customer experience index (cei) solution for net promoter
TRANSCRIPT
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 1 of 22
Customer Experience Index (CEI) solution
for Net Promoter Score prediction
July 2018
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 2 of 22
This material, including documentation and any related computer programs, is protected by
copyright controlled by Nokia. All rights are reserved. Copying, including reproducing, storing,
adapting or translating, any or all of this material requires the prior written consent of Nokia. This
material also contains confidential information, which may not be disclosed to others without the
prior written consent of Nokia.
© Nokia 2018. All rights reserved.
About Nokia
Nokia is a global leader in the technologies that connect people and things. Powered by the innovation of Bell Labs
and Nokia Technologies, the company is at the forefront of creating and licensing the technologies that are
increasingly at the heart of our connected lives.
With state-of-the-art software, hardware and services for any type of network, Nokia is uniquely positioned to help
communication service providers, governments, and large enterprises deliver on the promise of 5G, the Cloud and
the Internet of Things.
http://www.nokia.com || https://networks.nokia.com/software
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 3 of 22
CONTENTS
1 INTRODUCTION 4
2 CEI INSIGHT DESCRIPTION 6
BUSINESS PERSPECTIVE 8
USER INTERFACE & REPORTING 10
3 DATA MODELS 14
CUSTOMER EXPERIENCE INDEX DATA MODEL 14
DATA COLLECTION 15
4 CONCLUSION 17
APPENDIX A: NOKIA’S CUSTOMER INSIGHTS SOLUTION OVERVIEW 18
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 4 of 22
1 Introduction
Summary: A sophisticated set of Machine Learning algorithms use the network and non-
network data to predict the Net Promoter Score (NPS) for the entire subscriber base, and
on a per end user basis. The model is calibrated using periodic outside-in survey results to
provide an accurate proxy for NPS on an ongoing basis.
The key benefits are being able to
• Segment end users across clearly identified categories that reflect the level of
satisfaction of their experience with the service provider
• Drill-down (which end users, which locations, which devices, etc.) and serve multiple
stakeholders of the service provider organization, such as Operations with reasons/root
cause analysis for detraction enabling rapid and proactive corrective actions to improve
NPS, as well as Marketing and Strategy teams with insights on how to maximize
promoters end user usage and loyalty and generate new revenues opportunities
The traditional method of calculating Net Promoter Score (NPS) through customer surveys has
limitations in terms of the number of customers who can be surveyed, the cost of conducting the
survey, the time taken to get the survey results and the availability of detailed information on the
reasons for detraction. While the wide adoption of NPS across industries is an indicator of the
relevance of this measure, the problems listed lead to ask the following questions:
▪ Can NPS be calculated in near real-time instead of waiting for the results from a weekly
survey?
▪ Is it possible to cost-effectively calculate NPS for all customers instead of a sample of a few
thousand customers at a time?
▪ Can the reasons for detraction be proactively identified without having to ask the
customer?
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 5 of 22
▪ When corrective action is taken could the improvement in NPS also be measured
immediately?
▪ How to best capitalize on NPS scores that are satisfactory?
Communication Service Providers have the advantage of collecting a large number of KPIs at
different touch points such as each voice call, data session, customer service interaction and bill
sent. If this information is aggregated on single subscriber basis using a sophisticated algorithm it
can give a reasonably comprehensive view of customer experience at the individual subscriber
level. The algorithm would generate on a daily basis a single number on a 0-100 scale representing
how satisfied each customer is likely to be based on delivered service levels at various touch points.
Taking this one step further the algorithm can be calibrated so this single number reflects the
promoter score that a subscriber is likely to give i.e. predict how each customer would rate their
service provider on the 0 to 10 NPS scale using accurate historical information on the experience
for all individual subscriber service events with the operator over an extended period of time.
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 6 of 22
2 CEI Insight Description Nokia has designed and deployed a Customer Insights solution that predicts the promoter score a
customer is likely to give in response to the NPS survey based on that individual’s delivered service,
by calculating a near real-time overall NPS score for the entire subscriber base.
