2. • Noah Powers
– Principal Solutions Architect, Customer Intelligence, SAS
• Patty Hager
– Analytics Manager, Content/Communication/Entertainment, SAS
• Suneel Grover
– Solutions Architect, Integrated Marketing Analytics & Visualization, SAS
– Adjunct Professor, Business Analytics & Data Visualization,
New York University (NYU)
5. Information Management
“There is no better place to start than data, since
it is the fuel needed to make insightful decisions
that can drive your business forward.”
Information Management
ERP CRM EDW Online Social Other
Data Sources
6. Information Management & Analytics
“Being able to derive insights from data is the
key to making smarter, fact-based decisions that
will translate into profitable revenue growth.”
Customer Social &
Predictive
Segmentation Profitability & Network
Analytics
Modeling
LTV Analytics
Data Data Data
Quality Information Management
Integration Model
Metadata
ERP CRM EDW Online Social Other
Data Sources
11. DECISIONS
DATA ANALYTICS INSIGHTS
INFORMATION MANAGEMENT
12. OUR
PERSPECTIVE
Big Data is RELATIVE not ABSOLUTE
Big Data
When volume, velocity and variety of data
exceeds an organization’s storage or
compute capacity for accurate and timely
decision-making
13. THRIVING IN THE BIG DATA ERA
VOLUME
VARIETY
DATA SIZE
VELOCITY
VALUE
THE
TODAY
FUTURE
14. Which Category Are You?
Strategic
Data Managers
Aspiring Data
Managers
Competitive Advantage
Data
Collectors • Mature capabilities in data
management
• Attribute data management
Data to C-suite
• Embrace importance of
Wasters data
• 53% outperformed peers
• First to identify measurement
• Allow data to inform
& data points that align with
strategic decisions
corporate strategic goals
• Invest in technology
enablement
• 60% put 50% of data to
• Drowning in data use
• Misaligned IT and • Lack resources to
• Underperform Business leverage data
financially • Lack resources to
• Misalign IT and leverage data
Business
• Underuse data
• Mid-levels drive data
strategy Degree of Intelligence
15.
16. Big Data Marketing Challenges (1)
Source: 2012 BRITE/NYAMA Marketing in Transition Study
17. Big Data Marketing Challenges (2)
Source: 2012 BRITE/NYAMA Marketing in Transition Study
20. What If We Had A Set Of Master Keys?
ANALYST
CUSTOMER
21. Where We Want To Get To…
CRM Data Integrated Marketing Enrichment Data
Data Table
(Customer ID , 12345)
(Name , John Smith)
(Gender , M)
(Age , 42)
(Life Stage , FL)
(HH Income , 75K-100K)
(Children Ind , 1)
(HH Education, College)
(HH Value Score, Above Avg)
(CC Propensity, 0.57)
(Visit Recency, 12)
(Session Count, 7)
(Session Avg. PV, 4)
(Engagement, High)
(Content Goal, 1)
(Sticky Goal, 1)
Online History Data (Session Affiliate, Org Search) Current Session Data
22. Integrated Marketing Data Table
Discovery and
Marketing Analytic Modeling
Reporting
Data Queries Acquisition Predictive Analysis
OLAP Cube Discovery CRM Segmentation Analysis
Real-Time Model
Data Visualization Churn / Attrition Execution
The Integrated Marketing Table (also known as
“Customer State Vector”) is an analytic
approach designed for rapid retrieval of
customer-level data from any dimension.
23. Why Do We Care?
Act
Orient
YOUR
Decide Decide COMPETITIVE
MARKET ADVANTAGE
OPPORTUNITY Orient
Act
Observe
24. Big Data - Why Do We Care?
Video (Time: 0:00-5:00)
http://youtu.be/CrSX97elHDA?hd=1
25. DECISIONS
DATA ANALYTICS INSIGHTS
INFORMATION MANAGEMENT
26. Predictive Analytics
“Encompasses a range of techniques for collecting,
analyzing, and interpreting data in order to reveal
patterns, anomalies, key variables, and relationships.”
