IDC TECHNOLOGY SPOTLIGHT

Brought to You by NETSCOUT, Powered by IDC

Complexity from 5G and MEC deployments is extreme. Successful business operations

must involve automated data analysis to provide the directional insight for satisfying
customer needs and addressing changing market conditions. Ignoring ML- and AI-driven
data analysis is no longer an option for business success in the 5G era.

In the 5G Era, the Customer Experience

Matters, and Here's What to Do About It
December 2021

Written by: Karl Whitelock, Research Vice President, Communications Service Provider Operations and Monetization

Introduction AT A GLANCE
5G and multi-access edge computing (MEC) are often noted as disruptors
to the traditional ways that communications service providers (SPs) do » Net Promoter Score (NPS) is not the only
business. These technologies generate mounds of operational data while measure of customer service quality.
bringing an enhanced level of payload-carrying capacity for each message What is more important is how a customer's
end-to-end usage experience plays out.
stream and voice/video chat session. 5G/MEC also adds complexity to
Uninterrupted voice/video chat, smooth
already complex network operations, making service-level monitoring a streaming content, and gaming without
long-term front-runner to quality experience management. delay are some of the items on a growing
list of expectations. 5G only increases
Advanced service assurance and customer experience analysis takes on customer hopes of flawless service
many forms and can deliver value in several ways. However, data analysis performance.
within a hybrid 4G/5G/MEC network raises several concerns such as: » Automated data analysis with built-in ML
and AI is a must for any 5G deployment
» Do communications SPs have a way to collect, channel, and analyze strategy. Actions within this network occur
at a frequency and volume too great for
network data for customer experience insight based on the influx of
step-by-step human involvement.
data traversing their networks from 5G/MEC technology in concert
with existing 4G and fixed broadband data sources?

» How can the wealth of data for network operations and service quality analysis be placed into the customer
experience processes to deliver insight about device and network compatibility along with performance?

» With 5G network slicing, new operational needs must be addressed to ensure service quality and availability,
including data analysis at the telco edge. How can real-time network data be used to meet customer service
expectations and to avert problems within an SLA-driven operations environment?

» Can existing data collection, compute, and storage mechanisms create insight about user device behavior for a
section of the network or identify impacted customers when a service-level problem is detected?
This document notes several new assurance-related business and operational needs. It explores how customer service
problems can be isolated or even averted with the right combination of business strategy, a purpose-built toolset, and
industry expertise able to organize these capabilities to meet the strictest service quality controls. The document also
notes how the right service support attributes can deliver far-reaching benefits to customers — human and machine.
IDC TECHNOLOGY SPOTLIGHT In the 5G Era, the Customer Experience Matters, and Here's What to Do About It

5G Complexity Is the New Customer Experience Reality

In the 5G era, network slicing provides a way for customers to receive low latency, high bandwidth, and ultra-fast
connectivity services that can be unique in various configurations to the type of service needed and to the specific
customer requiring the service. Best fit no longer applies. Examples abound such as uniquely defined multiparty video,
virtual reality (VR) based on application emphasis, gaming with normal or "turbo boost" responsiveness, augmented
reality (AR), industrial automation by industry or machine requirements, connected events, surveillance/detection,
connected agriculture, nonsurgical "connected health" medical procedures, and connected public transportation.
For these examples and thousands of others, complexity is the norm with no two service offerings likely the same.
However, in each case, the fulfillment, assurance, and monetization functions must work collectively to deliver the
service, continuously monitor for performance anomalies, identify changes when they occur, and automatically adjust to
stay aligned with SLA-defined service conditions. The key is monitoring for inconsistencies followed by actions that are
executed because of complexity, not by manual methods, but by a data-driven, intent-based business approach.
Network data must be continuously collected, analyzed, and meshed with customer information for organizations to get
an accurate picture of the customer experience. However, this measurement approach also brings many new business
and process challenges. For example:

» How can customer happiness with each service be managed

consistently and reported regularly when the connectivity path involves Network data must be
a hybrid array of differing network pathways each with uniquely
defined SLA parameters?
continuously collected,
analyzed, and meshed with
» What type of data analysis is needed for each business scenario? customer information for
How will configuration and monetization flows be validated?
Can network data anomaly detection occur and corrective actions
organizations to get an
happen in near real time? accurate picture of the
customer experience.
» How can virtual network function usage be ensured for availability and
performance? How will end-to-end (E2E) service operability be
measured against SLA definitions?

