Now that we have explored a model to think about applying Data Analytics and Machine Learning to a closed loop system for engaging customer more proactively, let’s take a look how we continuously monitor and improve our system. Here at the Consortium for Service Innovation we are big fans of double loop systems since they are self-correcting.
Looking back at the Event Loop, it is built to help us think about the components that are needed to design a closed system for listening to data and providing contextual, actionable outputs.
To ensure ongoing success, we need to monitor, assess, and continuously improve the Event Loop. Each stage of the Event Loop has a corresponding Improve Loop stage that assesses its effectiveness.
Detection Effectiveness: Are we listening for and collecting the right data at the right time to feed the Analysis and Rules Engine?
Asset Quality: Is the data we have more than just the input data; is it accurate enough to apply in the Analysis and Rules Engine?
Analysis & Rules Effectiveness: Is the ‘engine’ effective in creating the actionable connections and outputs we need?
Engagement Effectiveness: Are the communications channels that are being used reaching the intended audience in a clear and actionable way? Is the audience taking necessary action?
Impact Assessment: When the action is applied, is it having desired impact on the system it is being applied to?
With these components and associated measures or checks in place, we have a robust double loop engagement model.
There are many dependencies and details to work though in making the double loop Predictive model a reality. Some key enablers beyond the technology are:
- Knowing a lot about our products and/or services and how they are used. In order to serve up information, solutions, suggestions, or fixes to customers, we need to know how our products or services are being consumed.
- Knowing a lot about our customers as companies and people. Context is as important as the information being served up to ensure it fits the need.
- Fostering a trusting relationship with our customers. Our actions must be for the benefit of our customer first and foremost.
- Being very clear in our intent. Much of the information we deliver today is polished and intended to drive buying behaviors. Going back to trust, we should be completely open on the intent of the system we are building to engage our customers.
In summary, “How can we provide information that we have, that customers would value, but don’t know to ask for?” is a question that takes some thought and design to untangle. More information about the Predictive Customer Engagement initiative can be found here.
Consortium Members continue to iterate on these ideas and exercise these models. Members can access notes and presentations from the March 11, 2020 Predictive Customer Engagement team meeting on the wiki.