An AYM Gael publication

Copyright  Jo Mills 2002

All rights reserved

Telecommunications Revenue Assurance

Implementing a Solution
More on Revenue Assurance

By Jo Mills

It is common for telecommunications service providers who do not have effective

revenue assurance tools to lose from 5% to 15% of total revenues. A one-time audit and
overhaul of processes will provide rapid payback by identifying major leaks, but as
systems and processes evolve following the audit, new leaks will appear.

This paper discusses the elements of a permanent reactive and proactive revenue
assurance solution and an approach to implement it. It touches on secondary benefits
from such a solution in providing invaluable input to product development, sales,
supplier management and network technology evolution.

Introduction
During the 1990’s, the focus was on growth. Carriers were throwing new technology
into the network, improvising systems and procedures and introducing untried products
in a race to keep up with competition and market demand. In the process, they
introduced many new points of revenue leakage.

In today’s more constrained industry there is greater emphasis on cost containment.

Carriers around the world are paying increasing attention to revenue assurance. Hard
numbers are difficult to come by and there are wide variations between carriers, but the
most conservative estimates are that on average a carrier will fail to collect at least 5% of
revenue due. In some cases, losses are much higher. The payback on systematic revenue
assurance programs is always high.

Even a one-time audit will provide many benefits, identifying recoverable revenue and
plugging gaps. An audit that focuses on just one area, such as retail billing, interconnect
settlements or service provisioning will have value in itself. But the greatest benefits will
come from an institutionalized revenue assurance program covering all aspects of
business operations from sales through to collection of payments. The program will
prevent or plug leakages, eliminate unwarranted payments and will incidentally provide
invaluable business planning information.
 Telecommunications Revenue Assurance
Implementing a Solution

Implementing a comprehensive revenue assurance program is a major undertaking and

will take time. The implementation approach, discussed later in this report, should focus
on first identifying and addressing the areas with the highest payback. Then the scope
can be progressively expanded to include secondary areas.

Revenue assurance should concentrate on identifying and correcting root causes of

revenue leakage. A carrier must not neglect recovery of unbilled revenue, but it is not
realistic to expect to more than 25% recovery of revenues lost in the past. The emphasis
must be on reducing ongoing revenue leaks. Even on this basis, cost justification should
be no problem.

It is not easy to quantify one of the largest benefits of a revenue assurance solution. The
analysis and instrumentation for revenue assurance provides valuable information on all
aspects of business operations, helping the carrier eliminate inefficiencies and focus
investment on maximizing profitability.

Causes of Revenue Leakage

There are many potential causes of revenue leakage. This section outlines some of the
more common causes to illustrate the extent of the problem.

Provisioning Weaknesses in the provisioning process may cause revenue loss

through delays in service activation, activation of features that the
customer did not order, delays in removing cancelled service,
failure to implement international call blocking and so on.

Audits between the features on the switch and the features

recorded in the billing system typically show a 5% - 7%
discrepancy, but higher levels are common.

Incomplete Records The records used to bill customers for non-recurring and
recurring charges, such as service activation charges, recurring
service feature charges and rental charges for data circuits, may
be incomplete. The customer receives the services free.

Trouble Tickets / Carriers often see trouble management and repair as a problem
Repair for Network Operations separate from customer care and billing.
But trouble management can cause revenue leaks. Examples:

• A customer calls to complain that a service feature is not

working. The repair person finds that it is not active, and
quickly corrects the “problem”. But the customer did not
order the feature and is not paying for it.

• A customer complains that their phone service is not working.

Page 2 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

The repair person finds the problem is with a customer-owned

handset. But the customer is not billed for the repair visit.

Customer Refunds Customers may find that it is easy to get refunds by calling up and
complaining about billing accuracy or quality of service. The
personnel handling complaints must have access to information
that lets them detect fraudulent patterns.

Call Detail Records Switch hardware or configuration problems may cause inaccurate
call details records to be generated, or may result in no call detail
record being generated for a chargeable call. Problems with
mediation software may result in similar problems. The 24 hours
= 0 problem is classic.

Transactions such as unmatched call details may be diverted into

a recycling file, and eventually be purged without ever having
been billed.