Customer
Segmentation
Predictive Profitability &
Social Network
Modeling Analytics
LTV
Data Data Data
Metadata
Quality Integration Model
ERP CRM EDW Online Social Other
Data Sources
28. OUR
PERSPECTIVE THE ANALYTICS GAP
Most organizations:
Can‟t generate the information they need.
Can‟t generate information fast enough to act on it.
Continue to incur huge costs due to uninformed
decisions and misguided strategies.
The opportunities afforded by
analytics have never been greater!
29. The Predictive Analytics Lifecycle
BUSINESS BUSINESS
MANAGER IDENTIFY /
FORMULATE ANALYST
Domain Expert EVALUATE / PROBLEM Data Exploration
Makes Decisions MONITOR DATA Data Visualization
Evaluates Processes and ROI RESULTS PREPARATION Report Creation
DEPLOY
MODEL DATA
EXPLORATION
VALIDATE
MODEL TRANSFORM
IT SYSTEMS / & SELECT
MANAGEMENT BUILD DATA MINER /
Model Validation MODEL STATISTICIAN
Model Deployment Exploratory Analysis
Model Monitoring Descriptive Segmentation
Data Preparation Predictive Modeling
30. Lifecycle Challenge…
20%
80% = :*(
IDENTIFY /
FORMULATE
EVALUATE / PROBLEM
MONITOR DATA
RESULTS PREPARATION
DEPLOY
MODEL DATA
EXPLORATION
VALIDATE “Data is the number one challenge in the
MODEL TRANSFORM adoption or use of business analytics.”
& SELECT
BUILD Companies continue to struggle with data
MODEL accuracy, consistency, and even access.
Bloomberg BusinessWeek Survey 2011
36. Customer Case Study: Telco
Handset vs. Network Compatibility
• Which customers should be upgraded to 4G?
• Which handsets should be pushed in which region?
Dropped Calls Analysis
• Do dropped calls contribute to churn?
• Are there handsets that are more likely to drop calls?
Handset Penetration Analysis
• Which cities have the greatest handset penetration?
• Which handsets have the greatest ROI in each market?
iPhone Launch Analysis
• Which markets are being hit the hardest by your competition‟s iPhone
launch?
• Which cities are the responding the best to your iPhone campaign?
37. Customer Case Study: Telco
Inner circle
represents %
of calls each
switch type Total number
carried. of drops that
occurred over
each handset
Outer circle type
represents
% of drops
each switch
type % of Drops is
the drop rate
carried. for each
switch.
Total calls and minutes Handset %s represent
are displayed for each the distribution of
individual switch by handset over each
region switch
52. Decision Trees
• Decision trees are a form of multiple variable (or
multiple effect) analyses
• Allow marketers to explain, describe, or classify an
outcome
– Use Case
1. After analyzing Dec 2011 campaign results, we
use Decision Trees to calculate the classification
probability of a prospect responding to the
acquisition campaign
2. Score “look-a-like” prospects for Dec 2012
campaign
54. Data Driven Segmentation Rules
Segment #1
#2
Recency Score: High
Engagement Score: Medium
Engagement Score: High
Age: Young Adult (25-44)
Affiliate: Organic Search
Affiliate: Email
Response Probability: Medium
Response Probability: High
55. Benefits Of Decision Trees
• The multiple variable analysis capability enables one to
discover & describe outcomes in the context of multiple
influences
• The appeal of decision trees lies in their relative power,
ease of use, robustness with a variety of data
and levels of measurement, and interpretability
Bootstrap Forests CHAID / C5 / RP Boosted Trees
56. Clustering
• Marketing can use cluster analysis to partition
prospects/customers into segments – without the
bias of a historical consumer decision
• Understand the organic synergies between
different groups
– Use Case
1. Marketing is planning a new campaign, and
historical information is not available
2. Tag prospects with cluster results for our Dec
2012 campaign, and influence creative execution
57. Clustering
Finding groups of observations such that the observations in a
group will be similar (or related) to one another, and different
from (or unrelated) to the observations in other groups
58. Data Table
Step 2
Step 1
Approach: K-Means
Number of Clusters: 3
59. Cluster #1 Cluster #2 Cluster #3
Weight Management Guilty Pleasures Health Management
Diet Focused Taste Focused High Fiber
60. Benefits Of Clustering
• Segmentations arise from varied business needs &
demands
– Marketing vs. Sales vs. Advertising
• Integrating data streams allows greater capabilities
– When combined, Marketing gains an increased understanding
of customer behavior, demographics and psychographics
Expectation-
Centroid Hierarchical
Maximization
61. Customer Profitability & LTV
“Customer lifetime value (CLV) is a prediction of the
net profit attributed to the entire future relationship with
a customer.”
