01
Step 1
Collect the catalog identifiers with current product types and tags, the approved taxonomy with the definition of every node, and the navigation filters and reports that depend on category values.
Back to topic
Spark playbook scenarios · Spark: AI Store Assistant
Classify products using an approved catalog taxonomy
Products entered the catalog faster than they were placed in the approved taxonomy, so navigation now depends on who entered the item. In this case, the immediate task is to classify products using an approved catalog taxonomy. The workflow should produce a category assignment table with exceptions listed, checked against the supplied taxonomy rules.
Topic
Product Tags and Categorization
Target keyword
automatically categorize shopify products
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 classify products using an approved catalog taxonomy.
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: [catalog identifiers with current product types and tags, the approved taxonomy with node definitions, and the navigation filters and reports that depend on category values]. Map each product to exactly one approved node using the node definitions, recording as exceptions any product that fits no node or matches several at once, and never forcing an exception into a nearby category to clear the queue.
Return a category assignment table with exceptions listed, checked against the supplied taxonomy rules 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
Map each product to exactly one approved node using the node definitions, recording as exceptions any product that fits no node or matches several at once, and never forcing an exception into a nearby category to clear the queue.
03
Step 3
Deliver a category assignment table with one row per product plus an exceptions section stating why each item cannot be auto-assigned, then verify the mappings against the taxonomy rules and confirm dependent filters still return coherent results.
FAQ
Common questions when running this scenario
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
How are multi-match products handled?
They are listed as exceptions with every matching node named, and a primary node is chosen only where the taxonomy's own rules allow a selection; forcing a single label by guesswork breaks filtering later, so the exception stands until the taxonomy or the product scope is clarified.
Why can exceptions not be force-filed into nearby nodes?
Because a wrong category propagates into navigation, filters and reports that trust the taxonomy, and the error is expensive to find once buried; an explicit exception queue keeps misfiled products visible and reviewable instead.
What is checked before old category values are removed?
The filters, reports and navigation entries that read the old values are listed and confirmed to work on the new assignments, and the removal is recorded in the change list; a value is only dropped after its dependents are verified, so nothing fails silently.
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.
Assign consistent product tags using explicit rules
Add approved tags to a selected product group
Remove specified tags without disturbing other labels
Prepare scoped updates to product metafield values
Plan a product video that explains the item clearly
Add valid variants to an existing product structure
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.