Identity resolution attempts to recognize people across channels and devices to create one unified customer profile. Connecting the dots across omnichannel customer journeys is one of the biggest challenges that marketers face today. With unified profiles, you can gain deeper insights which will help you deliver more personalized messages and offers.
Events SDK feature identifies your website’s visitors as customers if they fill in the email or phone number field. We can identify if a customer is a visitor, prospect or existing customer by tracking their online behaviour.
- If cookie is created on device and collected add new Customer in Customer API as visitor
- If same customer has new cookie on new dev ice it is treated as new visitor in Customer API until we match it with existing Prospect or Customer
- If cookie has a new registration with no purchases or fills contact form, it is defined as prospect in Customer API
- If prospect purchase is identified as event, we turn it into customer
- If we identify new visitor with existing customer, we match the new cookie and delete a new visitor.
Note:
- Customers who don't have transactions are still CUSTOMERS
- People who bought something on e-shop but haven't registered are CUSTOMERS
- Visitors who left any id (email or phone number) are PROSPECTS
- Visitors who didn't leave any id (email or phone number) are VISITORS
- Prospects who didn't give consent are PROSPECTS
- Prospects who gave consent are PROSPECTS
Exact matching → deterministic |
Fuzzy matching → probabilistic |
Exact Matching is a deterministic approach to linking together customer actions and attributes using first-party data that the customer has provided (app login, newsletter subscribe, loyalty program data, web shop login, etc). Exact Matching brings as close as 100% accuracy since it's based on the explicit customer actions and attributes. |
Fuzzy Matching is a probabilistic approach to linking together customer actions and attributes using complex statistical and predictive algorithms, and variety of data points available in the system. Although the accuracy is lower than with Exact Matching, Fuzzy Matching provides more matches across the customer landscape, with the tolerance for accuracy being controlled depending on the use cases. |
Comments
0 comments
Please sign in to leave a comment.