Customer Social &
Predictive
Segmentation
Modeling Profitability Network
Analytics
& LTV
Data Data Data
Metadata
Quality Integration Model
ERP CRM EDW Online Social Other
Data Sources
63. Value of Your Company = Value of Your Customers
The only value your company will ever create is the
value that comes from customers–the ones you
have now and the ones you have in the future.
To remain competitive, you must figure out how to keep
your customers longer, grow them into bigger
customers, make them more profitable and serve
them more efficiently.
By Don Peppers and Martha Rogers, Ph.D.,
Founding Partners, Peppers & Rogers Group
63
64. Perils Of Ignoring Customer Profitability
• 20% of the customers represent 80% turnover
• Some customers repeatedly contact the call-center
• Sales channels are incented by revenue
• Identification and retention of the profitable customers is a challenge
Situation • Marketing campaigns segment customers without considering profitability
• Profitable and loyal customers are not recognized/rewarded
• It is not the profitable customers who are retained
• It is not the most profitable products which are offered to the customers
• Sales and call-center staff spend their time on the unprofitable customers
Consequence • Sale of unprofitable products result in losses and wasted resources
• Low return on sales and marketing activities
64
65. Competitive Advantage & Profitable Growth
Focus resources on gaining and retaining the most profitable
customers with the most relevant offers at the opportune time.
Positive & Negative Profit: Predict & Execute Proactively:
• Many are profitable customers • Identify customers most at risk
• Other customers reduce profits • Identify customer influence factors
• The key is to understand which Customer • Execute proactive customer retention
customers fall into each category
Profitability
Revenue Customer
Growth Retention
Customer
Relevant conversations:
Centric
• The way the customer prefers
• At the time they prefer
65
66. Path to Optimized Profitable Marketing
Harness customer insights that result in smarter more personalized
marketing execution to improve customer profitability.
Define Execute
Define
Consolidate and Analytically Optimized
Customer Value
Organize derived Marketing
and Cost
Customer data Customer Based on
Metrics
Segmentations Essential Insight
66
67. Define Customer Value
Challenges
Expenses are allocated with broad strokes to Costs Revenue
customer segments
Lack of visibility into the true drivers of
profitability
Solution: An advanced profitability costing
and allocation engine
A full cost view of individual customer
profitability to uncover profit drivers and
detractors
Understanding the root causes of adverse
trends for margin, revenue, and cost for
individual customers and segments.
Profit Retention Potential
Predicting future profitability including various
scenarios for customers and segments
Lifetime
Understand role and influence of social network Value
67
68. Costs At The Customer Level
In order to determine customer profitability in a reliable and
repeatable way, a comprehensive source of cost data at
the lowest possible level of granularity is required:
The data should be available on product, service and customer
level, where appropriate.
Aggregated costs need solid decomposition algorithms, accepted by
business and financial analysts
Average costs might be misleading, as the same product sold to two
different customers may have differing cost profiles
Customer, product and service profitability are not universal and
transferable across the entire database
Other costs to serve should be calculated using a proven
methodology, like Activity-Based Costing
68
69. Define Analytically Derived Customer Segmentations
Create individual segmentations for each of the profit levels
Uncover profit drivers or profit detractors for each profit level
Segment Name Description
Top 20%
Most • Uncover Why they are Most Profitable
Profitable • High influencer/ leader? Usage? highest churn rate?
High • Uncover Why they are Profitable
Value • Is it High usage? How high is the churn rate?
