Harnessing advanced analytics to improve customer experience

Subscribers are becoming more and more demanding. They expect personalised offerings, ubiquitous access, broad choices and a reliable, seamless experience. Meeting these needs requires a fundamental rethink of a service provider's CEM (Customer Experience Management) infrastructure.

Subscribers are becoming more and more demanding. They expect personalised offerings, ubiquitous access, broad choices and a reliable, seamless experience. Meeting these needs requires a fundamental rethink of a service provider's CEM (Customer Experience Management) infrastructure. 

Today, most service providers use offline data analysis for reporting, planning and CEM purposes. Their current analytics infrastructure does not provide them with the kind of real-time advanced analytics capabilities that can help them continuously monitor and respond to VIP customer issues in real time, engage in personalised marketing in real time, conduct sophisticated network planning and proactively detect and prevent fraud. 

Advanced analytics driven CEM solution must have an underlying smart data infrastructure to efficiently support the data volumes, concurrency and query complexities involved, along with capability to support complex event processing. While the data infrastructure is critical, this must be surrounded by an information management environment that feeds it clean, trusted information.

Critical services in this environment include data integration, data quality, data profiling and master data management. The final component is an advanced analytics driven environment that allows both interactive users and automated processes to efficiently access and derive insight from the data at a granular level to drive optimised decisions.

The advantage of a service provider using advanced analytics is that it can help service multiple functions across an organisation. An effective analytics driven strategy involves:

  • Creation of an architecture which enables the collection, storage, and integration of data sets, from a variety of systems

  • Applies advanced analytics techniques in order to identify patterns of significance across those data sets (perhaps also providing root cause, predictive and outcome analysis, undertaking complex event processing and providing multi-variant business activity monitoring)

  • Facilitates the delivery of actionable, context-specific insight to end users

An effective analytics solution must be able to access data, analyse it and provide the results of that analysis on-demand, so that end users (which can be either people or technology systems) have the insight they need to make the better decisions, without delay. In a world where more and more customers interact online, and talk about their experiences and issues online, online brand management has become big business. 

Operators who ignore what customers say about them in unstructured environments risk swift and widespread brand damage. Making sense of structured and unstructured data to understand the mood and transaction pattern of customers is therefore critical, as is social network and sentiment analysis. 

Human beings as well as IT systems are generally more comfortable and adept in handling known problems or issues. Both look for commonality of situations and feel at ease taking measured approach to well understood problems. However, both humans and IT systems are not adroit at handling situations or conditions which are unstructured and relatively unknown to them. These blind spot areas need to be discovered, analysed and then addressed which requires critical judgment and efficient real-time advanced analytics driven solutions. 

So the larger question which looms is what operators can do to be prepared about the unknowns. Operators believe that there are few things they can do better - have a systematic process, technology and analysis team in place so that constant analysis can result in discovery of unknown problems, have resolution to known issues automated as much as possible so that staff can focus on unearthing unknown issues. 

Advanced analytics can help operators unearth those unknowns and help take preventive actions so they can avoid churn or customer dissatisfaction by providing targeted promotions or pre-emptive service assurance. There is need for solutions which can combine customer usage and subscription data with insight into the network, cost, revenue, supply chain, stock control, customer mood, social network sentiment and customer preference data to trigger specific actions which helps in accurate network planning, intelligently and cost effectively manage content caching, enhance customer experience and impart customer loyalty. 

Advanced analytics can properly identify those customers who have a higher propensity to churn and possibly take those in their social circle with them. Using advanced analysis allows CSPs to shift their business intelligence focus from looking at old data to looking at current data in a predictive and preventative fashion. The key to an advanced analytics solution providing optimal churn mitigation will be its ability to process information about all interactions that impact the customer experience, including network coverage, bandwidth consumption, billing information, support history, and device type. 

Quick responses to customer issues can help keep a subscriber happy throughout his or her customer lifecycle. Dynamic real-time or near-real-time offer management capabilities based on subscriber network usage and traffic-based promotion, loyalty points, event-based promotion and rules-based promotion will be critical for an operator’s revenue optimisation strategy. Use of analytics effectively can provide service providers key benefits in business areas of concern, such as marketing campaigns, contract negotiations, churn management, customer loyalty and operational processes. 

CSPs need to tailor pricing appropriately to strategic customer segments for both new and renegotiated contracts. Analytics driven pricing strategies can assess the different combinations of customers, competitors, products and offerings. The resulting modified pricing models can be tested using analytic means to ensure that operations can scale to support anticipated growth. In previous years CSPs typically focused on retaining customers based only on profitable ARPU. However, the models now include non-financial factors as well. Customer profitability analytics can help to streamline operational processes and reduce associated costs. 

One obvious key factor here is agility as differentiating and building a competitive edge in communications industry demands quick time-to-market changes. Fast translation of analytics results into operations will make a timely and sizable difference when acquiring new customers and reducing attrition risks with informed interactions. In addition, there is an opportunity to identify which operational processes affect customer value and how to streamline costs to improve profitability.