Customers report that pages hang or stop halfway, yet the same URLs sometimes open normally on another device. In this case, the immediate task is to diagnose page-loading failures from available evidence. The workflow should produce an error inventory with repair and verification steps, putting each failure down to rendering, assets or server response.
First replace every [bracketed] placeholder with real store scope and constraints, then add any actual records or data you have, and paste it into your preferred AI assistant or Spark agent. Spark will follow this page's workflow to produce structured output and verification steps.
Act as a Shopify operations specialist.
Help me diagnose page-loading failures from available evidence.
My store context is [describe the business, products and target audience].
My scope is [specify the affected products, pages, customers or reporting period]. Use these inputs: [each reported URL with the device class, browser and connection used, whether the visitor was signed in or relying on an app-injected feature, and the timings separating time to first byte from asset completion]. Read the failed request and resource timings to place each case in one class, calling it a server response problem when nothing renders and the wait for the first byte dominates, an asset failure when the page arrives but listed images, fonts or scripts never complete or return their own errors, and a rendering delay when content appears late or only after interaction, leaving conflicting cases marked undetermined.
Return an error inventory with repair and verification steps, with each failure tied to the measurement behind it for the specified task.
Show the relevant evidence or before-and-after examples and explain how I can verify the result.
If essential data is missing, ask for it and mark unsupported conclusions as unknown.
Do not invent metrics, product claims or customer facts.
For any publication, message, account or store change, use only connected tools within my explicitly authorized scope and report success only after verification; otherwise provide the draft or proposed steps.Split the work into input, judgment, and deliverable phases. Verify each step's output before moving to the next.
Collect each reported URL with the device class, browser and connection used, whether the visitor was signed in or relying on an app-injected feature, and the timings on record separating the wait for the first byte from the time images, fonts and scripts take to finish.
Read the failed request and resource timings to place each case in one class, calling it a server response problem when nothing renders and the wait for the first byte dominates, an asset failure when the page arrives but listed images, fonts or scripts never complete or return their own errors, and a rendering delay when content appears late or only after interaction, leaving conflicting cases marked undetermined.
Deliver an error inventory with repair and verification steps tying every failure to the measurement that supports it, require the same URL to be retested on the same device class and connection until the behaviour repeats consistently, and list alongside it any reported page whose timings were never captured so absence of evidence is not read as success.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
A slow page still returns everything requested, just later, while a failed one never completes a listed request or comes back with an error status; measure time to first byte and full asset completion separately and report them as distinct findings, because the fix and the symptom shoppers see are not the same.
Device class, browser version and connection quality change how much can render before something gives up, so an image-heavy page can stall on one combination and load on another; retest under the conditions actually reported instead of your own, and record both outcomes.
Mark the case undetermined rather than choosing whichever class tells the tidiest story, then retest under identical conditions and note what changed between runs; a failure swinging between slow first byte and stalled assets usually means two causes, which deserve separate entries.
Start with scenarios under the same topic, then expand into cross-topic matches suggested by similarity.
Ciwi Spark builds Shopify-native AI tools and agents around the workflows merchants actually run: research, ads, translation, and customer support. Stop rewriting the same prompts by hand.