01
Step 1
Collect the customer records with their identifiers and order references, the matching rule the merchant accepts, and the address, phone and name fields that disagree between candidate pairs.
Back to topic
Spark playbook scenarios · Spark: AI Store Assistant
Identify likely duplicate customer records for review
Several customers look like the same person, but merging them on a guess would destroy order history that cannot be restored. In this case, the immediate task is to identify likely duplicate customer records for review. The workflow should produce duplicate groups with reasons and retained source identifiers, leaving anything the records cannot settle open.
Topic
CRM and Contact Management
Target keyword
shopify customer deduplication
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 identify likely duplicate customer records for review.
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: [customer records with their identifiers and order references, the matching rule the merchant accepts, and the address, phone and name fields that disagree between candidate pairs]. Score each candidate pair only on the fields that rule names, recording the evidence that makes a pair confident or doubtful, and refuse to group two records when the only agreement is a weak field such as a shared last name.
Return duplicate groups with reasons and retained source identifiers, leaving ambiguous pairs open for review 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
Score each candidate pair only on the fields that rule names, recording the evidence that makes a pair confident or doubtful, and refuse to group two records when the only agreement is a weak field such as a shared last name.
03
Step 3
Deliver the duplicate groups with the matching evidence and both source identifiers retained, mark every pair the records cannot settle as open for review, and leave the underlying records unmodified.
FAQ
Common questions when running this scenario
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
When is a match confident enough to propose?
When the fields named in the merchant's matching rule agree and the order evidence points the same way; agreement on a single weak field such as a last name is recorded as doubtful instead.
What happens to order history when records are grouped?
Nothing during this pass, because the output only proposes groups and keeps both identifiers; merging stays a separate decision the merchant makes once the evidence has been reviewed.
How are disagreements inside a group handled?
Each conflicting field is shown with both values, precedence is applied only where the merchant has stated a rule, and everything else stays open rather than silently picking one value.
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.
Organize customer information for practical CRM use
Organize customer contact records for reliable follow-up
Compare new and returning customer behavior
Identify patterns in customer purchase behavior
Draft useful blog content from a search-intent brief
Identify possible churn using relevant purchase history
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.