The prediction, referred as Customer Experience Index (CEI), is based on data that is captured non-
intrusively from the various network data sources collected via probes (transactional data / call
detail records), Billing and Care insights from IT systems (number of interactions with CC over
certain period, etc), CRM information (end user profile, tariff plan, etc). Other data sources such as
sentiment analysis, social media analysis, etc. can be added flexibly into the model. Network data
collected relates to both connectivity/accessibility (Control Plane), Usage and Quality of the service
(User Plane) as shown in Figure 1 below.
Methods have been developed to validate the quality of data arriving from the data sources,
process the outliers and impute the missing values to make it robust to imperfections of the
incoming data. In addition, elaborate methods and metrics to validate the performance of the
machine learning models result in consistent prediction performance. Building and deployment of
the machine learning model is automated (AutoML) to a large extent resulting in reduced time to
deployment as indicated in Figure 1 below.
Figure 1: Most complete Customer Experience Index (CEI) in the industry powered by Auto ML
The accuracy of the calculated NPS has been verified by comparing system calculated proxy-NPS to
actual NPS survey results for the same time-period. An actual comparison between the Net
Promoter Score given by the customer in a survey and the proxy-Promoter Score calculated using
Machine Learning is shown in Figure 2.
C u sto m e r S u r vey
�
CRM & location dataAge, GenderSubscription PlanLocation, DeviceLife time value, … �Network DataVoice, VoLTESMSData ÂOther SourcesCall center trouble ticketsBilling & Charging tickets
�
Data Auto ML
Model definition
Data Ingestion & pre-processing
Model building & training
Model evaluation & best model selection
Model deployment
C E I
Customer experience index is calculated
for all subscribers
Promoters
Passive
Detractors
One score is computed per subscriber
reflecting actual experience correlated with NPS surveys
9 0 8 0 6 7 5 4 2 0 3 4
Further aggregated by time, location,
device, segment, etc. to allow detailed analytics
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 7 of 22
Figure 2: Promoter Score versus Customer Experience Index
Please note that Nokia’s experience in creating CEI score has been a long-standing experience
acquired over the years. The model has been constantly evolving to increase accuracy first by
ingesting other type of data than network data, and second by using more sophisticated
technology until machine learning applied today. The combination of knowledge, experience and
constant innovation is what differentiates Nokia in our ability to produce an accurate reflection of
the customer experience. Our Machine Learning model has been deployed with a high level of
accuracy with several Tier 1 Service Providers across the globe.
The CEI framework is productized and adaptable such that service providers be able to benefit
from it right away while being able as well to consider and test different types of machine learning
models.
Figure 3: Customer Experience Index evolution
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 8 of 22
Business Perspective
Data collected from the network, customer contact points as well as many other data sources are
analysed and translated into concrete actions to improve the perceived experience, the customer
life time value, as well as identify new revenue opportunities.
Nokia is committed to build an outcome based approach when producing our CEI framework.
Types of actions that can be generated are depicted in Figure 4.
Figure 4: The Power of the Customer Experience Index
In the following customer reference example, the calculated CEI score acting as NPS proxy has been
used as the starting point for NPS improvement use cases. A drill-down on detractors could
identify dropped calls as the reason for detraction and specific actions were taken at these
locations on improving call continuity.
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 9 of 22
Figure 5: CEI Contributing factors drilled down to and corrective actions recommended
In this case voice quality was chosen as the drill-down criteria. Sites with more than 30% of
subscribers having Mean Opinion Score (MOS) less than 3 were identified. The Net Promoter Score
of these sites was 19.7% and the voice CEI was 82. The identified sites and the corresponding
scores are shown in Figure 4 below.
Figure 6: Detractors due to voice quality problems. Identifying location with a large proportion of
subscribers having low MOS
Further investigation was made at these sites by the network operations team and actions such as
frequency change, mechanical tilt change, TRX addition, neighbour cell optimization, jammer
closure was taken. The MOS improved and as a result voice CEI and overall CEI increased. As a result
of improved KPI’s for individual subscribers and therefore a better promoter scores, the NPS for
these sites improves significantly.
