01
Step 1
Gather sales history across consecutive periods, sessions and orders, the availability log, and a dated list of the pricing, assortment and campaign changes made inside the same window.
Back to topic
Spark playbook scenarios · Spark: AI Store Assistant
Investigate why store sales are falling
A merchant has watched revenue slide for several periods and holds a list of plausible suspects, from seasonality to sold-out variants, none of them checked. In this case, the immediate task is to investigate why store sales are falling. The workflow should produce a list of candidate causes, each marked supported or contradicted by the records.
Topic
Sales Decline Diagnosis
Target keyword
why are my shopify sales dropping
Copyable AI prompt
A prompt you can drop directly into Spark
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.
ai-prompt.txt
Act as a Shopify operations specialist.
Help me investigate why store sales are falling.
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: [sales history across consecutive periods, sessions and orders, the availability log, and a dated list of the pricing, assortment and campaign changes made inside the same window]. Test each candidate cause against the records and mark it supported, contradicted or unresolved, treating a cause as supported only when the signature it predicts appears in the segments and dates it implies, as contradicted when the opposite signature appears, and as unresolved when the records cannot separate the two.
Return a decline explanation with each candidate cause marked supported or contradicted 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.How to solve
A recommended resolution workflow
Split the work into input, judgment, and deliverable phases. Verify each step's output before moving to the next.
02
Step 2
Test each candidate cause against the records and mark it supported, contradicted or unresolved, treating a cause as supported only when the signature it predicts appears in the segments and dates it implies, as contradicted when the opposite signature appears, and as unresolved when the records cannot separate the two.
03
Step 3
Deliver the candidate list with a verdict per cause, the evidence line that decided each one and the next check for anything unresolved, and verify that the supported causes together account for the observed movement or that the report names what remains unexplained.
FAQ
Common questions when running this scenario
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
Is a traffic fall the usual cause?
No single cause is usual, so each candidate is tested rather than assumed. Traffic, conversion and order value are all measured, and a cause is reported as supported only when the signature it predicts shows up in the segments and dates it implies.
How is a cause marked contradicted?
When the records show the opposite of what the cause predicts, for example sessions holding flat while a claimed traffic loss should have cut them. Marking contradicted causes matters as much as listing supported ones, because it stops the team re-litigating the same suspects next period.
Which marketing changes should be reviewed?
List every pricing, assortment and campaign change inside the window with its date, then check each against the timing of the decline. A change starting after the fall began cannot explain it, and that boundary removes most suspects quickly.
Related scenarios
Continue with adjacent or same-topic scenarios
Start with scenarios under the same topic, then expand into cross-topic matches suggested by similarity.
Next step
Want these AI scenarios embedded directly into daily operations?
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.