New Key-Influencers Tool Helps Refine Mobile Engagement Score Model

Customer_Engagement_Score-Thumbnail“What gets measured gets improved,” a notable saying goes, and it seems that in few industries is this saying more applicable than in the mobile industry. New technology advancements and new mobile services have fueled an explosion of data that can be can be aggregated and analyzed in any number of ways to help improve quality of service. And this granular insight is opening new possibilities for the way that mobile operators can serve the needs of users in real time.

Yet trying to assess all today’s mobile services can be both wide-ranging and complex. For this reason, Syniverse recently developed an approach that uses a “customer engagement score” that we think offers huge promise to help mobile operators better understand and serve their customers. The score is based on the concept that at the core of mobile services is a subscriber’s engagement with an operator’s network, and that subscriber decisions, like how frequently they use mobile service each day and to what degree they use specific mobile services such as text messaging or data service, can reveal this level of engagement.

Using this score, Syniverse’s Global Business Analytics team has refined a model that seeks to measure how subscribers use roaming services on a foreign network on a typical day. The model assesses four key areas:

  • Data, text messaging, and voice usage patterns
  • Depth of engagement
  • Frequency of engagement
  • Average time between roaming events on the network.

Using this score, we’re beginning to help MNOs better drill down into various segments and better understand customers on a number of levels. And as part of this, we recently integrated a new tool to analyze “key influencers” that is designed to evaluate how certain key aspects of mobile usage can influence a customer engagement score. Specifically, this key-influencers tool measures seven areas of roaming usage:

  • Total number of mobile “events” that a subscriber initiates
  • Average engagement time for each mobile event
  • Total data sessions
  • Total megabytes used
  • Total text messages sent
  • Total minutes of voice service used
  • Total calls placed.

To understand the key influencers in even greater depth, we’ve taken the data from these seven areas and separated subscribers into three groups – a low-engaged roamer group, a medium-engaged roamer group and a high-engaged roamer group.

As an example, the pie chart below provides a snapshot of a medium-engaged roamer group for one customer and a breakdown of the metrics that influence the score (behavior) of this roamer group. In this case, 69 percent of this group’s engagement score is being influenced by three metrics. The first influencer is the total event count, which is a measurement that is taken based on the time stamp of every event pertaining to the subscriber. An event can include any number of separate actions by a subscriber, including placing a call, receiving a text message, accessing the Internet, or turning off a phone. The second influencer is the number of data sessions. And the third influencer is the average engagement time between events, which involves a complex algorithm that was devised by the Global Business Analytics team to measure this metric.


Based on our initial results for this key-influencers analysis tool, we’re optimistic about the potential of continuing to improve the development of a customer engagement score and other similar measurements now emerging in the mobile industry. This granular insight is opening new possibilities for better serving the needs of users in real time. For example, MNOs can use it to enhance everything from the response rate to targeted roaming service offers and marketing campaigns, to the reaction to price plan changes or new service offerings, to the health of a mobile network (traffic levels, trouble reports, roaming registration attempts and failures, etc.). And in this way, this data demonstrates that “what gets measured” can truly “get improved.”

I would love to know your thoughts. What do you think are some of the most important ways that measuring customer engagement in more detail can help us significantly improve service to mobile users?

Tags: , ,


About the Author ()

Rasha Salaheldeen joined Syniverse in 2013 as Systems Engineer, Business Intelligence, on the Global Business Analytics team in Syniverse’s Enterprise Solutions organization. In her role, she focuses on supporting Syniverse’s global sales teams in different regions with solutions to analyze and improve different areas of mobile usage. These have included high-performance business intelligence solutions and statistical models that have been automated for specific customers to enhance areas related to mobile device type, machine-to-machine (M2M) communication, mobile intelligence, mobile engagement and text messaging. Rasha holds a bachelor’s degree in management information systems from the University of South Florida and is currently completing a master’s degree there in business intelligence.

Leave a Reply