Middle • Determine which customers have potential to move up in profit.
Middle 70%
• Learn why they have lower margins
Low
• What is the churn rate?
Bottom 10%
Negative • Determine why they are negative value?
69
70. Accumulated Profit Curve
A smaller percentage of your customer base is driving the
majority of the profit.
Migrate /
Spend Keep &
Shift to
to keep migrate
lower cost
May be some of
your largest
customers
Source: Gartner
70
71. Customer Profitability – The Life Cycle
Acquisition Development Retention Churn/ Win-
back
Net Margin
71
72. Customer Profitability – The Life Cycle
Acquisition Development Retention Churn/ Win-
back
Net Margin
Decisions points during acquisition:
• Looking at products and offers
• Comparing pricing
• Company can be scoring - credit worthiness
72
73. Customer Profitability – The Life Cycle
Acquisition Development Retention Churn/ Win-
back
Decisions points during relationship
development:
Net Margin
• Service & product usage
• Customer user experience
• Cross & up-sell
• Bad debt detection and collection
• Customer service
73
74. Customer Profitability – The Life Cycle
Acquisition Development Retention Churn/ Win-
back
Net Margin
Decisions points during retention:
• Targeted retention activities
• Complaint handling
• Renewal pricing, discounting & bundling
• Reactive retention
74
75. Customer Profitability – The Life Cycle
Acquisition Development Retention Churn/ Win-
back
Net Margin
Decisions points during churn/win-back:
• Win-back discount and bundle pricing
• Trigger campaigns for future reacquisition
75
76. Examples of Elements Affecting Customer
Lifetime Value (CLV)
- +
(1) – Start-up of customer case
(2) + fee income
(3) – Continuing “cost to serve”
(4) + Sale of additional products, “cross-selling”
(5) – Advice
Opportunities
(6) – Marketing Through
Customer‟s
“Lifetime”
(7) – Initiatives for retention of customer
(8) – Influence others to churn
= Customer lifetime value = CLV
76
77. How Is Competitive Advantage Created?
Retention of the profitable
customers
Profitability Realization of the
per customer customers’ potential
Profitability Pricing of
per product products/services
and service considering profitability
Development of new
Insight in profitable products
profitability
through the Profitability Restructuring of
entire per market organization according
business segment to the segment’s
model profitability
Make processes
more efficient
77
78. Broaden Use for Profitability Metrics
Once Profitability Metrics are calculated, the information
can be leveraged across departments.
Sales/Marketing
• Offer Strategies
Finance
• Improved information for business
• Promotion strategies
analysis
• Product portfolio management
• Interconnection rates
• Customer segment management
• Cost control
• New product intro
• Process improvement
• Channel effectiveness
• Proper capital investment
• Marketing direction
Operations
• Network optimization strategy
• High cost process that needs to be reengineered
• Utilization review
• Infrastructure decisions
• Optimize contact center strategies
• Prioritize service treatments
78
79. Case Study: Verizon
• Business Issue: Needed to analyze
and understand shared expenses and
overhead costs such as sales,
engineering, and product development
and meaningfully allocate those costs to
the products sold and the sales revenue
generated. Lacked right information and
ability to do this on a timely basis “The cost and profitability initiative
at MCI, and subsequently Verizon
• Results/Benefits Business, supported by SAS
Activity-Based Management,
• Created P&Ls used to hold business
provided key information in the
leaders accountable for financial results
transition of the business through
by sales-channel segment profitability.
acquisition and continues to
• Expanded model to calculate more provide value that only cost and
detailed profitability information on a profitability insight can deliver.”
monthly and annual basis in:
• Channel profitability, Customer segment
profitability, Product or service
profitability, Cost of business processes
and Cost of shared services (such as IT)
80. Social Network Analytics
“Social network analysis views social relationships in terms
of network theory, consisting of nodes (representing
individual actors within the network) and ties (which
represent relationships between individuals).”