Late-Arriving Records may arrive after the statutory period beyond which they
Transactions become un-billable. In Europe, the operator cannot bill for
charges more than three months after they were incurred.

Interconnect Interconnect charges and credits represent a large part of any

Charges & Credits carrier’s cash flow in a competitive market. Weaknesses in the
interconnect charge calculation and reconciliation tools can result
in massive loss of revenue for call terminations and overpayment
of charges for call placement.

Rating & Errors in configuration of rating and discounting software, bugs

Discounting in the software and errors in manually applied discounts can
cause large loss of revenue. Customers will quickly point out
cases where incorrect rates are resulting in excess charges. They
are less likely to notify the operator when the reverse applies.

Special billing routines for large customers are a common source

of serious loss of revenue. When large amounts are involved,
small errors are magnified.

The list above gives cases where the operator directly loses billable revenue. Some other
“soft” causes of revenue loss include weaknesses in:

Credit Assessment Inaccurate credit assessment may result in excessive debt write-
off later in the cycle.

Treatment / Failure to properly manage accounts receivables or to apply

effective collection/treatment process may result in major losses

Page 3 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Collections from write-off of bad debt. These losses can be reduced through
careful monitoring of overdue accounts to detect a customers’
financial problems early on.

Fraud A major subject in itself. Customers or employees may defraud

the operator, obtaining service free of charge or obtaining
payments from the operator for goods and services that were
never in fact supplied.

Product Profitability Lack of data about the underlying costs of products, including
Analysis cost structures in different geographical areas and different
demographic segments, can lead to major loss of revenue.

Sales Targeting Service representatives need training and information to ensure

that they propose the optimum products to maximize revenues
based on customer location and demographic profiles.

Network Outages Excessive network outages may have a revenue impact far
exceeded the cost of renovation. Outages may be localized and
recurrent, indicating underlying physical or procedural problems.
Analysis tools are required to identify the patterns that indicate
these underlying problems.

These “soft” areas are outside the scope of this paper. However, information gathered for
revenue assurance purposes will provide insight into areas where credit verification, debt
collection and fraud detection procedures can be improved.

Instrumentation installed for revenue assurances purposes will also assist in calculating
the profitability of different territories, demographic markets and product lines, and will
provide insight into the business case for network upgrades. The revenue assurance team
should review the solution with the departments responsible for these functions to ensure
that all potential synergies are exploited.

Page 4 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Sales
Sales/ / Debt
Debt Account
Account
Troubles
Troubles
Ordering
Ordering Collection
Collection Adjustments
Adjustments

• Unprofitable Products • Lost Service Fees • Bad Debt • Invalid Refunds

• Poor credit risks

Interconnect
Interconnect
Provisioning
Provisioning Repair
Repair Billing
Billing Settlement
Settlement

• Unbilled Services • Free Services • Incorrect Charges • Excess Charges

and Features • Lost Revenues

Network
Network Usage
Usage
Elements
Elements Processing
Processing

• Lost Usage • Lost Usage

Tracking Issues
A naive view of the revenue assurance problem is that the basic requirement is to track
each revenue-generating transaction from origin through to receipt of the corresponding
payment. If every transaction can be tracked from start to end, there can be no leakage.
Unfortunately a number of factors make end-to-end transaction tracking impossible in
many cases. These include the work pool issue, splitting and filtering, identifier changes,
automatic transactions and calculations.

Work Pool Issue

There will be many points in the provisioning, repair and billing flows where transactions
are placed in a work pool, and leave that pool in a different sequence from the sequence
in which they entered it.

In the provisioning and repair work flows, orders or tickets may be held up while they
wait for available facilities, materiel or technicians with the required skills. Orders may
simply be held until close to their due date. In the billing flow, a batch of call detail
records generated by the switch will be distributed over invoices issued anywhere from
one day to six weeks later.

Page 5 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Day 1 Day 2 Day 3 Day 2 Day 3 Day 4

A D G A D E
B E H Work Pool
C G I
C F I F H …

In the diagram above, what happened to transaction B? Has it been permanently lost or
is it just being held back for some legitimate reason?

