CSPs face many challenges including intense competition, pressure to reduce costs, and changing customer demands. Advanced analytics can help CSPs better understand customer behavior to improve customer retention. Network analytics can also help CSPs monitor network performance, identify bottlenecks, and utilize their networks more efficiently. Overall, analytics can provide insights to help CSPs improve customer experience, maximize network efficiency, and remain competitive.
How U.S. Telecoms Can More Effectively Convert Data to Foresight
1. • Cognizant Reports
How U.S. Telecoms Can More Effectively
Convert Data to Foresight
Executive Summary a more programmatic use of advanced analytics
The business environment has never been more to provide superior customer service, resulting
challenging for communication services providers in improved customer retention. This requires
(CSPs). Buffeted by a growing threat from non-tra- the backing of the organization’s leaders and a
ditional players, pressure to reduce costs, drifting cultural shift toward fact-based decision-making.
customer loyalties and a dynamic technological As part of the due diligence process, CSPs should
landscape, they must also deal with the perennial consider working with partners capable of provid-
issues of billing abnormalities, subscriber churn, ing analytics as a service, which not only delivers
revenue leakage and call failures, among other cost advantages but also leverages experts who
issues. Above all, intense competition has CSPs can jump-start the process and accelerate time
venturing outside their comfort zones to improve to value.
customer experience and avoid being outmaneu-
vered by competitors.
Driving Forces
CSPs are banking on the customer data gener- CSPs operate in an increasingly tough environ-
ated by operational systems day-in, day-out to ment. The industry is characterized by intense
provide valuable insights about their custom- competition, a growing threat from non-tradi-
ers’ tastes and preferences. However, this data is tional competitors, a rapidly changing technologi-
complex and runs into terabytes, putting intense cal landscape and increasing customer demands
pressure on existing systems and processes. (see Figure 1, next page). Among the challenges:
The end result: CSPs often have little or no clue The U.S. market is highly competitive, with
about their customers’ actual requirements. With CSPs vying for customer affection by offer-
increasing customer demands and the availability ing services at competitive prices, resulting
of facilities that provide easy movement to the in reduced margins. Average revenue per
competition, CSPs need to undertake proactive user (ARPU) from traditional voice services is
measures to minimize customer attrition. steadily declining, countervailing the increase
in data ARPU1 (see Figure 2, next page).
Advanced analytical solutions can help CSPs The rise of over-the-top (OTT) players such as
understand and predict customer behavior and Google, Yahoo and Skype, which use CSP net-
make quick and effective decisions. Combined works to provide voice, video, messaging and
with network analytics, which help in utilizing other services freely to consumers, directly
networks more efficiently, CSPs can benefit from impacts CSP revenues.
cognizant reports | september 2011
2. CSP Challenges
Intensifying competition 57%
Efficiency improvements 40%
Growing customer expectations 37%
Reduced time-to-market 30%
Reducing complexity of the company's structure 30%
Keeping pace with new technology trends 20%
Sustaining profitable growth 20%
Network congestion 13%
Dealing with regulatory issues 7%
0% 10% 20% 30% 40% 50% 60%
Source: Frost & Sullivan
Figure 1
Average Revenue Per User Deep-Dive
$60
$50
$40
$30
$20 Q2 2013
$10
$0
Q310
Q304
Q105
Q205
Q305
Q405
Q106
Q406
Q207
Q307
Q407
Q108
Q308
Q408
Q109
Q209
Q409
Q110
Q210
Q410
Q111
Q404
Q206
Q306
Q107
Q208
Q309
Data ARPU Voice ARPU Linear Data ARPU Linear Voice ARPU
Source: “U.S. Wireless Data Market, Q1 2011 Update,” Chetan Sharma Consulting
Figure 2
CSPs must keep pace with growing technol- Following the global downturn, CSPs have
ogy complexity and equip themselves to meet been under immense pressure to improve
the growing demands of their subscribers. operational efficiencies and cut costs, while
The convergence of IT, telecom and media has ensuring quality of service.
