shopify

Each run drives a real browser through the buying journey: open a product, add it to cart, go to checkout, and reach the payment step. It runs on a desktop viewport and a mobile viewport, because most traffic is mobile and many breakages (a popup over the buy button, a sticky bar covering checkout) only happen on small screens. Every run records a video and a Playwright trace, so a failure comes with proof you can watch, not just an alert.

There are two ways to run it. The default is scheduled: the test runs every few minutes on its own, so a break is caught within minutes whether or not anyone deployed. The second is on demand, which is where the developer tools come in. After a risky change, a theme edit, an app install, a content update, you can trigger a run from the command line:

rs run <test-id>                 # queue a run now
rs runs <store> --json          # read the result for a CI gate

That lets you wire checkout testing into a deploy pipeline so a release that breaks checkout fails the build with a link to the video.

You can, and under the hood these are Playwright runs. The hard part is not writing one test; it is keeping it green. Storefronts change, products go out of stock, cookie and password gates appear, and bot protection blocks headless browsers. Revenue Shield generates the test for a store, injects the handlers for cookie banners, password pages and mobile menus, runs on infrastructure that gets past the render gates that block a plain headless browser, and self-heals the spec when the store changes. The tools give you the trigger-and-read surface over all of that, so you get the benefit of Playwright end-to-end tests without owning their maintenance.