Splitting & Filtering

As transactions flow through the system, a single input to a process may generate several
outputs or none at all. Or several inputs may generate one output. For example, a data
circuit order may generate a number of work orders to connect facilities at the A and B
ends of the circuit and at the intermediate network locations. A voice installation order
may or may not generate a field visit depending on whether facilities are in place. Two
trouble tickets may be matched, creating a single repair order.

In these situations, there is no ready way of determining whether the absence of an output
transaction corresponding to the input transaction is legitimate or represents leakage. If
several outputs are generated from one input, there is no ready way to tell that all outputs
have been generated and none lost.

Identifier Changes
Many transactions do not preserve a common identifier as they flow through the process.
A customer service order number may be translated into a work order number. A rental
telephone set may have one identifier for materiel tracking purposes and an unrelated
identifier for billing purposes. It is often impossible to match these identifiers – and there
may be no good business reason to do so.

Automatic Transactions
A service order for a voice feature such as call waiting or calling line identification may
not result in an immediate billing charge, but instead will cause the billing system to
generate a charge for the feature each month. Short of replicating the ordering and
billing flows and logic, there is no ready way to confirm that all such charges are being
correctly generated.

Calculations
Usage charges are usually subject to standard rating based on origin, destination, duration
and time of day or day of week. They are then subject to discounting rules that depend

Page 6 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

on the customer class and billing plan. With corporate VPN arrangements, usage below a
certain threshold may not generate any individual charges at all. If a usage record fails to
generate an invoice item, this may or may not be legitimate.

Conclusions
The revenue assurance problem is not simply one of tracking transactions from start to
end. In many cases this is not practical. The next section will describe alternative ways
of identifying and preventing revenue leakage, including database audits, mirrored
calculations, statistical analysis and spot checks.

Solution Outline
Every carrier has a different mix of network technology, products and services,
procedures and computer systems, but there is great similarity between carriers in terms
of their internal functions and the revenue assurance issues they must address. Overall,
the solutions are also the same.

This section describes the typical components of a revenue assurance solution for a
telecommunications carrier.

Database
Database Statistical
Statistical Spot
Spot Mirrored
Mirrored
Audits
Audits Analysis
Analysis Checks
Checks Calculations
Calculations

Error
Error Dispute
Dispute
Analysis
Analysis Monitoring
Monitoring
Results

Performance
Performance
Follow-Up
Follow-Up Reporting
Reporting

Database Audits
Database audits compare the data managed by different operational systems and by the
network elements, and identify discrepancies. The operator should run database audits

Page 7 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

routinely, perhaps once per month or once per quarter. The diagram on the next page
shows a typical (simplified) audit process.

In this diagram, programs extract relevant data from two different sources. The two sets
of data are then matched, producing statistics and reports on discrepancies. Because of
processing delays, it is common for databases to be out of synch. However, the audits
will show up revenue assurance problems in two ways:

Database A Database B
e.g. Switch e.g. Billing

Data
Data Data
Data
Compare
Compare
Extract
Extract Extract
Extract

Stale
Statistics Discrepancies
Updates

1. Stale Updates If, for example, a customer service is active in the switch but does
not yet show in the billing system, the discrepancy may be simply
due to an update delay. But if the service has been active for three
months, it indicates a lost billing update.

1. Gap Creep If, over time, the rate at which records are being added to or
removed from one database does not match the rate at which
corresponding records are being added to or removed from a
matching database, it indicates a systemic problem.

Database audits have value independent of revenue assurance in providing a statistical

view of the operator’s business for planning purposes.

Database audits are essential, but may not be easy to build. One common problem is that
the keys used to identify records are not compatible. Complex algorithms may be needed
to match records for audit purposes. The databases may also have different scopes. One
database may hold all current and pending customers while another holds only current
and historical voice customers. Many problems will be encountered.

But the fact that it is difficult to build perfect database audit programs should not be seen
as an obstacle to doing anything at all. It is better to throw together a rough audit
program, then to refine it. Some results are better than none. Matching algorithms can

Page 8 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

be refined. Legitimate discrepancies can be filtered out. A trial and error approach will
in itself yield valuable insights into the business processes and potential causes of
revenue leakage.

