01
Step 1
Sample tickets together with the labels currently applied, the routing destination each label leads to and the handling action that destination performs, so categories can be judged by the action they trigger rather than by their names.
Back to topic
Spark playbook scenarios · Spark: AI Store Assistant
Classify support tickets using explicit categories
Topic labels were added whenever someone new joined the team, leaving overlapping tags that make the volumes impossible to read. In this case, the immediate task is to classify support tickets using explicit categories. The workflow should produce a ticket category scheme inside a support operating workflow validated on unseen tickets.
Topic
Customer Support Operations
Target keyword
shopify support ticket classification
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 support tickets using explicit categories.
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: [sampled tickets with current labels, the routing destination and handling action behind each label, and a held-back ticket set for testing]. Define categories that are mutually exclusive at ticket level and complete for the request types present, write a precedence rule for tickets fitting two, then test the definitions on tickets held back from the ones used to write them.
Return a ticket category scheme inside a support operating workflow 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
Define categories that are mutually exclusive at ticket level and complete for the request types present, write a precedence rule for tickets fitting two, then test the definitions on tickets held back from the ones used to write them.
03
Step 3
Deliver the category scheme inside the support workflow with a definition and an example per category, then report the share of held-back tickets that classify cleanly under exactly one category and list those that fit none.
FAQ
Common questions when running this scenario
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
How is a held-back sample used?
It tests whether the definitions hold on tickets they were not written from, which is the only real check on completeness; tickets fitting no category are evidence for adding one or widening a definition, never grounds for forcing a fit.
What does mutually exclusive mean here?
One category per ticket chosen by the written precedence rule, so category counts add up to the ticket count; secondary labels may be kept for search but must not feed routing or reporting, or the totals stop reconciling.
When should a category be retired?
When it no longer changes routing or reporting and its volume has moved elsewhere, since a category that decides nothing only adds a choice; historical tickets keep the old label so earlier counts can still be reproduced.
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.
Route support tickets to appropriate owners
Design a reliable customer support workflow
Surface recurring customer complaints requiring attention
Draft consistent replies to common customer requests
Create support reply templates grounded in policy
Transfer unresolved chatbot requests with useful context
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.