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 10 of 22
Figure 7: Detractors due to voice quality problems. Actions: frequency and tilt change, TRX
addition, neighbor optimization, jammer closure were taken
User Interface & Reporting
User interface reporting presents data to the end user in meaningful report format which can be
used for quick decision making. The result set from data model is shown as graphs in reporting
module. Queries are optimized to improve reporting performance.
Figure 8.1: Predicted results from the ML model applied – per subscriber
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 11 of 22
Figure 8.2: Predicted results from the ML model applied – per subscriber
The Customer Experience Index (CEI) application provide for various reports such as (not
exhaustive):
▪ Customer
Experience Index by
region & city
▪ Customer
Experience Index with
time (trends)
▪ Comparison
▪ Top X details
Region and City Analysis
A location analysis shows the measure of Experience Index by Region and Cities. By selecting a
region in Region analysis, the analysis moves to analysis by Cities.
Numbers inside a bubble represents the Customer Experience Index for a particular location and
the size of the bubble represents the total number of subscribers. Color code is used to reflect the
quality of the experience per location, e.g. in this case a poor experience is affecting Perth area.
Refer to the figure below.
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 12 of 22
Figure 9: Location Analysis - Customer Experience Index by location
Comparisons
This report enables user to compare multiple dimensions e.g. models of devices, location, and
subscriber segments etc.
Figure 10: Comparison Report
Top X details
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 13 of 22
This report shows dynamically the top x subscriber having a high index for selected criteria in the
global filter. Similarly, analysis can be done for the top ‘x’ - subscriber segment, devices and
locations.
Figure 10: Top X Details Report
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 14 of 22
3 Data Models
Customer Experience Index Data Model
An overview of Nokia key KPIs used by Nokia’s common data model and Key KPIs for CEI calculation
is shown in the following figure. The picture depicts a Mobile network. Similar model applies and is
available as well as commercially deployed for Fixed network.
Figure 3: Customer Experience Index Data Model - Key KPIs (mobile domain)
The KPIs in the above model are supplemented by the following data in the data model:
▪ Radio conditions (signal strength, signal to interference ratio) in idle and connected mode
▪ Outbound roaming status (% of time spent as outbound roamer)
▪ Usage information (E.g.: number of calls, number of SMS, data volume, time of use)
▪ Mobility information (# of cells visited)
▪ Customer Static data (*)
- CRM (Corporation, Department, Segment, Gender, Age, Contract type, Plan)
- Device (OS, device model, vendor)
- Location topology (cell / address)
- Technology
- Customer life time value
- APN
CEI (Customer Experience Index)
NEI (Network Experience Index)
Voice
SMS Data
CSI
(Customer
Service Index)Voice CS VoLTE
Call Drop MO (%) - CoreCall Drop MO (%) - Subscriber
Call Drop MO (%) - Radio
Call Drop MT (%) - CoreCall Drop MT (%) - Subscriber
Call Drop MT (%) – RadioAverage Call Setup Time( s)
Call Setup Success Rate (%)
Average MOS (#)Average Echo Path Delay (ms)
Average Jitter (ms)Average Noise Level (dB)
Mute Call Rate (%)
VoLTE Registration Failure (%)VoLTE Call Setup Failure (%)
VoLTE Average Call Setup Dur (s)
VoLTE Default Bearer Act Fail (%)VoLTE Dedicated Bearer Act Fail (%)
VoLTE Call Drop MO (%)VoLTE Call Drop MT (%)
VoLTE Average MOS Tx (#)
VoLTE Average MOS Rx (#)VoLTE Average MOS (#)
VoLTE Min MOS Tx (#)VoLTE Min MOS Rx (#)
VoLTE Min MOS (#)
VoLTE Redial Call Rate (%)VoLTE Long Call Setup Rate (%)
VoLTE Mute Call Rate (%)
Receive Failure Network (%)Receive Failure Subscriber (%)
Send Failure Network (%)
Send Failure Subscriber (%)
Attach Failure Rate (%)Activation Failure Rate (%)
Default Bearer Activation Fail (%)
Handover Failure Rate (%)NI Detach Rate (%)
NI Deactivation Rate (%)Dedicated Bearer Activation Fail (%)
Average PDP Setup Time (s)
Average Default Bearer Setup Time (s)Average Dedicated Bearer Setup Time (s)
Throughput Browsing (Mbps)Throughput Streaming (Mbps)
Application Access Time Browsing (s)
Application Access Time Streaming (s)Service Access Failure Rate (%)
Average Repetitive Calls (#)Escalation (%)
Average Ticket Age (days)
First Level Resolution (%)Mean Customer Wait Time (s)
Completed Request (%)Abandoned Calls (%)
Average Resolution Time
(days)
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 15 of 22
(*) CEI also utilises important data for customer related information that is imported into the data
model, usually from Customer Relationship Management (CRM) systems. Its nature is “static”
compared to the network traffic (counter & KPI) data processed through the southbound network
interfaces. The customer related information is assumed to remain mostly unchanged, and no
historical records are kept for this data.