led to the development of new products (e.g.,
smartphones, tablets, etc.) that are driving
In addition, CSPs must regularly contend with
demand for both data and personalized ser-
maintaining customer loyalty, a challenge made
vices. For instance, data will account for 51%
more difficult with the introduction of portable
of overall wireless spending by 2014, accord-
mobile numbers. Other challenges include billing
ing to the Telecom Industry Association (see
abnormalities, revenue leakage and call failure,
Figure 3, next page). This requires CSPs to plan
among other issues, which need to be addressed to
and invest in infrastructure to accommodate
protect revenues. Under pressure to improve prof-
increased demand for network bandwidth.
itability and quality of service, CSPs have placed
cognizant reports 2
3. Traffic Growth
4,000
3,500
3,000
Percent
2,500
2,000
1,500
1,000
500
0
2009 2010 2011 2012 2013 2014
Average Data Growth Traffic Growth per Site* Tech-Adjusted Traffic per Site**
Source: “TIA 2011 Playbook,” Telecommunications Industry Association, and “FCC Report – Mobile Broadband: The Benefits of
Additional Spectrum,” October 2010.
* Data demand adjusted for additional cell sites.
** Data demand adjusted for special efficiency improvements.
Figure 3
renewed focus on customer retention and, along bottlenecks, address capacity concerns and uti-
with it, the tools to improve customer service. lize network infrastructure intelligently. Further,
by deploying analytics to streamline business
operations, CSPs can reap sustainable efficien-
Customer-Centricity Imperative cies, which could prove to be a game-changer.
According to a Bain & Co. study, a 5% increase
in customer retention can improve profitability
by 75% for a company across many industries. Analytics for Improved Customer
With traditional communications services being Relationship Management
increasingly commoditized by the emergence of Building long-term profitable relationships with
IP-based networks, CSPs have put greater empha- customers is the key for any organization. Consid-
sis on improving the customer experience in a ering CSPs’ huge, diversified customer bases, with
bid to extend customer retention. Achieving this their varying service requirements, customer rela-
requires a greater understanding of customer tionship management becomes an arduous task.
behavior and needs, as well as superior service.
Prohibitively high customer churn rates in the
The good news is that CSPs already possess rich early 2000s forced CSPs to invest significantly
and abundant customer data. Leveraging this in CRM systems, with the hope of achieving radi-
data through advanced analytics will enable CSPs cal improvements in customer relationships and
to develop deeper insights into customer usage profitability. However, a majority of these CRM
and preferences, predict their future require- initiatives failed to realize their full potential, as
ments and make more effective business deci- CSPs struggled to mine the insights contained
sions. CSPs should also ensure that customers within these systems.
are connected and have access to persistent ser-
vices around-the-clock, which requires efficient Merely collecting customer information through
network capabilities. CRM systems does not aid in decision-making.
Consolidating and analyzing data related to sales,
Networks are the key to CSPs’ business strategy, marketing and service provides deep insights into
and their performance is vital to customer satis- what customers want. Analytics provide CSPs
faction. That’s why CSPs focus on network health with the power to analyze enterprise-wide data
as their top priority in order to avoid sudden and make effective tactical decisions that can
outages. Network analytics allow CSPs to con- improve customer satisfaction at various stages
tinuously monitor network performance, identify of the customer relationship, from acquisition,
cognizant reports 3
4. to improving customer spend and retention. If Further, marketers can identify prospective cus-
properly configured, CRM systems provide a tomers, predict customer needs in advance,
360-degree view of customers and enable differ- design targeted marketing campaigns and pro-
ent groups in an organization to leverage a single vide customized services, which will improve cus-
version of the truth. tomer experience, ultimately increasing customer
loyalty and profitability (see Figure 4).
