The same customer can appear as a store purchaser, an ecommerce account and a loyalty member. If those records stay disconnected, different teams can make decisions from different versions of the relationship.
Identity resolution use cases start with a decision that improves when the right records are connected. Define the entity, the history and the receiving workflow before choosing an audience or promising a business result.
Key Takeaways
Retention workflows need customer history and eligibility rules that hold across channels.
Analytics needs a clear distinction between people, accounts, households and transactions.
Loyalty use cases require explicit links between membership and purchase activity.
Paid-media audiences need usable destination identifiers as well as resolved profiles.
Identity resolution supports these workflows but does not, by itself, authorize contact or establish campaign lift.
Retention: recognize a purchase across channels
Consider an illustrative retailer whose ecommerce audience targets customers who have not bought recently. A customer who purchased in a store yesterday may still appear lapsed if the store record isn't connected.
The workflow needs resolved customer relationships, purchase timestamps, returns and the team's definition of recency. It also needs eligibility and suppression rules for the planned channel.
Test known examples across both channels. Confirm whether a qualifying purchase changes the profile, audience membership and receiving system as expected. Use a subsequent experiment to evaluate campaign performance.
Analytics: measure customers at the right grain
An analyst may need to count distinct customers, calculate repeat purchase or compare cross-channel behavior. Each question depends on what the customer identifier represents and which transactions belong to it.
Keep a person-level view distinct from a household or account view. Reconcile source totals and investigate identity changes before interpreting a shift in customer counts as a change in business performance.
Our guide to identity resolution techniques explains matching approaches. The analytical acceptance test still needs a clear entity definition and a documented calculation.
Loyalty: connect membership with purchase history
A loyalty team may need to understand members who purchase through more than one account or channel. The source set should include membership identifiers, relevant transactions and the business rules governing attribution.
Define how shared household details, duplicate enrollments and changed contact information should be handled. Connecting records should not silently transfer benefits or overwrite membership policy.
For a published example, Wyndham's customer story describes profiles for loyalty and non-loyalty guests and more specific audiences. Treat that as context for the decision, not a guarantee of program growth.
Paid media: build audiences and exclusions
Resolved purchase histories can help teams define customer audiences or identify existing purchasers for exclusion. The destination still needs identifiers it accepts, and the account must meet its own audience requirements.
Test the route from a known source record to the audience in the ad platform. Distinguish a correct customer grouping, an available identifier, a successful delivery and a destination match.
A higher destination match rate does not prove that every underlying profile is correct. Keep the quality test separate from the reach and campaign-performance measures.
Customer experience: prepare context for supported workflows
A service or personalization workflow may need to recognize a customer and retrieve relevant context. Determine which identifiers the surface receives, which fields it may access and how current those fields need to be.
Identity recognition during an interaction and broader identity resolution are different operations. Amperity's real-time documentation separates real-time concepts from batch processing.
Test known, unidentified and newly identified customers. Avoid assuming that every historical relationship or profile aggregate is recalculated at the moment an event arrives.
Make each use case testable
For each use case, write a short acceptance plan:
Decision: what the team will do differently with connected records.
Data: which sources, identifiers, history and eligibility fields are required.
Identity: which entity is resolved and which difficult cases need review.
Output: which profile, report or audience the workflow produces.
Verification: how the team checks the result and observes downstream behavior.
Preserve source lineage so reviewers can explain the output. Include an owner for investigating exceptions and a reason to repeat the test after material source or policy changes.
Prioritize one complete workflow
Choose a use case with clear business ownership and observable results. The build-versus-buy framework helps identify operational responsibilities, while the cost of poor data quality helps frame avoidable reconciliation work.
Explore our identity resolution capabilities with that workflow in mind. The first useful milestone is a decision your team can support with connected, reviewed customer evidence.