» How can a problem be noticed before it causes less than desirable service quality issues for a section of the
network, a group of customers, a specific individual, or a device from a group of thousands?

» How will E2E revenue flow be accounted for and adjusted if SLA definitions are violated? How will detailed billing
for partner contributions be validated? How will SLA insight be communicated and corrective actions taken?

» Will existing data collection and analysis solutions still be useful for managing customer value?
Beyond consumer voice and data connectivity services, some communications SPs are expecting the density of
connected devices to approach 1 million per square kilometer as 5G/MEC plays out. At this level, how will network
operability parameters be monitored, managed, maintained, reported, and analyzed for issues that affect customer
experience, especially when the customer is not likely to be a human?

#US48449921 Page 2
IDC TECHNOLOGY SPOTLIGHT In the 5G Era, the Customer Experience Matters, and Here's What to Do About It

Such interactive involvement requires not only network operations to be in real time but also service-level delivery,
assurance, and monetization functions to operate at the network edge to satisfy application-sensitive latency
expectations and then periodically "synch up" with the core to maintain operational integrity. Such a requirement is
brand-new territory for the industry and opens a new dimension of assurance needs, especially with partner-level
business concerns tied to component delivery, end-to-end service integrity, and access security.

Automation and AI Are the Key to Real-Time Customer Experience Management

The customer experience processes consist of many attributes and can yield a variety of outcomes according to customer
need and business intent. Data-driven tools using machine learning (ML) and artificial intelligence (AI) algorithms require
situational awareness matched to key performance indicators (KPIs) in much the same way as humans engage when
troubleshooting network and service issues. ML/AI-driven assurance and customer experience insight creates the right
workflow that can detect and isolate a particular problem. Then, based on initial findings, the tools use advanced
workflows to identify groups of customers with common attributes. For example, service issues with a particular type of
device may be detected in only a subsegment of the network. By noting other device characteristics such as software
release version and last update status compared with all network registered devices, the data-driven tools can help
organizations better understand the potential scope of an expanding problem. Mashing up network details with
customer information then creates a list of customers that can be sent to the customer experience processes for further
treatment, whether to directly reach out to customers or to place their accounts on a watch list for matchups when
service troubles are reported.
Traditional visibility, trending, and reactive assurance management functions must transform into proactive, predictive
and, eventually, self-optimizing capabilities (see Figure 1). Automated service delivery and real-time experience
management bring new requirements, all of which carry a real-time focus around meeting customer expectations.

FIGURE 1: 5G Real-Time Customer Experience Management

Traditional Service Delivery
Resource
Inventory and
Configuration
Automated Service Delivery Evolution
Catalog
and Activation
Dynamic Resource Event-Based Service Self-Supporting Service
Management Orchestration Perpetuation
Traditional Service Assurance
Proactive Predictive Self-
Visibility and Reactive Node Assurance Assurance Optimization
Trending Management
Real-Time Experience Management Evolution
Visibility and Trends Reactive Management
Customer Experience Management
▪ Dynamic Discovery ▪ Fault Monitoring
▪ Physical and Virtual Paths ▪ Performance Monitoring Company/Employee Experience Management
▪ Network Core-Edge- ▪ Network SLA Mgmt.
Customer Edge-Enterprise ▪ Service Monitoring Proactive Assurance Predictive Assurance Self-Optimization
▪ E2E Service Visibility ▪ Issue Isolation • RT Device and Net Data Collection • Behavioral Learning • Dynamic Anomaly Detection
▪ Multitenancy ▪ Reactive Root Cause Analysis • Statistical Learning • Probabilistic Algorithms • Prescriptive Replacement
▪ Real-Time Analysis ▪ Network Planning • Customer Hierarchy Insight • Inference Probabilities • Intent-Based Operations
• Automated Data Correlation • Automated Service Identification • Service Profitability Mgmt.
Network Management • Anomaly Pattern Detection
• Issue Isolation and Identification
• Automated Notification
• Customer Alert
• Business Assurance
Example:
Network and User Device Monitoring with Automated Customer-Level
Correlation, Trouble Identification, and Notification