When the causes of database audit discrepancies are not clear, the revenue assurance
group should lead a root cause analysis to identify the cause of the problem.

Root Cause Analysis: There are weighty books on this subject, but the essence of the
approach is contained in two maxims: Ask “why?” three times and Processes cause
problems, not people.

For example, if clerks are incorrectly entering billing information, the problem is not that
the clerks are making errors. Why are they making errors? Are they lacking training or
tools? Are they poorly motivated? Why is that? A root cause analysis is not complete
until it has identified the fundamental source of problems in the process and has
identified corrective actions. Fixing the symptoms will not cure the problem.

Mirrored Calculations
An excellent method of checking the billing system is to implement another system that
performs the main calculations in parallel.

Transaction
Records

Primary
Primary Mirror
Mirror
System(s)
System(s) System
System

Results
Results
Comparison
Comparison

The mirror system does not have to include the full functionality of the billing system’s
charge calculation routines. For example, it may be difficult for it to replicate “friends
and family” discount plans. But it should at minimum be able to replicate the base rating
calculation and produce results for use as a sanity check on the primary prepaid and
postpaid billing systems.

Preferably the mirror system takes the usage data from the closest point possible to the
switch, so it can audit the full rating chain.

Page 9 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

It is usually sufficient to compare results manually. When there is no ready explanation

for mirrored calculation discrepancies, the revenue assurance group should lead a root
cause analysis to identify the cause of the problem.

Statistical Analysis
It is usually not difficult to either modify systems to accumulate control totals as they
run, writing these statistics at the end of each run, or to write small programs that scan
through intermediate transaction files generating statistics. As a rule of thumb, count
anything that is easy to count, even if the value for revenue assurance is not immediately
apparent. Low-level statistics (e.g. number of calls for a given distance band from a
given switch) are more useful than high-level statistics. They make it easier to zero on in
the area where leakage is occurring.

Network Statistics: The network includes a number of useful sources of statistical

information. Switches produce operational measurements or peg counts that can be used
to cross-check call detail records and meter readings. Network monitoring equipment
also generates data on usage volumes. Network quality measurements should correlate to
levels of quality complaints from customers.

The statistics all go into a central database with a simple and flexible structure. A single
table holding statistic origin, name, date, cycle number and amount will be enough to
start with. Analysis programs can then run against the database.

System
SystemAA
“Hook”
“Hook” Counter File E1
Counter

System
SystemEE
System
SystemBB
“Hook”
“Hook” Counter
Counter File E2
Statistics
Database
System Counter File F1
SystemCC
“Hook” Counter
“Hook”

System
SystemFF

System
SystemDD
“Hook”
“Hook” Counter File F2
Counter
Analysis
Reports

Statistics can provide a basic check that the number of records produced by one process
matches the number of records fed into a downstream process. This will catch computer
operational errors: dropped or duplicate file processing, or failure to process a
replacement file after a re-run.

Page 10 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

More interesting uses for statistics include correlation, trend analysis and detection of
anomalies. For example, one would expect that the weekly number of calls and the ratio
of long distance calls to local calls in a given city or neighborhood would be relatively
stable, although there may be slow shifts over time. A sudden change in the volumes or
ratios may indicate a revenue assurance problem. A discrepancy between the ratios
reported by the switch and the ratios reported by the billing system would certainly be a
cause for concern.

When there is no obvious explanation of a statistical anomaly, the revenue assurance

group should lead a root cause analysis to identify the cause of the problem. Even where
the anomaly turns out to be legitimate, this type of analysis can provide invaluable
insights into the business. The analysis programs can be refined if needed to filter out
legitimate anomalies.

As soon as statistics are available from any system, they can be fed into the database and
initial analysis programs developed. As more information comes available, the database
will become increasingly valuable. But the key to success is to produce early results and
quickly build up interest and support for extensions.

Once again, count anything that is easy to count, even if the use is not immediately
obvious. The database should certainly not be limited to purely financial data. Better to
have too much data than not enough.