Embedding Analytics Across the Customer Lifecycle
Prospecting and
Growth Manage Retain
Acquisition
Customer
Social Media
Profitability Customer
Analytics
Analytics Segmentation Churn Analysis
Social Network
Customer Product Customer Risk
Analysis
Lifetime Value Affinity Analytics
Analysis Analytics
Tariff Modeling
Identify Uncover
Develop Develop a
profitable up-sell and
personalized robust customer
customers cross-sell
offerings retention strategy
opportunities
Source: Cognizant Research Center
Figure 4
Customer Lifetime Value profitable customers and retaining them is impor-
Customer lifetime value (CLV) is the profit or loss tant to sustain and also gain referrals.
generated by a customer during a business rela-
tionship. It reveals how much a new customer Customer profitability analytics allow CSPs to
is worth based on specific inputs — for example, determine who their most profitable customers
how long the customer will stay with a company are by comparing the revenues generated from
(estimated value), a customer’s purchase behav- these customers with the cost of providing ser-
ior and profitability. Based on the CLV, the CSP’s vices to them during a particular period. This is
marketing departments can decide how much useful for understanding what causes the lack of
can be spent on acquiring a particular customer profitability of some customers, as well as how to
type. By calculating segment-wise CLV, CSPs convert them into more profitable ones by devel-
can understand which segments provide better oping new products or tweaking existing ones and
opportunities and direct their customer acqui- developing targeted marketing campaigns.
sition and retention strategies. For instance,
CSPs can focus on providing better services to Targeted Marketing Campaigns
a few high-value (post-paid) customers vs. many Competitors’ marketing campaigns are one of the
less-valuable customers, although the latter could prime factors for customer defection. It is, there-
be targeted later to buy more services. fore, important to consistently engage customers
with timely and relevant marketing campaigns.
Customer Profitability Analytics
Based on the Pareto Principle — which states The first step is to break down complex customer
that 80% of a company’s revenues is derived data. By analyzing customer calling patterns —
from 20% of customers — identifying the most including the operator services they use, their
cognizant reports 4
5. demographics and other personal information — in the case of new customers from whom the
the subscriber base can be divided into segments company is yet to gain return on investment.
based on behavioral patterns. Such segmentation
allows CSPs to understand issues unique to each Customers depart for a number of reasons, rang-
group and address the group as a whole. ing from poor customer service and inappropri-
ate billing, to lucrative offers from competitors.
Customer segmentation analysis allows CSPs New offerings such as mobile number portability
to design marketing campaigns tailor-made have eased defection to a competitor’s service,
to addressing the needs of each segment. For making it possible to complete the move in a mat-
example, high-value customers who account for ter of hours. It is, therefore, important for CSPs
a major chunk of revenues or consume higher to understand the current level of satisfaction of
profit services can be offered special tariffs and their customers and identify in advance custom-
other services that incent them to stay longer ers who are most likely to switch loyalties.
without compromising profitability. Further, pre-
dictive analytics can be used to understand which Churn management solutions, including social
customer segment is most likely to buy a product network analysis, allow organizations to iden-
and during which period, resulting in better ROI tify customers who are most likely to defect
for campaigns. based on their behavior. Further, by combin-
ing and analyzing the data of defected custom-
Campaign Analysis ers, CSPs can understand the factors that influ-
CSPs apply huge amounts of resources to market- enced these departures and take steps to prevent
ing their offerings across various channels. With further churn.
the growing cost and complexity of marketing, it
is important that CSPs understand how effective Social Network Analysis
their marketing campaigns are. Campaign analysis Customers share varying degrees of relationships
can be used to study the efficacy of a marketing with other members in a group. Social network
campaign in generating sales against the costs analytics help in identifying proximities and rela-
incurred, based on key success criteria for a cam- tionships between people, groups, organizations
paign through a channel. It is also useful for under- and related systems. It reveals the strength of the
standing the impact on other related products relationships, how information flows within the
and design efficient future campaigns. Real-time groups and who the influencers are in the group.
campaign analysis allows marketers to measure It is observed that influencers can cause mass
each and every aspect of a marketing campaign churn within four to seven days of their move-
and take immediate corrective actions, resulting ment to another service provider. By appeasing
in efficient utilization of budget and resources. group influencers via customized and innovative
tariffs, CSPs can prevent mass churn and attract
Cross-Selling and Up-Selling new customers, including those from competi-
Affinity analytics or market-basket analytics tors. Further, group influencers can be used to
enable CSPs to understand products that are quickly spread the news about a new service or
often bought together. It also provides the right a product.
combination of services for customers. By iden-
tifying customers who have not bought any of Social Media Analytics
these services, CSPs can offer bundled services Social networking sites have emerged as a major
(up-selling) or new services (cross-sell). This will channel for customer engagement. Two out of
not only lead to improved customer spending three people in the U.S. use social networking
(increased ARPUs), but it will also reduce cam- sites; about 63.7% (147.8 million) of Internet users
paign costs, as the right customers are targeted. are members of social networking sites, and that
number will increase to 67% by 2013, according
Churn Management to eMarketer.
