Sessions are being recorded while the order count stays empty, and there is no trace of how far those visits actually got. In this case, the immediate task is to diagnose why visits are not becoming orders. The workflow should produce a conversion diagnosis and experiment backlog, following each visit through the events the merchant already captures.
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 why visits are not becoming orders.
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: [session records with the ordered event sequence per session, the order table for the same window including cancelled, draft and test orders, the tracking and consent configuration in force, and checkout and payment gateway records]. Trace each session to the last event it produced and count sessions by that stopping point, first ruling out a measurement failure by confirming that the events the store is expected to send do arrive for a test session, and only then treating a stopping point as a behavioural one.
Return a conversion diagnosis and experiment backlog with the stopping point checked against the supplied 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.
Pull the session records for the window with the ordered events captured for each one, the order table for the same window including cancelled, draft and test orders, the tracking configuration and consent settings in force, and the checkout and payment gateway records, so that an absent order and an unrecorded one can be told apart.
Trace each session to the last event it produced and count sessions by that stopping point, first ruling out a measurement failure by confirming that the events the store is expected to send do arrive for a test session, and only then treating a stopping point as a behavioural one.
Deliver a conversion diagnosis and an experiment backlog naming the step where sessions stop with the counts behind it and the tracking checks already cleared, then reconcile the session count with the merchant's own visit record and the order count with the order export, and list any step whose events are missing entirely as unmeasured rather than as a zero.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
Before any behavioural conclusion, run a test session end to end, check its events against what the store should send, and query the order table for cancelled, draft and test orders. Ruling measurement out first stops a broken tag being read as a shopper problem; if the check fails, the finding is a tracking fault, not a conversion one.
Each session is followed through its own event sequence, and the last event it produced is the step where it stopped, counted across all sessions for the window. Sessions whose sequence is incomplete because an event is missing are counted separately, because a gap in the record is not the same as a shopper leaving at that step.
Test orders, draft orders and cancelled orders are identified in the order table and reported apart from the live count. Including them can make a period look as if it converted when it did not, and excluding them without saying so can hide a real order; the rule applied is stated in the output so the figure can be reproduced.
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.