The conversion rate has been falling across recent periods, and theme, app and price changes were made in that window. In this case, the immediate task is to investigate a declining store conversion rate. The workflow should produce a conversion diagnosis and experiment backlog, matching each dated change against the rate values in the merchant's records.
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 investigate a declining store conversion rate.
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 rate by day or week with the sessions and orders behind each value, the traffic mix by source, device and landing page, a dated log of theme edits, app installs and removals, price and shipping changes and checkout settings, and the rollback value for each change]. Separate a mix shift from a store change by first recomputing the rate within each source, device and landing-page segment and checking whether the fall persists there, then match the surviving movement against the dated change log and mark a change as a candidate only where it precedes the movement and no segment explains it.
Return a conversion diagnosis and experiment backlog with the timeline verified against the merchant's 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 rate by day or week for the affected periods together with the sessions and orders behind each value, the traffic mix by source, device and landing page for the same periods, and a dated log of every store change in the window, including theme edits, app installs and removals, price and shipping changes and checkout settings.
Separate a mix shift from a store change by first recomputing the rate within each source, device and landing-page segment and checking whether the fall persists there, then match the surviving movement against the dated change log and mark a change as a candidate only where it precedes the movement and no segment explains it.
Deliver a conversion diagnosis and an experiment backlog listing the candidate changes in date order with the segment evidence and the rollback value for each, then reconcile the orders behind every rate with the order export and confirm the rate definition and window are unchanged before the timeline is published.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
The rate is recomputed inside each source, device and landing-page segment for the same periods. If the total falls while every segment holds, the mix moved and the store did not; if the fall appears inside the segments too, a store-level change is the candidate. This check runs first, because a mix shift needs no rollback.
Every change that can alter what a shopper sees or pays, including theme edits, app installs and removals, price and shipping changes, checkout and payment settings, and tracking edits, each with its date and the person who made it. Undated changes cannot be tested against the timeline, so they are listed as undated and handled last.
A movement is treated as a decline only when it holds across the consecutive periods the merchant reviews and beyond the range of ordinary movement for that traffic volume. A single period is reported as an observation with its raw counts; if it persists across the reviewed periods, it enters the diagnosis as a confirmed movement.
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.