Data Collection
Nokia plan the data sets and build the model, after exploring what KPIs and data can be provided.
In partnership with the service provider, and based on the KPIs and/or other data from the list
provided above, the solution can use existing Data Lakes. In such a model where data from existing
data lakes are used, Nokia expects such data to be made available via a Hive connector on the Data
Lake. Data not available from existing lakes would be collected from the network or other sources
as mentioned in previous sections.
Hadoop Storage nodes will provide the storage for customer KPIs and customer event data. The
customer KPI data and event storage enable the customer analytics functionality; default reports
and content for background processing reporting.
The functions performed by data collection and analysis, in conjunction with the Data Lake are:
▪ Collection of the subscriber event data from the network
▪ Storage of the subscriber event data
▪ Forming of the subscriber individual quality and use counters
▪ Default content for analysis
▪ Dynamically updatable data analysis
The customer definition in the customer repository requires data from CRM systems to be
collected. This can be from a mixture of different systems deployed and different interfaces to
these systems.
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 16 of 22
Network Data Sources
The collection of Network Data sources that are included as input to calculate the network part of
the CEI are the following:
Service Connectivity/Accessibility Usage Quality
Voice 2G, 3G IuCS, A IuCS, A Nb/Nc
SMS 2G, 3G IuCS, A IuCS, A NA
VoLTE S1-MME, S11 Gm (logical interface) Mb
Data 2G, 3G Gb, IuPS Gn UP Gn User plane, DPI
Data 4G S1-MME S1U S1U, DPI
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 17 of 22
4 Conclusion
Nokia’s approach provides a powerful method to identify actions that will improve NPS and witness
results in near real-time after actions are taken. It addresses the traditional problem of NPS
surveys which may tell you that experience is deteriorating but cannot explain why. The NPS
surveys also identify only a small sample of detractors since the survey methodology has an
inherent limitation on the number of subscribers who can be surveyed and the frequency of
surveys. Nokia’s CEI NPS proxy on the other hand is available in near real-time for the entire
subscriber base and provides a detailed drill-down to the likely cause for a customer to be a
detractor. Nokia has achieved a level of accuracy that makes this NPS proxy usable for NPS
improvement initiatives from real customer deployments and the Machine Learning algorithms and
calibration approach are being continuously improved to achieve higher levels of accuracy.
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 18 of 22
Appendix A: Nokia’s Customer
Insights Solution Overview
▪ Reveal high value customers with low customer experience as input for retention
activities
▪ Analyse service usage and quality for dynamically defined customer segments
▪ Identify best candidates for targeted campaigns
The challenge
Evolving technology, more and mode devices connecting to the network, increasing amount of
generated data result in the following challenges:
▪ Information on customer experience is scattered across organisational silos
▪ No holistic view of customer experience or ability to measure impact of initiatives across the
organization
▪ Inadequate understanding of customer experience touchpoints and KPIs and how to
improve them over time
▪ Transitioning to analytics driven data organisation can be hampered by people, tools and
processes
▪ Working in reactive mode rather than using predictive analytics and automation
The solution
Nokia’s Cognitive Analytics for Customer Insight provides a platform to accelerate the customer
experience journey, from any starting point. It helps increase customer satisfaction by creating a
holistic view of fixed and mobile subscriber experience. It provides uniquely actionable insights
down to every customer or cell. It has a built-in analytics framework with machine learning
algorithms for tuning insights.