Customer Social
Predictive
Segmentation
Modeling
Profitability & Network
LTV
Analytics
Data Data Data
Metadata
Quality Integration Model
ERP CRM EDW Online Social Other
Data Sources
82. What is Social Network Analysis (SNA)?
Overview
The practice of identifying and
measuring the relationship structure
that exists between individuals within a
social network..
This is most commonly used in the
telecommunications industry where
it is used to understand the links
formed through voice, text and picture
messaging. Individuals can be
differentiated by the number and nature
of their connections to others.
82
83. Business Value of SNA
Social Network Analysis provides both a deep and
broad understanding of customer behavior. When
combined with proven advanced analytics this enables
the development of many powerful business focused
solutions which help build strong and measurable
customer advocacy.
83
84. SNA Based Business Solutions
Below are examples of business solutions that rely on SNA:
Social Network Propensity Scores
- eg. improve churn prediction, average $, or customer advocacy.
Persistent Individual Identification
- Enables multi-SIM use, prepaid SIM recycling, and improved churn
reporting.
Customer, Household, and Life-Stage Segmentation.
Customer Value
- Understood in terms of relations and influence upon purchase behaviour
of others.
Acquisition Of High ARPU Prospects
- And competitor customers through referral and highly targeted viral
campaigns.
Agile Campaigns
- Insights and data provided which indicates when specific customer actions
occur (enables a shift from monthly routine of mass campaigns).
84
85. Better Customer Understanding
Most mobile providers perform customer segmentation,
usually based upon call usage behavior or profile.
Also predictive analytics to identify churn risk customers.
Social Network Analysis reveals relationships and
measures the influence customers have upon others.
Churn
Churn
2
85
86. Agile Customer Management
Social Network Analysis is used to develop event-based
campaigns and customer management strategies.
Churn is an example;
- contact friends immediately after a customer churns.
SNA enables a move from traditional monthly batch
analytics.
Churn High Risk
High Risk Churn
2
High Risk
High Risk High Risk
High Risk
86
87. Community Detection
In addition to better understanding of individual
customers SNA can be used to create or enhance
household segmentation by identifying communities.
The purpose of Community Detection is to identify the
strongest relationships within the customer base.
2
87
88. Communities Detection
The allocation of communities need not be mutually exclusive.
These can be hierarchical communities which may first represent
immediate family and then extended friendships.
Supporting hierarchical communities is essential when solving
conflicting business goals such household segmentation (which
requires close communities) or viral marketing (which requires
larger communities for optimum results).
2
88
89. Household Segmentation
Because Community Detection finds the natural social groupings
of all customers it is a powerful mechanism for Household
Segmentation.
Using analytics to combine information about social links with, for
example, customer age, gender or location it is possible to
accurately infer household type and customer life-stage.
Male & Female Postpaid (age 40 yrs)
Single Prepaid (age 19 yrs)
Mature Family Segment
2
Different Surnames
Matching Address
Age Group 25-30 yrs
Young Couple Segment
89
90. Know True Customer Value
Customer advocacy is critically important in today‟s
marketplace. SNA is used to track adoption and spread
of new services and identify key influencers.
Community detection is used to attribute $$$ value that
is not visible at an individual customer level. Households
that span competitor networks indicate share-of-wallet.
I‟m a high value
2 customer on a
competitor network
I just bought I influence my partner‟s
an Android I‟m a highest purchasing decisions…
It looks cool, value customer
now I might
buy an Android..
90
91. Not All Links Are Created Equal
Customer relationships can be distinguished and
analyzed by
Their strength (e.g. number of calls)
Their interval class (e.g. days between calls)
2
We chat everyday
We chat everyday
I‟m a high value
We discuss sports
customer on a
scores on the weekend
competitor network
91
92. Identification Of Roles
Customers are categorized by links and position within the entire
social network (in some cases roles are relative to the community).
Leaders: Highest number of links and centrality measures.
Followers: Similar to Leaders, to a lesser extent. Usually directly
connected to a Leader.
Marginals: Similar to Followers, but not often connected to a Leader.
Outliers: Few links and often low centrality measures.
Bridges: Connect Communities and isolated individuals
2
92
93. Improve Retention of “Leaders”
Capability Marketing Action Benefit
Identify highly Target retention More efficient targeting of
connected strategies to marketing spend.
