Data Driven Customer Experience

How many times you feel frustrated with user experience? You have to use the exact same steps to perform a transaction and some of the steps are completely unnecessary (for example to make a payment, you need to go to the Account Details first). Won’t it be nice if the customer experience becomes more personal?

Well, it is time we use the data as goldmine and provide best in class customer experience. In many organizations, data strategy and customer experience strategies are totally isolated from one other, but if they collaborate, we can enhance the customer experience exponentially which will help with customer retention and loyalty (by the way loyalty is hard to come by in the digital age especially with millennials). In the end this means higher revenue and long term growth. Let us see some of the strategies which can be employed to execute that?

1.   Close Collaboration Between Analytics and Customer Experience Team: While creating the scrum teams around the user journeys, add an analytics expert for that journey. The role of the analytics expert will be to perform number crunching using machine learning and predictive analytics techniques, consult the user journey team on how the customer experience should be and make it intuitive and personal.

2.   Invest in Training Data Set: I cannot over-emphasize the need of this. Unless the model can predict right, decision taken for the customer experience will not be right. And, for the model to be right, it needs large amount of training data. In many organizations, this is the most neglected part of the organization, but my advice is get the right tool set, create the right infrastructure and build large samples of training data set.

3.   Self-Healing Infrastructure: One of the best uses of the data will be to predict the scalability and spot issues before they happen in the production systems. It is not a customer facing feature to customer, but it is basic hygiene to be in business 24x7 to serve customers. Data collected from various sources like server usage, network bandwidth, customer volume, transaction types etc can play a huge role. Create predictive model using this data can really help to build intelligent and rugged infrastructure which can be self-healing ( respond automatically and take corrective actions)

4.   Employ Fog Computing Technique: For some of the instant customer gratification, you can employ fog technique in customer experience channels (like mobile or internet). What I mean is that create intelligent data adapters which runs close to the customer facing systems.  These adapters can take certain data from data store and based on what customer is doing on the customer facing channels, provide immediate contextual help or offering. One of the example is, if the customer is looking at the overdue payment and if there is history that customer is behind on the payment most of time, offer customized loan to customer immediately. In addition to showing the loan offer, also show the reduction in the interest rates and personal saving to the customer by switching to this loan.

5.   Plan for A/B testing: When you are ready to roll out data driven customer facing features to your customers, start slow, as this need to be tweaked till you get it right. Most of the time, you have to run it with multiple set of training data to prove your hypothesis. I would also recommend creating a robust A/B testing framework where you can gradually roll out to customers.