01
Step 1
Collect the recurring request types ranked by how often they appear, the approved policy answer behind each, and the order fields a reply must quote such as order reference, dispatch status and the delivery window shown in the record.
Back to topic
Spark playbook scenarios · Spark: AI Store Assistant
Draft consistent replies to common customer requests
Order-status questions dominate the inbox, and answers written on the spot commit to different refund wording depending on who replies. In this case, the immediate task is to draft consistent replies to common customer requests. The workflow should produce a support operating workflow and sample ticket outcomes tied to each request type.
Topic
Customer Support Operations
Target keyword
shopify automated customer replies
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 draft consistent replies to common customer requests.
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: [recurring request types by frequency, the approved policy answer for each, and the order fields a reply must quote]. Write one approved reply per type that uses the same wording for the same facts and takes every value from the order record, and record for each type the conditions under which an agent must answer personally instead.
Return sample ticket outcomes inside a support operating workflow for every request type 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
Write one approved reply per type that uses the same wording for the same facts and takes every value from the order record, and record for each type the conditions under which an agent must answer personally instead.
03
Step 3
Deliver the sample ticket outcomes inside the support workflow with each reply shown against a real ticket, then check every reply against the policy text it came from and flag any sentence committing to something the policy does not guarantee.
FAQ
Common questions when running this scenario
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
What makes two replies consistent?
The same facts produce the same commitments in the same words, even where greetings differ; a reply that adds a promise the policy does not make is a defect however helpful it reads, because the customer will hold the store to whichever version they received.
How are requests covering two topics handled?
They are split into one reply per topic or handed to an agent, since merging two policies into one message creates a combined commitment neither policy supports; the ticket keeps both topics recorded so the second one is not lost.
Who approves a change to an approved reply?
The role that owns the underlying policy, with the effective date recorded against the reply; replies built on a superseded policy are retired rather than edited in place, so a customer's earlier answer can still be explained.
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.