“Leaders” within “Leaders”. Reduced attrition / improved
customer base. retention.
Communications rapidly spread
throughout the customer base.
2
93
94. Improve Retention of “Followers”
Capability Marketing Action Benefit
Identify Implement highly Minimise viral churn.
“Followers”. reactive event-driven Efficient timing & targeting
Know when a retention strategies for of marketing $‟s.
“Leader” churns. “Followers” at-risk Reduced attrition / improved
retention.
Churn
Churn
2
High Risk
High Risk High Risk
High Risk
94
95. Use Viral Effect For Acquisition & Growth
Capability Marketing Action Benefit
Identify influential Target cross / up-sell Understand acquisition
"Early Adopters" & strategies to "Early value of campaigns and
“Bridges” to better Adopters". Leveraging indirect outbound
understand viral viral power of “Bridges” communications. Improve
adoption of new to competitor customer timing & relevance of new
products. bases. offers.
2
95
96. Persistent Customer Identification
By examining a customer‟s position within the social
network it is possible to infer persistent identification even
after churn, mobile service number, or address changes.
This approach can, for example, also be used to identify
Prepaid SIM recycling and multi-SIM use.
Accurate reporting of monthly „Churn & Adds‟ numbers are
critical to correct strategic decision making.
96
97. CLA In Banking / Financial Services
Data is different and does not capture a true social network
Pseudo-social network (PSN) where consumers are linked if they
transfer money to the same entities
Effectiveness of targeting network neighbors can be attributed to
similarity rather than to social influence
97
98. SNA in banking / financial services
An analytic framework that enables marketing analysts to enhance
customer insight by identifying and incorporating consumer purchasing
similarities and their strength in profiling and segmentation.
Use SNA derived variables to generate superior customer
understanding and improve campaign effectiveness:
Target those individuals that are strongly connected to key
individuals
Enhance campaign management process by introducing new
consumer variables and methodology (e.g. campaign selection
and response attribution).
Data can be exploited in a privacy-sensitive way, since it is not
necessary to know the identities of the connected consumers or the
institutions that connect them
98
100. Saturday Afternoon Preview
• Know how to gain efficiencies and boost ROI with
marketing automation.
• Recognize the keys to achieve real-time relevance in
both inbound and outbound channels.
• Understand how to plan, prioritize and execute to
maximize profits.
101. Orchestration & Interaction
Marketing
Decisions
Multi-Channel Campaign Management
Real-Time Decisions
Marketing Optimization
Case Studies
Information Management & Analytics
ERP CRM EDW Online Social Other
Data Sources
Data wasters. These companies underperform financially, and their business and IT functions are not aligned. They collect data, but severely underuse them. Found in every industry, these companies are most likely to put a mid-level manager in charge of their data strategy.* Data collectors. These companies are submerged in data. They recognise the importance of data, but lack the resources to do anything about them, beyond storing them. They suffer from poor IT/business alignment, with nearly one-quarter maintaining that IT does not understand the importance of data; another quarter says the same of the business side. Companies in the healthcare and professional services industries are likely to be found in this category* Aspiring data managers. These companies have fully embraced the importance of big data to the future of their company. They allow data to inform strategic decisions, and invest in them aggressively. But they still lag behind the leaders. Sixty-six percent of them put only about one-half of their data to good use. Companies in the communications and retail industries are most likely to be found in this category.* Strategic data managers. This is the most advanced group of big data managers, with the most mature capabilities. Fifty-three percent of these strategic data managers say they outperformed their peers in the last fiscal year, 44% say they are on even par and only 1% say they underperformed. They are most likely to be found among manufacturing, financial services or technology companies. Strategic data managers first identify specific measurements and data points that align closely with corporate strategic goals.