Source: IDC, 2021

#US48449921 Page 3
IDC TECHNOLOGY SPOTLIGHT In the 5G Era, the Customer Experience Matters, and Here's What to Do About It

Analytics can also be used to understand if a potentially noted problem shows up only in the radio access network (RAN), or
if it involves the core — or both. What is most important is quickly determining if a detected problem affects all customers
or just some. If the problem affects only some customers, it is also important to understand the level of correlation between
a network or service problem, the customer devices involved, and the geographic locations where each issue occurred.

New Types of Service and Customer Experience Assurance Are Forming

With 5G/MEC, service and device incompatibility issues are likely to multiply. Complexity can arise from the architectural
separation of the user plane and control plane that is now possible in the SA 5G core or from the needed ability to
dynamically deliver and ensure each network slice. Complexity may also be created from the coordination needed between
network components and partner-provided capabilities, as well as from the service delivery and service assurance
processes. The service assurance and customer experience processes can sometimes become misaligned between
technology generations or even disengaged from the monetization processes due to increasing levels of complexity and
dynamic change.
With complexity and new real-time operating requirements, external forces and sometimes internal personnel will scan
for ways to fraudulently obtain value from network and/or partner delivery points. As such realities gain momentum,
understanding how the network interacts with devices of all varieties at a deep level is paramount for long-term service
quality efforts and legitimate service delivery not easily associated with group-level network behavior. Some of the less
familiar 5G business assurance functions needed for service effectiveness include:

» Dynamic partner availability and delivery assurance. Partner

value-add resources such as solution-enhancing capabilities, digital ML/AI-driven assurance and
content, or even physical goods, when included within a connectivity customer experience insight
pathway, are what makes 5G business solutions appealing. How will a creates the right workflow
communications SP know if each contributing partner delivers what it
contracted to do with the right resources and the right configuration at
that can detect and isolate
the right time for each service definition? How will a partner know if the a particular problem.
network connectivity resources are configured to its specifications for the Then, based on initial
value-added capabilities it brings? What degree of auditing is necessary findings, the tools use
to minimize operational risk? Can all partners be implicitly trusted? advanced workflows
» Real-time slice charging and quality assurance. A network slice may be to identify groups of
defined for single time use when certain network conditions are customers with
applicable. Slices could also be defined to address specific customer common attributes.
needs. How will the right pricing definition and charging rules be
monitored for accuracy? How will quality assurance attributes be
checked against pricing definitions? How will slice-specific charging be applied and monitored for accuracy,
especially if different SLA definitions are engaged on time-based usage strategy increments? How will SLA
definitions be dynamically "changed out"? For example, service operation pricing between 6:00 a.m. and 11:00
a.m. uses a 3x normal pricing multiplier, while service operation pricing between 4:00 p.m. and 8:00 p.m. uses a
0.8 pricing multiplier. The SLA is defined as latency sensitive during the early shift and normal on the evening shift.
These changeable parameters could be altered according to network demand, time of day, day of week, local
events, traffic type, data instead of voice, and an endless stream of related parameters.

#US48449921 Page 4
IDC TECHNOLOGY SPOTLIGHT In the 5G Era, the Customer Experience Matters, and Here's What to Do About It

In the 5G era, business assurance must be an automated function. 5G/MEC enables a large partner ecosystem
for contributing ancillary service capabilities, including compute capacity, data storage, AI/ML applications,
firewall security, on-demand digital content, and even physical goods. This environment is a new playground for
fraudsters, and due to complexity, most 5G/MEC services are a significant threat to the rating/charging/billing
and customer experience processes. Additionally, the level of detail and the rapid response needed make other
methods, such as the traditional approach of batch-based data stream comparisons to find anomalies, an
antiquated remnant of the past.