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 19 of 22
Customer Insight’s technology utilizes Artificial Intelligence, to provide:
▪ Holistic view of customer experience and a platform to drive actionable use cases
▪ Key focus on improving the customer’s experience in mobile & fixed networks, increasing
the levels of automation in processes and increasing revenue
▪ Analytics to understand and improve customer experience across different touch points
using all available data sources
▪ Secured data collection and remote access to the system
▪ Increasing number of applications and use cases that can be selected from the Nokia’s Use
Case library
Customer Insights Introduction
Insight into how users are experiencing their services and the capability to take action based on
this insight are more important than ever. To address this, we need key capabilities to gather real
time and non-real time data from multiple sources, generate holistic insight about the customer
experience, and use it to prioritize and trigger concrete actions to improve the customer
experience and their business results.
Nokia’s Customer Insight combines an innovative dashboard view of the customer-centric
performance indicators with application for specific departments. You are able to access and share
information across the entire organization from operations to marketing to customer care.
Because everyone uses the same product, they speak the same language, without the need for
specialist knowledge.
Nokia’s Customer Insight is not another data warehouse. It is ready to use, with industrialized
telecommunication knowledge available to collect the large amount of data from the network,
customer contact points and other data sources to enable a view of each customer. The
information retrieved is analysed and consequently translated into concrete actions to improve the
perceived experience, and the customer life time value. In addition, we can identify new revenue
opportunities. To this, it requires us to turn the retrieved data into value.
The initial step is to collect relevant customer data from different data sources. This can be for
example collecting traffic data on a corporate customer level. In the second step this data is
processed and analysed in order to convert the data into information and knowledge
Customer Insights Functionality Overview
Customer Insights provides a centralized control of rich, near real time data produced by multiple
data sources like the Nokia real-time network analytics, third party probes, billing systems, and
many more. The data collected from those multiple data sources is stored for long term, correlated
and is provided a way of translating enriched information into actionable insights. With this, we are
developing a way of allowing users accessing customer relevant data, and complementing it with
wider end to end view and greater historical perspective than the one provided by a real time
environment. We bring this intelligence for different user groups by having a set of applictaions to
address different needs in terms of information type and detail. Different departments can then
benefit from different functionalities and end user data by means of specific use cases in order to
reveal relevant insights and derive actions in order to blend and improve existing processes/
business practices.
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 20 of 22
Customer Insights architecture and functionalities are specifically designed to deal with the key
challenges mentioned, that go from the point where data is collected simultaneously from multiple
data sources to actionable insights tailored for different user groups. The data model is TMF SID
certified (Shared Information Data) while the dynamics of the market allows adopting new data
source quickly as modelling follows a bottom up design that facilitates to develop schemas and
data loading procedures for new integration. Data Integration services help in creating the business
logic for data loading, aggregation and data quality checks. Data Integration application offer many
off-the-shelf adapters including various database adapters (Hadoop HDFS, Oracle, MySQL, Sybase,
ODBC etc.,) and file adapters (ASCII flat file, XML file, CSV files etc.,) to load the data from external
sources. In addition, it also includes auditing services to ensure data quality, data reconciliation and
security.
Customer Insights is offered in small scale deployments to large scale deployments with many
nodes. A staging server to perform data quality checks like identification and deletion of duplicate
data, identification of missing data, identification and deletion of bad records and data source
identification (add source id to incoming records) etc., is offered for robust data loading. The data
after cleansing, quality checks is loaded into raw data base tables and stored for the defined period
of time. Data can be loaded either in real time or in batch for faster performance. The raw data can
be used later for data aggregation and also for reporting. Data reconciliation is available as part of
the platform and be able to report if there are issues identified with data availability.