Retaining customers is a challenge, but not as
much as acquiring new ones. According to stud- Tracking of social media (using tools such as
ies, acquiring new customers is five times more text analytics) allows CSPs to understand cur-
expensive than retaining existing customers. High rent customer sentiment and obtain a deeper
customer attrition affects profitability, especially understanding of their products and services.
cognizant reports 5
6. For example, by analyzing the chatter created on traffic routing, free capacity can be identified and
social media about a new advertising campaign used elsewhere, driving down Op-Ex.
or special offer, CSPs can know which features
the customers liked or disliked and, based on this Network Monitoring
information, make adjustments to their offering. In a hypercompetitive arena, providing uninter-
rupted connectivity is of paramount importance
to CSPs, and doing so requires providers to
Analytics for Efficient Network maintain network health, identify glitches that
Management impact network performance and proactively
Network operating costs account for 45% of address them.
CSPs’ total Op-Ex, which itself is about 50% of
revenues. But networks are utilized only at 30% Network monitoring tools track network behavior
of their capacity, as traditional solutions employed and help in identifying stress points before they
by CSPs provide limited insights into network uti- impact the network and connectivity. Further,
lization, hindering their ability to utilize capacity by conducting root cause analysis of past net-
optimally, according to Yankee Group. With the work breakdowns using correlation techniques,
proliferation of smart devices driving the demand future disturbances can be averted, resulting in
for more network bandwidth, CSPs need to not improved quality of service.
only utilize their current network capacities more
efficiently but also predict future capacity addi-
tions accurately. Network analytics help address Analytics for Revenue Assurance
capacity utilization issues and predict future Fraud Management
requirements accurately, allowing CSPs to provi- Global communications fraud — the practice
sion bandwidth more efficiently. of using telecommunications products or ser-
vices with no intention of paying, according to
Capacity Planning the Communications Fraud Control Association2
The emergence of new technologies (3G, LTE, — results in the loss of $72 billion to $80 billion
etc.) is driving the demand for more bandwidth, (about 4.5% of revenues) annually. Fraud analyt-
as CSPs have to support wireless, VoIP, video, ics tools use customer bill payment schedules,
high-speed data, a host of digital media applica- demographic details, calling patterns and other
tions and other services across numerous con- information to proactively identify customers
sumer and business devices. This requires CSPs who are likely to exhibit fraudulent behavior. For
to invest in building new capacities. However, they example, the tools generate alerts upon finding
must ensure not to overbuild, which would lead a sudden abnormality in the usage or payment
to increased Cap-Ex and Op-Ex, nor under-build, patterns of customers, which can be used to plan
which often leads to poor quality of service. CSPs remedial measures.
should first understand their current network uti-
lization and then plan capacity expansion. Customer Risk Management
Some customers may not be able to pay their bills,
Real-time network analytics allow CSPs to under- resulting in bad debt. An increase in the number
stand current network usage — capacity utilized at of such customers puts pressure on revenues and
various locations at different time periods — and margins. CSPs need to be careful when imposing
identify regions where network usage is expected credit limits, in order to minimize the impact on
to grow vis-à-vis others due to a new marketing customers in good standing. Blanket policies can
campaign or a change in tariffs, among other and often do negatively impact satisfaction levels.
influences. This allows CSPs to plan in advance Analytics help identify risky accounts proactively
and provision the right amount of bandwidth at and design less expensive tariffs, which lessen the
the right place and at the right time, leading to burden on such customers without affecting CSP
cost savings. By combining this with dynamic profitability. This will not only reduce potential
revenue loss but also improve customer retention.
cognizant reports 6
7. Revenue Leakage
CSPs globally lose about $100 billion annually in revenue leakage, primarily due to inter-carrier settle-
ments, which is among the major revenue sources for some CSPs. According to Frost and Sullivan, inter-
carrier revenue leakage accounts for 3% to 5% of total costs for traditional voice products and 7% to
11% for broadband services.