NETSCOUT Advanced Analytics Solution

To meet the challenges of the new digital age, communications SPs need a new approach for understanding,
maintaining, and continuously improving the level of service quality offered to customers and in understanding
the traffic traversing a communications SP network. Traditional human-centered approaches are ineffective
due to the amount and complexity of data that must be analyzed and the speed at which insight is required.
Methods involving AI/ML algorithms are becoming more popular, but they typically have shortcomings tied to a
lack of "Smart Data," minimal data analysis expertise, knowledge concerning data origination or usage, or
decision-making capabilities when conditions dictate.
The NETSCOUT Advanced Analytics solution combines NETSCOUT's packet-level "Smart Data" framework, AI/
ML technologies with domain expertise, and data knowledge. NETSCOUT states that this holistic analytics
solution not only can identify what's broken or impacting business but can also highlight what is working well
and how services and customers are acting. The NETSCOUT Advanced Analytics solution is designed to provide
a complete RAN-MEC-core view of the network 24 x 7 x 365 by performing deep impact analytics — a view
involving any network, any datacenter, any cloud, and any architecture, 5G or otherwise (see Figure 2).

FIGURE 2: NETSCOUT nGenius Smart Data Platform

Source: NETSCOUT, 2021

#US48449921 Page 5
IDC TECHNOLOGY SPOTLIGHT In the 5G Era, the Customer Experience Matters, and Here's What to Do About It

The NETSCOUT Advanced Analytics engine harnesses NETSCOUT's Smart Data by recognizing business impact
across any communications SP network. By injecting specific domain knowledge into its AI and ML algorithms,
the NETSCOUT Advanced Analytics engine is designed to drive workflow automation, which in turn can isolate
business impact across multiple domains — from access to core. Providing actionable business intelligence that
might have taken days if not weeks to isolate in the past, the NETSCOUT Advanced Analytics solution is designed
to automatically deliver through a visually intuitive display.
The NETSCOUT Advanced Analytics solution can help resolve the increasing amount of network complexity with
5G in several ways, including:
» Smart Data can be used to simplify underlying changes in the 5G network, including virtualization and
cloudification of the core, mobile edge computing, network slicing, and the RAN. The NETSCOUT Advanced
Analytics solution is set to analyze millions of investigative iterations, regardless of network deployment,
to find the issues with the highest impact and their underlying causality.
» Network and operational engineering efficiency can be increased by removing the data overload that
engineers often face when focusing on issues with the highest impact to subscribers and services. The
NETSCOUT Advanced Analytics solution automates discovery of service-impacting events by removing
many of the time-consuming, manual steps now employed.
» The customer experience can be improved by decreasing resolution times. The NETSCOUT Advanced Analytics
solution is designed to help communications SPs deliver increased service quality and an improved customer
experience. Chasing individual trouble tickets is tedious and time consuming and often requires customer
involvement in the quality process. Automated workflows can lower the need for customer interaction within the
troubleshooting process.
Industry and Solution Challenges
Challenges abound anytime there are changes to the status quo. 5G/MEC and new solutions to help address
the complexities that come with new architecture are not immune to challenges given the flexibility 5G brings
on the one hand and the levels of network and service complexity on the other. From customer experience
and data management perspectives, challenges can become opportunities that lead to further challenges.
Some aspects of intent-based customer experience management are necessary for full 5G success, but several
factors can make data-driven experience measures difficult, including:

» E2E service-level complexity. Complexity tied to network composition and partner contributions, along with
managing a growing partner ecosystem, is a real challenge for all 5G/MEC environments. Correlation of
functionality is difficult, and the likelihood that the customer experience will be less than ideal is high.
Understanding how customers are relating to their device-based services and the quality of experience between
their device and the network is very important. AI-driven decisioning is critical in most 5G situations due to time
responsiveness, partner interaction needs, and customer expectation parameters.
» Real-time and dynamic operations. Many actions and reactions need to happen in near real time. This makes
human involvement a less than desirable means for meeting business and customer expectations. Process
automation and intent-based decisioning are the only ways to effectively manage the network, but AI-driven
assurance is complex and takes experience to develop the right focus points for making the customer experience
successful and each service offering profitable. Solutions designed to engage in the customer experience process
also need to learn acceptable and unacceptable boundaries, which takes time and experience.