Business Reporting services allow configuration of rich visual content packages. Customer Insights
also offers numerous off-the-shelf content applications like High Value Customer Insights. Reports
can be scheduled at periodic intervals and can be visualized via web access or distributed by email
in various formats like HTML, PDF, Word etc.,
Customer Insights uses Hadoop Big Data HDFS based database and it is tuned to meet the
performance, capacity and scalability needs and at the same time offers customized services for
further database fine tuning. The system can scale vertically and horizontally based on capacity
needs in runtime.
Customer Experience Index
The Customer Experience is the overall psychological impact of a customer's interactions with the
Communications Service Provider (CSP) and the services provided. This includes direct experience
from the quality of product and network services delivered.
One of the major findings from a number of surveys and studies carried out by Nokia Networks is
that a significant gap exists between perceived assumption of delivered customer experience and
the "real" experience of end customers. The perception gap is substantiated by the fact that, while
80% are convinced of delivering a superior customer experience, only 8% of customers believe
that they receive it.
At the same time, we collect a tremendous amount of data from the network, customer contact
points and other data sources, that would enable them to create an overall, yet detailed, view on
the customer experience. The information retrieved should be analyzed and consequently
translated into concrete actions to improve the perceived experience and retention of subscribers
and also identify new revenue opportunities.
Customer experience index (CEI) enables understanding “customer experience” from the
subscribers’ point of view.
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 21 of 22
Application
The Customer Experience Index application can be used by many including product management,
marketing and account management personnel. It provides:
▪ Customer segmentation
o Easy segmentation and micro-segmentation of subscribers
o Identification of the right candidates for targeted campaigns, increasing the hit-rate
and saving costs.
▪ Cross-functional customer experience index
o Summarizes the customer experience in one index that crosses the organization -
one business KPI.
▪ Value customer analysis
▪ For calculation of overall experience index, platform supports flexible weight setting for
premium customers. That means - we can give more importance to a certain set of
customer segment over others.
Customer experience is impacted by many services and processes, both technically and
organizationally. CEI calculates and collects customer specific measurements from these services
and processes. Example: “Trouble ticket resolution time, Dropped data session” etc. The
measurements are collected over the customer’s life cycle, from the order to the termination of
the contract.
Customer Experience Index Customer
Contract
Subscription
Quality
Usage
Invoicing
Customer Service
Timeframe
Different thresholds are applied on customer specific measurement and further normalized based
on their importance to the customer experience. Each service is weighted against each other and
the Customer Experience Index is calculated for all customers individually. The Customer
Experience Index is a uniform KPI reflecting the overall experience of each subscriber.
Following the trend and distribution of customers according to the Customer Experience Index
enables the CSP to understand the reasons for a perceived quality difference against the measured
quality. Adjusting the thresholds of the KPIs and weights for the service builds a more reliable
Customer Experience Index.
© 2018 Nokia CEI as NPS prediction - Solution Paper
Page 22 of 22
The Customer Experience Index uses only data that has an impact on the customer experience
itself, emphasizing the soft factors. For example, dropped calls are considered, but the number of
hand overs might be ignored.
Modeled customer experience can be displayed in the CEI distribution report where the number of
customers Vs the Customer Experience Index graph is plotted. The user access to the application is
provided by web clients. The solution is designed to support a multi user environment where the
system has multiple concurrent users performing actions.
Though data analysis is distributed, the system provides a single access point to the data so that
the user can gain uniform understanding of the points of interest over the full network and all
technologies. User access to the data is authenticated and can also be restricted at a user level.
The customer repository forms subscriptions based on the measurements’ relationship to the
customer in person and to the contract, age and gender of the customer based on the integrated
data sources e.g. CRM. This additional data is used to group and analyze customer level data by the
new aspects provided.
The services offered by the system functions are the following:
▪ User authentication
▪ User management
▪ Central point of access
▪ Raw & long-term storage
▪ Customer repository
▪ Integration templates
System wide functions are also scalable by distributed processing architecture. When the number
of users, data storage time, data volumes or number of customers managed by the system is
changed, the solution can scale accordingly. The customer repository therefore enables the flexible
analytics of the customers data based on points of interest without report and database changes.
The customer repository is populated with key data from existing systems such as CRM, invoicing
and trouble ticketing. Integration is done via Integration templates.