Moreover, the increase in traffic, combined with the huge variety of services that CSPs provide, has
resulted in increased billing complexity. This, in turn, puts pressure on legacy systems when analyz-
ing huge volumes of call detail records (CDRs), hindering the accurate capture of relevant data that is
required to charge customers for their network usage.
Advanced analytics can be used to filter billions of CDRs quickly and efficiently, identify and plug
revenue leakage sources across the revenue chain and enable more accurate billing and inter-carrier
settlements.
Lastly, the benefits of customer and network analytics cannot be fully realized if internal processes
are not efficient. Often, there is a huge time lag between data capture, analysis and action, which costs
CSPs dearly. By measuring operational performance against key business drivers, process optimization
analytics help CSPs identify areas where capacity is unused or underutilized, or where changes in staff-
ing and budget can yield better results, etc. For instance, the sales department can check whether the
sales force assigned to a particular area is sufficient to handle the volumes and make changes to the
team if necessary.
Roadblocks ment that can effectively deal with data overflow.
CSPs need to analyze huge volumes of data that Efficient data management, combined with pow-
is generated continuously, such as CDR, SNMP erful predictive and real-time analytics that con-
(Simple Network Management Protocol) and sider customer, network and other critical aspects
IPDR (Internet Protocol Detail Record) data. With of the business, can result in just-in-time under-
millions of subscribers utilizing a range of voice, standing of operational issues, effective forecast-
video and data services, there has been an expo- ing and more meaningful and timely decisions.
nential growth in such data. However, very little
has been done on the database management
front to handle such an explosion, putting pres- Analytics for Competitive Advantage
sure on existing systems. Through the mid-2000s, analytics was synony-
mous with reporting. This has changed as data
Data at many CSPs typically resides in disparate proliferated and competition grew, which is forc-
databases running on numerous independent ing CSPs to adopt data-driven decision-making.
legacy systems, which often results in data incon- Further, the rapid development of IT and the avail-
sistency. In this situation, even advanced analyt- ability of industry-specific analytics are enabling
ics cannot aid in identifying problem areas or the organizations to embrace analytics to make tacti-
root cause of a problem. It is, therefore, important cal, operational and strategic decisions.
that data structures across the organization be
standardized and data issues resolved. Resolving CSPs acknowledge the benefits of analytics. In
data issues forms 70% to 80% of the effort in fact, the telecom industry leads many others in
implementing an analytics project, according to the adoption of analytics (see Figure 5). CSPs are
research firm IDC. in an enviable position when it comes to know-
ing their customers when compared with many
To leverage the benefits of analytics, CSPs need other industries. However, the information is
to focus on new approaches to data manage- not typically harnessed or used to inform more
cognizant reports 7
8. proactive decision-making the way it should be. an inability to cope with today’s dynamic market
The existing analytics infrastructure (legacy sys- conditions. In an ultra-competitive era, future
tems) at a number of CSPs has resulted in huge success will depend on how effectively CSPs
storage and related infrastructure costs, long lead leverage analytics to exploit data and build sus-
times for loading data and running queries and tainable competitive advantage.
Analytics: An Industry Scorecard
Q: Has your organization implemented a business intelligence/analytics solution?
Securities and investment 87%
Process manufacturing 84%
Insurance 83%
Life sciences 83%
Communication and media 82%
Transportation 81%
Government 79%
Banking 79%
Discrete manufacturing 79%
Utilities 78%
Oil and gas 77%
Healthcare 70%
Retail 70%
Professional services 68%
Consumer products 66%
Education 52%
0% 20% 40% 60% 80% 100%
(% of respondents)
Source: IDC’s Vertical Research Survey, 2010
Base: 2,771 respondents
Figure 5
It is often debated whether analytics can provide Embracing Analytics as a Service
sustainable competitive advantage when the Telecom data is complex. Analytics, with its
technology can be easily replicated by the com- wide application and the ability to meet the
petition. While competitors can duplicate the data increasingly complex decision-making needs of
collection process and the kind of data collected, telecoms, will play a crucial role in address-
the key lies in how much more quickly and effec- ing the issues. However, handling such huge
tively the organization can gain unique insights volumes of data poses a significant chal-
into its subscriber base vis-à-vis the competition. lenge for telecommunications providers and
Further, applying analytics to improve operational requires them to invest in people, process,
efficiencies is unique to each CSP. IT tools and infrastructure.