Error Analysis
When a target system rejects data from a mechanized source there is a process problem.
One system or the other is not functioning correctly, either because of logic errors,
because of synchronization failures or because of incorrect configuration. If the switch
reports a call but the billing system has no record of the originating number, there may
well be a revenue leakage problem.

The revenue assurance group should continuously monitor error and recycling files.
Simple mechanized tools may be required to help their analysis.

Some errors will be predictable, due to processing cycle delays, and will disappear when
the rejected records are recycled. When there is no ready explanation of the cause of
rejected transactions, the revenue assurance group should lead a root cause analysis to
identify the cause of the problem.

Spot-Checks
The carrier should institutionalize spot checks – sending trial transactions into the
system, tracking them and manually verifying all calculations.

Page 11 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

There are specialized tools that can generate calls for different types of test account,
recording call details independent of the switch. The testers can then compare the final
invoices to the data recorded on the trial calls to check whether usage records were in fact
generated and processed correctly, including correct rating and discounting.

Before the carrier launches a new product, live trials should be undertaken using test
accounts, and data related to the product checked from end to end. The checks are to
confirm that the product can be correctly provisioned and that bills for the product can be
produced accurately in a variety of scenarios. The revenue assurance group should be
well-placed to conduct these tests. They will have an overview of the total information
flow, they will have no stake in proving that the product is ready, and they will be
experienced in running spot checks. A product is not ready for launch until the delivery
and billing processes are free of potential revenue leaks.

The revenue assurance groups should also conduct spot checks on a routine basis for the
service ordering and provisioning flows for different types of product, and for the trouble
ticket and complaint processes. Trial accounts should be set up and allowed to fall into
arrears to check on the collection process. All aspects of business operations should be
subject to random sampling and verification that there are no points of revenue leakage.
This will not be a popular job, but it is essential to pro-actively identify and plug points
of leakage.

Dispute Monitoring
Customer refusal to pay bills and complaints about over-billing or incorrect service
provide important information about process integrity. Correcting the root causes of
these problems will often also correct revenue leakage problems. Customer complaints
and collection activity are often related. A customer may refuse to pay their bill because
they feel they have a legitimate and unsatisfied complaint. Or a customer may make
complaints to provide an excuse for not paying their bill.

In almost all communications companies, the collections, invoice adjustment and trouble
management (service complaints) processes are handled by different departments, with
little sharing of information. A simple shared database fed by the systems supporting
these departments may have quick payback in improving effectiveness.

Page 12 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Collections Adjustment Trouble

Trouble Complaints
Collections
Collections Billing
Billing
Clerk Clerk Management
Management Clerk

Shared TN: 123-4567

Database 2002/05/23 15:23 Collection Attempt …
2002/05/23 15:45 Complaint …
2002/05/23 15:57 Billing Adjustment …

In a sense, all complaint, adjustment and collection activity represents revenue leakage.
Customers are not satisfied with their service or are refusing to pay for it. It costs the
carrier money to correct the problem or collect the receivable. Better to avoid the
problem in the first place.

The collections, adjustment and complaints processes require careful monitoring,

correlation and root cause analysis.

Follow up
Database audits, mirrored calculations, statistical analysis, spot checks, error analysis and
dispute monitoring will all uncover anomalies. Root cause analysis of these anomalies
may identify points of revenue leakage. Follow-up activities have three goals:

1. To recover lost revenue where possible by adjusting invoices.

2. To prevent continued revenue leakage by correcting records, for example by updating

a customer’s billing records to show the services in use or disconnecting services that
are not being invoiced1.

3. To prevent future revenue leakage by correcting the underlying problem that made
the leakage possible in the first place.

The line departments will often be responsible for executing follow-up activities.
Follow-up may involve significant projects to implement software and/or procedural
changes. The revenue assurance group is responsible for initiating, monitoring and
reporting on the success of all follow-up activities, ranging from minor invoice
adjustments to major projects of several months’ duration.

Performance Reporting
There are two aspects to performance reporting:
1
The value of freeing up network resources used by unbilled services may in itself be more than sufficient
to cover the cost of implementing a complete revenue assurance solution.

Page 13 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Corporate The revenue assurance department will gather a wealth of statistical data
Planning on all aspects of the company’s operations. The revenue assurance group
should publish this information monthly for use in product planning,
sales campaigns, network planning and so on.

