The store wants recommendations driven by shopper behaviour but is unsure which signals it actually records. In this case, the immediate task is to recommend products using supported behavior signals. The workflow should produce recommendations that name the signal behind each suggestion and stay within what the merchant's records support.
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 recommend products using supported behavior signals.
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 store's recorded behaviour signals including product views, cart additions, past orders, wishlist entries and search terms, the system holding each, its consent coverage, retention window and the identifier linking it to the shopper]. Weight each signal only as far as the supplied data supports it inside that retention window, corroborate a single-event nudge such as a repeat view against a second record before shipping it, and discard any signal whose identifier cannot be matched reliably.
Return the recommendations with the signal behind each one 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.
Inventory the behaviour signals the store truly records, covering product views, cart additions, past orders, wishlist entries and search terms, noting for each the system holding it, whether consent covers it, its retention window and the identifier used to tie it to a shopper.
Weight each signal only as far as the supplied data supports it inside that retention window, corroborate a single-event nudge such as a repeat view against a second record before shipping it, and discard any signal whose identifier cannot be matched reliably.
Deliver the recommendations naming the signal, record and timestamp behind each suggestion, verify them against the merchant's own event data, and list every signal considered but excluded for staleness, missing consent or weak attribution.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
It supports a suggestion only when the event exists in the supplied data and sits inside its retention window; browsing alone often reflects comparison or research, so a second corroborating signal is preferred first.
Not by themselves. Repeat views are checked against cart additions or past orders in the same category first, since without that context the assistant risks pushing a product the shopper already ruled out.
It must fall inside the retention period the merchant stated for that signal; older events are excluded even when technically present, because acting on expired intent produces recommendations that read as intrusive.
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.