The Unified Data Architecture is what makes Plumb5 most powerful, allowing business to scale exponentially using automation, speed and accuracy. The architecture is designed by building relationships between data parameters into a structure which behaves like a Turing tape. Such a structure can solve any business related problem arising from data or data automation.
It is important to build a data aware model where captured data is directly stored against the tag. For new data, it is important to build the relationship as it’s being stored. With this, each time data is collected, it can store against the pre-existing tag. If done for each customer/user, it would create a holistic view across all channels and sources. If data is sequenced using time-series, it can create complete customer journeys without gaps. If objective states are set, it can create tags based on stages of the customer lifecycle. With this we achieve a holistic view of a single customer without any gaps. This makes way for accurate analysis.
Now that we have the customer journey of a single customer, successful and unsuccessful patterns can be identified. Assigning meaningful scores to each event can help in creating auto segmentation based on behavioral parameters which broadly comprises of customer interactions, responses, transactions and user sentiments.
The behavioral data provides insights that help in delivering recommendations based on past purchases, wish lists or by patterns derived from users with similar profiles. We can use real-time recommendation models to arrive at product list by scores based on association parameters.
Using a lifecycle grid analysis, you can create states across the customer lifecycle and use the behavior or recommendation scores to trigger communication. A rule library for each state or segment type will allow the machine to trigger communications if a particular rule is satisfied.
After the communication is triggered, the system needs to loop back the responses of the customer with respect to the communication and store data against the pre-existing tag. This ensures that the new responses are added to the existing insights and the new derivative is rendered for next action. Storing against the tag will help maintain single customer data at all times.