It is, therefore, important that analytics adop- A partner with the ability to handle complex ana-
tion be driven by top management and defined lytics tasks can help CSPs more quickly leverage
by specific goals, such as improving profitability the promise of advanced analytics. With virtual-
or reducing errors by a certain percentage. Top ization and cloud computing, opportunities now
management must focus on creating a strong exist for cost-cutting through an emerging ser-
organizational culture that emphasizes data- vice delivery model called business process as
driven decision-making. This can be achieved by a service (BPaaS).3 This approach makes avail-
closely aligning the business units with the teams able on-demand telecom analytics applications,
handling analytics, which creates a collaborative such as revenue assurance, fraud management
environment that can have a ripple effect across and customer experience management, which
the enterprise. can save critical Cap-Ex by eliminating the cost
cognizant reports 8
9. of acquiring expensive hardware, software and that CSPs stand to benefit greatly by associating
key talent. Moreover, enterprises pay only for with partners that have mature cloud computing
results achieved through outcomes-based and/or capabilities.
pay-for-use fee models.
To realize the full potential of analytics services,
In this context, analytics are delivered as a knowl- CSPs need to do the following:
edge process outsourcing (KPO) service, eliminat-
ing key technology barriers (continuous applica- Develop an enterprise-wide data architecture.
tion development, deployment and maintenance Identify key areas for deploying analytics.
issues), as well as talent-related obstacles that Design a comprehensive strategy for adoption
can often hinder performance. KPO allows CSPs and implementation of analytics, including
to deploy resources tailored to their needs that information technology.
can be increased or decreased as business Develop a fact-based decision-making culture
requirements dictate, providing Op-Ex flexibility. focusing on achieving specific goals.
Formulate customized strategies to capital-
When delivered as a service, analytics KPO ize on unique data, instead of copying the
presents options that are more extensive than competition.
traditional business process outsourcing (BPO), Continuously renovate and renew analytics
which focuses primarily on labor, cost arbi- implementation.
trage and continuous process improvement Enter into relationships with partners that are
to elevate business performance. As analyt- capable of providing mature analytics services
ics processes become standardized and can to aid in their attempts to gain competitive
be uniformly delivered via the cloud, we believe advantage.
Footnotes
1
Wireless ARPU decreased $0.46 in 2010. While data ARPU increased by $2.37, voice ARPU fell by $2.86 per user.
2
The Communications Fraud Control Association (CFCA), headquartered in Roseland, N.J., is a not-for-profit
global educational association that is working to combat communications fraud. The global fraud loss figures cited
are based on the initial results of a CFCA survey conducted in 2009. The figures represent a 34% increase over
2005 survey results.
3
BPaaS refers to the provision of business services encompassing underlying IT infrastructure, platform and skilled
manpower to run specific business processes in a virtual, globalized and distributed operating model.
References
Swati Sinha, “Improve Customer Retention with CRM Analytics,” The Decision Factor, June 28, 2011.
Kalyan Hariharan, “Revenue Assurance – Why This is a Critical Function in a Telecommunication
Industry,” India Telecom Online, May 12, 2011.
Eric Chan, “Five Ways to Win Back the Voice Market,” Wireless Week, April 26, 2011.
“Decision Analytics for Operational Excellence,” TM Forum, March 31, 2011.
Ari Banerjee, “BI Critical for Network Planning,” Telecomasia.net, March 21, 2011.
Michael S. Hopkins, “Interview: What’s IT’s Role in Analytics Adoption?” MIT Sloan Management Review,
Jan. 12, 2011.
“Big Data Integration Solutions: Turbocharge Analytics with Data Virtualization,” Composite
Software, 2011.
cognizant reports 9