3 Steps to Go Deeper in Touch Point Personalization for Telcos
By : Naveen Chandra
AGM- Product Management
Customer touch points are arguably the most important interfaces to Communication Service Providers (CSPs) for subscriber engagement. It is not just a platform to provide excellent service, but also an opportunity to enhance the relationship and increase the Customer Lifecycle Value. This is where it becomes extremely important how the subscribers perceive a CSP’s marketing efforts on any touch point. If the marketer’s action or communication does not appeal to a subscriber’s preferences, it runs the risk of irritating them, leading to dissatisfied and angry subscribers.
On the other hand, if the actions are personalized based on anticipated needs of the subscriber, they walk out with more value than they expected. A happy subscriber is bound to stay loyal and keep coming back for more. Marketing actions over a touch point typically relates to extending offers. This blog will explore how managing subscriber intent, through machine-driven offer recommendation and offer prioritization, can potentially take subscriber experience to a whole new level.
Step 1: Offer basket recommendation based on historic persona
With advanced packaged analytics it is now possible to discover hidden insights by analyzing a subscribers past behavior. A holistic view of the subscriber enables designing marketing communication that are well-tailored to a subscriber’s persona and addresses his or her likes and dislikes. With increasing complexity and myriad products and services, there are hundreds of relevant offers that are mapped to a single user. When a subscriber approaches a subscriber touch-point, there is a very small window of opportunity within which the best matched offers must be communicated to them. If the offers fail to grab the subscriber’s attention in the short span, that opportunity to serve and provide a superior experience is lost.
This challenge is tackled by offer basket recommendation, which uses machine-learned algorithms to filter out only the best fit offers that have the highest likelihood of acceptance by a subscriber. The resultant offer basket is a set of uniquely personalized offers that are most suited to a subscriber. For instance, suppose a subscriber, with a non-data enabled handset, logs in to the service provider’s web portal for voice recharge. The browsing history tells the system that he or she has been looking at bundled smartphone offers in the recent past. The subscriber’s webpage will be populated with various data pack offers as offer basket recommendation algorithms comprehend that the subscriber may soon graduate to a data user.
Step 2: Offer refinement based on current context
Subscriber behavior is dynamic and contexts change within minutes. While in one moment the subscriber is engaging in voice calling, the next moment he is browsing cricket scores. Yet at another time he is busy interacting on social media. For marketing communication to be truly effective and experience enhancing, it is important to keep track of these in-the-moment behaviors. To further align marketing communication to changing subscriber behavior, offer refinement algorithm comes into the picture. This algorithm dynamically changes the subscriber’s offer basket based on the constantly changing consumption pattern of the subscriber.
To illustrate this with an example, let us consider a subscriber who is categorized as a heavy local caller among many other attributes. The subscriber approaches the customer care call center with an inquiry. The call center executive, in addition to addressing the subscriber’s query, has a list of best fit offers to read out to the subscriber. In normal circumstances, the offer recommendation basket will feature a ‘local call special package’. However, on checking the real-time context, the machine detects that the subscriber is on roaming. This new status will trigger the automated machine-driven offer refinement algorithm to de-prioritize the ‘local call offer’ with a ‘roaming pack offer’, adapting to subscriber’s current context.
Step 3: Adaptive ranking based on recent touch point behavior
Offer recommendation basket and offer refinement algorithms ensure that the subscriber receives well-tailored as well as contextual and relevant communication across customer touch points. However, at times even the most relevant offers based on subscribers past behavior and anticipated needs may not find favorability with the subscriber. Human beings are impulsive in nature and their response may be triggered by a momentary dislike or a changed circumstance, among innumerable other reasons. In such a situation, badgering the subscriber with the offers he or she declined may ultimately hurt the subscriber experience. Adaptive offer ranking based on recent touch-point behavior enables the machine to learn preferences and re-arrange options accordingly.
For instance, if a subscriber uses only to 2G data services on his or her 3G-enabled smartphone and has declined 3G data pack offers in the past, the machine will direct the offer ranking algorithm to move 3G offers down the list and limit recommendations only to 2G data offers for certain time period. This ensures that subscriber’s preference or non-preference for a certain product is honored.
Machine-driven automated offer recommendation, prioritization as well ranking through packaged analytics enables CSPs to keep constantly in tune with the subscribers needs and delivering intuitively what they want, ensuring proactive communication at all times.
A Case Study
A global CSP with more than 170 million subscribers in India was troubled with high offer-rejection rates at its various touch-points. To address the issue, it integrated Flytxt’s Intent Management Application with its customer care portal. The application uses advanced behavior analytics – automated recommendation and offer prioritization capabilities to identify customer needs in real-time and provide highly personalized service recommendations across various touch-points.
With relevant, contextual and dynamically adaptive offer communication, the CSP was able to provide significantly superior customer experience and as a result, their marketing offers raked in high acceptance rates. Conversion rates shot from 2.1% to 11%, call hold time improved by 40%, and offer decline reduced from 5% to 0.17%. The application also enabled the preparation of in-depth performance reports of agents and products so that a profit-based evaluation could be done more precisely and informed business decisions could be made.
Understanding subscribers at a deeper level and enabling a high degree of touch-point personalization helps elevate subscriber experience, which in turn paves way for stronger relationships. And the stronger relationships CSPs build, the better is success rate.