See what changed in your Shopify storefront apps

Third-party apps add scripts to your storefront and update on their own schedule, without asking you first. On each scan, Revenue Shield records the storefront scripts it can detect, a signature for the scripts an app loads, their byte size and request timing, and any console errors seen in a real browser. Comparing one scan to the next surfaces when a script changed, when an error appeared where there was none, or when a detected script came or went.

This is the most important thing to understand about app observations. A changed script or a console error is a clue, not proof that an app caused a checkout failure. Confirm the customer-facing behaviour and investigate the relevant change before assigning a cause. Console errors are common on healthy stores, scripts change for reasons unrelated to any failure, and correlation in time is not causation. Use these observations to narrow where to look — not to close the investigation.

App observations are most useful next to a purchase-flow result. If a checkout monitoring run fails and a script or the theme changed around the same time, that change is a reasonable first suspect to check — but you still confirm it by reproducing the customer-facing behaviour. The observation tells you where to start; the purchase-flow test and your own reproduction tell you whether it is actually the cause.

The figures in app monitoring measure specific things, and it is easy to over-read them:

• Resource bytes are not full app size. The byte size Revenue Shield records is the storefront resources it observed loading, not the full size of the app's package or its server-side code.
• Request timing is not experienced page speed. How long a script took to load in a monitored request is not the same as how fast the page felt to a customer on their connection and device.
• Detected scripts are not your installed-app list. Revenue Shield sees publicly detectable storefront scripts. Apps that run server-side or add no detectable script will not appear, so this is not a complete inventory of what you have installed.

Purchase-flow monitoring answers whether a customer can reach checkout. App monitoring adds context about what changed on the storefront around that time. Neither replaces the other, and neither replaces your own reproduction of the issue. Together they give an investigation a starting point and a record to compare against, while the limits above keep those observations honest. See the app-by-app guides for specific tools.