Supported use cases

Learn what use cases can be built and deployed using Joule

Introduction to supported use cases

Joule enables business developers build business and technical use cases from pre-build and custom data integrations, event processors and analytics.

The flexibility of Joule offers a wide range of use case enablement from data encryption, real-time metrics, custom analytics to machine learning predictions. Presented below are some use cases which can be used to inspire

Business use cases

This category of use cases generally refers to how to apply customer facing insights, alerts and next best actions.

• Customer consent management Apply static or dynamic opt-In / opt-out lists to drive event filtering.

• Geospatial Intelligence Built geospatial aware applications that track entities within locations of interest, such as shopping malls or stadiums, with respect to time.

• Proactive inventory management Capture live inventory levels to manage and forecast inventory capacity with respect to location to drive sales revenue.

• Sensitive customer data management Remove or mask sensitive customer information (e.g. credit card number) before distributing to consuming systems.

Analytics driven use cases

The widespread use of analytics within use cases is now becoming key to differentiate customer facing offering.

• Analytic driven alerting Combine running metrics and live events to proactively alert on when a system is close to failure within a manufacture setting.

• Stream based predictive analytics Leverage machine learning models within a stream to predict in real-time; e.g. Likelihood to convert to being a valued customer.

• Real-time metrics Define and monitor custom metrics to drive proactive customer support and drive higher brand satisfaction and stickiness; e.g. Improve customer experience for mobile phone usage

• Contextual analytics Join events with slow moving customer profile data to determine next best action; e.g. customer website dynamic promotions.

Data Orchestration

This class of use cases are mainly focused to support the above categories and therefore considered more technical in nature.

• Once only processing Process events only once using event deduplication; Improve customer event processing by only processing critical events once an thereby reduce exposure to erroneous processing.

• Data transformation Prepare data for further pipeline processing by transforming event attributes in to a target state; e.g. Machine learning predictions may require input data elements to engineered in to a suitable target variable using methods such as normalisation.