#US48449921 Page 6
IDC TECHNOLOGY SPOTLIGHT In the 5G Era, the Customer Experience Matters, and Here's What to Do About It

» 5G network slicing. 5G slicing brings numerous concerns that ultimately impact the customer experience, network
operations, and revenue flow. Network complexity and real-time responsiveness are just two elements in an
expanding perfect storm of issues that can lead to the proliferation of fraud opportunities from inadvertent errors
and omissions. Constant and consistent monitoring of the customer experience is essential for enabling business
success but can be very complex. This places added challenges on any type of automated solution strategy.

» Ability to maintain customer loyalty. As business interests in 5G/MEC solutions increase, so will customer
expectations. Gaining access to the right data to validate and verify customer experience measures will always be an
operations challenge. If SLA expectations can be verified through actual service quality measures and show
connectivity without incident, customer loyalty will increase. Failing to measure service quality because of lack of data
or showing service results that are below SLA expectations will result in customers looking for something better.

Conclusion
5G is changing how companies think about doing business by enabling the connectivity of everything at a level that
makes life simpler, easier, and better, often without the end user even realizing that change is happening. 5G is also
enabling customers to expect increased value from the solution offerings created by nearly every industry worldwide.
Central to this change are increased amounts of data generated from devices providing new service offers and from
customers creating more data as they take advantage of those offers.
Without a means for analyzing data in near real time, businesses everywhere are stuck using approaches from the past
that are now misaligned with the realities of today's markets. Automation is the key for addressing business tasks that are
close to impossible for humans to perform. Customer experience approaches involving automation, such as the
NETSCOUT Advanced Analytics solution, can drive workflow automation by harnessing the power of ML and AI
to correlate and analyze millions of data points to produce insights that can lower opex, eliminate most
customer churn, and reduce the time to action when applied to several business processes.

About the Analyst

Karl Whitelock, Research Vice President, Communications Service Provider
Operations and Monetization
Karl Whitelock leads IDC's Communications Service Provider Operations and Monetization global
practice. He offers strategic insight and global perspectives concerning service operations and
monetization functions, formerly known as OSS/BSS. Areas covered include rating and charging, policy
management, partner management, customer experience, revenue assurance and fraud management,
service assurance, network data analytics, service orchestration, and network operations.

#US48449921 Page 7
IDC TECHNOLOGY SPOTLIGHT In the 5G Era, the Customer Experience Matters, and Here's What to Do About It

MESSAGE FROM THE SPONSOR

About NETSCOUT
NETSCOUT SYSTEMS, INC. assures digital business services against disruptions in availability, performance, and
security. Our market and technology leadership stems from combining our patented smart data technology with
smart analytics. We provide real-time, pervasive visibility, and insights customers need to accelerate, and secure their
digital transformation. Our approach transforms the way organizations plan, deliver, integrate, test, and deploy
services and applications. FY 2021 reported revenue was $831M.

The content in this paper was adapted from existing IDC research published on www.idc.com.

IDC Research, Inc. This publication was produced by IDC Custom Solutions. The opinion, analysis, and research results presented herein are drawn from
140 Kendrick Street more detailed research and analysis independently conducted and published by IDC, unless specific vendor sponsorship is noted. IDC
Custom Solutions makes IDC content available in a wide range of formats for distribution by various companies. A license to distribute
Building B IDC content does not imply endorsement of or opinion about the licensee.
Needham, MA 02494, USA External Publication of IDC Information and Data — Any IDC information that is to be used in advertising, press releases, or promotional
T 508.872.8200 materials requires prior written approval from the appropriate IDC Vice President or Country Manager. A draft of the proposed
F 508.935.4015 document should accompany any such request. IDC reserves the right to deny approval of external usage for any reason.
Twitter @IDC Copyright 2021 IDC. Reproduction without written permission is completely forbidden.
idc-insights-community.com
www.idc.com

#US48449921 Page 8