Tie back to IMM and foreshadow the predictive modeling, segmentation & SNA work
Customer and product profitability mutually depend on each other, i.e. two customers having exactly the same product and services package may differ heavily with their profitability, also the same product sold to two different customers may cost the operator differentlyTwo flat-access-fee product users may demonstrate different attitude to the available basket of services and generate highly different product costs; Two customers generating similar revenue may utilize different services bundles;Customer, product and service profitability are not universal and transferable across the entire database; more granular profitability calculator is necessary to propose right product to right customer micro-segments
Which customers are profitable?Those who keep generate positive Gross Profit*/ Net Profit ** month-after-monthThose, who’s cumulative Gross Profit/ Net Profit is already greater then Acquisition Costs (SAC) and Retention Costs (SRC)* Gross Profit (Return On Sales) = Gross Revenues from Services – Costs of Goods & Services Sold + Revenues from Interconnect and Interworking connections and messaging**Net Profit = Gross Profit – Costs to Serve (call center calls, claims, HLR & Network Costs, etc. directly related to customer activities and services used)
Which customers are profitable?Those who keep generate positive Gross Profit*/ Net Profit ** month-after-monthThose, who’s cumulative Gross Profit/ Net Profit is already greater then Acquisition Costs (SAC) and Retention Costs (SRC)* Gross Profit (Return On Sales) = Gross Revenues from Services – Costs of Goods & Services Sold + Revenues from Interconnect and Interworking connections and messaging**Net Profit = Gross Profit – Costs to Serve (call center calls, claims, HLR & Network Costs, etc. directly related to customer activities and services used)
Which customers are profitable?Those who keep generate positive Gross Profit*/ Net Profit ** month-after-monthThose, who’s cumulative Gross Profit/ Net Profit is already greater then Acquisition Costs (SAC) and Retention Costs (SRC)* Gross Profit (Return On Sales) = Gross Revenues from Services – Costs of Goods & Services Sold + Revenues from Interconnect and Interworking connections and messaging**Net Profit = Gross Profit – Costs to Serve (call center calls, claims, HLR & Network Costs, etc. directly related to customer activities and services used)
Which customers are profitable?Those who keep generate positive Gross Profit*/ Net Profit ** month-after-monthThose, who’s cumulative Gross Profit/ Net Profit is already greater then Acquisition Costs (SAC) and Retention Costs (SRC)* Gross Profit (Return On Sales) = Gross Revenues from Services – Costs of Goods & Services Sold + Revenues from Interconnect and Interworking connections and messaging**Net Profit = Gross Profit – Costs to Serve (call center calls, claims, HLR & Network Costs, etc. directly related to customer activities and services used)
Which customers are profitable?Those who keep generate positive Gross Profit*/ Net Profit ** month-after-monthThose, who’s cumulative Gross Profit/ Net Profit is already greater then Acquisition Costs (SAC) and Retention Costs (SRC)* Gross Profit (Return On Sales) = Gross Revenues from Services – Costs of Goods & Services Sold + Revenues from Interconnect and Interworking connections and messaging**Net Profit = Gross Profit – Costs to Serve (call center calls, claims, HLR & Network Costs, etc. directly related to customer activities and services used)
Social Network Propensity Score - using centrality and the network structure we can generate highly predictive propensity scores.These propensity scores can be for customer actions such as churn, acquisition, prepaid to postpaid migration etc.Persistent Individual Identification- using links and calling pattern mechanisms to assign similarity scores to individuals. We can track individuals over time, even through number or address changes.The assumption/requirement is that they continue to talk to the same people/telephone numbers over time.Customer, Household, and Life-Stage Segmentation.- Using community detection, in addition to demographics (age, gender etc), location, and address information where available we can allocate customers into family segments.Customer Value - An accurate metric of value is also a product of your influence upon others. If a low value customer is highly connected and a great customer advocate they may be responsible for significant acquisition of many customers (whom may be high value) and reduces churn, and hence marketing costs for retention.Acquisition Of High ARPU Prospects - Refer-a-friend and ‘member-get-member’ offers often yield better results when you are aware of the $ value of off-network friends and the potential market share (number of off-network friends) each customer can bring.Agile Campaigns- By having customer link analytics prepared, it can be matched daily to recent churners. The next day (or even hourly) the friends or any churned customers can be contacted to prevent viral churn.