Reported conversions do not match the order count, and no one has checked the purchase path in one pass. In this case, the immediate task is to audit conversion measurement across the purchase journey. The workflow should produce a tracking diagnosis and a repeatable test plan, with each expected event matched once against the merchant's order identifiers.
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 audit conversion measurement across the purchase journey.
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: [the expected event for each journey step with its required parameters and trigger, the tag and pixel configuration with install locations, order identifiers for the window, the consent configuration, and the timezone and currency used by the records]. Walk the journey once with consent granted and once under the default setting, confirming at each step that the expected event fires exactly once with the order identifier, value and currency matching the record, and classifying every deviation as missing, duplicated, late or fired without a required parameter.
Return a tracking diagnosis and a repeatable test plan with events reconciled against the supplied order records 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.
Write down the expected event for each step of the purchase journey with the parameters it must carry and the page or action that triggers it, then collect the tag and pixel configuration with where each is installed, the order identifiers the merchant can supply for the window, the consent configuration, and the timezone and currency the records use.
Walk the journey once with consent granted and once under the default setting, confirming at each step that the expected event fires exactly once with the order identifier, value and currency matching the record, and classifying every deviation as missing, duplicated, late or fired without a required parameter.
Deliver a tracking diagnosis and a repeatable test plan naming each deviation with its step, its likely cause among configuration, consent, theme load order or double installation, and the check that confirms the fix, then reconcile the purchase events captured with the merchant's order records for the window by order identifier.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
One clean pass means every expected event fires exactly once, in the documented order, carrying the identifier, value and currency that match the test order. Anything else is a deviation written down with its step; a pass that fires an event twice is not clean, and a pass run while tracking is already broken proves nothing about the fix.
Each purchase event is matched on the order identifier recorded with it, and where that differs, on the value, currency and timestamp within the window. Matching on counts alone hides both duplicates and missing records; events that cannot be matched to an order are listed as unmatched rather than counted as conversions.
Consent state decides which tags are allowed to send, so a journey run without consent shows what a real first-time visitor produces. If events appear only when consent is granted, the gap is expected behaviour and not a fault; testing both states keeps the merchant from chasing a tag that is working as configured.
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.