Revenue The revenue assurance department should also publish the results of their
Assurance work, outstanding follow-up actions and so on.

AYM Gael does not recommend exhaustive discussions over the types of published
metric. It is best to start by publishing the most readily available information and then
refine the output based on feedback from the recipients. There will be plenty of
feedback.

Page 14 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Implementation Approach
The diagram below shows the main steps in implementing a revenue assurance operation.

Project
Project
Initiation
Initiation

Situation
Situation Framework
Framework
Analysis
Analysis Implementation
Implementation

Procedure
Procedure Mirrored
Mirrored Statistics
Statistics
Audit
AuditTools
Tools
Design
Design Calculations
Calculations Collection
Collection

Analysis
Analysis
Institutionalization
Institutionalization Tools
Tools

A series pf projects will implement audit tools, mirrored calculations, statistics collection
and analysis tools, prioritized based on the situation analysis. The emphasis will be on
addressing the primary areas where there seems to be highest payback first, then adding
coverage of secondary and tertiary areas as funding permits.

Tertiary
Primary

Effort Tertiary Payback Primary Secondary

Secondary

Packaged Solutions
With the growth of awareness of the critical nature of revenue assurance to a carrier’s
financial health, many vendors have developed generic services or software to address
the problem. They include:

Page 15 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Consultants / The main integrators have all adapted their generic models to meet
Integrators specific telecom needs.

CC&B Vendors All the major Customer Care and Billing system vendors provide
some mix of revenue assurance products and services.

Mediation The vendors of usage mediation software have almost all extended
Vendors their products to include revenue assurance tools, usually focused on
usage analysis.

Specialists Many smaller companies provide a mix of consulting services and

tools.

All these vendors provide attractive sales material and promise large and rapid returns.
Although their claims may be exaggerated, they should not be dismissed out of hand.
Almost any revenue assurance project will provide good payback. However, AYM Gael
considers that before selecting any vendor the carrier should consider the following:

1. Most of the effort in a revenue assurance project will consist of building many small
software building blocks to gather and correlate information from the existing
operational systems, files and databases. The best people to do this work are the
people who maintain these systems at present.

2. Although there are useful analysis tools, they will require extensive configuration or
customization to match the carrier’s business environment, the available information
and the unique coding schemes used by the carrier. Simple SQL reports from the
base information may give most of the benefits that more elaborate tools provide.
The carrier should not rule out using analysis tools, but should treat claims for their
benefits with some skepticism.

3. Given the broad scope of a revenue assurance project, there will be benefit to using
neutral external consultants to provide overall coordination and direction. The
consultants must have solid telecom revenue assurance experience. But no matter
how knowledgeable the consultants are, employees who understand the unique local
operational environment must do much of the detailed analysis.

4. Ongoing revenue assurance will be the responsibility of an internal department, and

the nucleus of the staff for this department must be involved in the revenue assurance
project from day one.

Bottom line is that there are no miracle tools, and no external consultant can do the whole
job for the carrier. The carrier’s internal staff must do the bulk of the work, with
assistance and coordination as needed, particularly at the front end, from outside
consultants with deep understanding of the industry and the problem domain.

Page 16 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Project Organization
A revenue assurance operation touches on almost all aspects of the carrier’s business,
directly or indirectly affecting many different departments. Minor software changes will
be required to many operational computer systems, competing with other high-priority
enhancement requests. Implementation of the revenue assurance processes will depend
on obtaining buy-in and ongoing cooperation from many different groups.

The project must include full-time or part-time participants from the main departments
affected, so the organization will depend to some extent on the carrier’s existing
departmental structure. This report provides a baseline organization assuming that only
three main departments are involved:

Customer Service Handles sales, ordering, account inquiries and complaints

Network Operations Handles provisioning, repair and central office equipment

Finance Handles billing, collections, settlements and financial reporting

In practice, the division of responsibilities may be different. For example, there may be a
case for separate representatives for consumer and major accounts customer service, or
for voice and data network operations if different groups handle these functions.

The diagram below shows a simple baseline project structure:

Steering Group

Project
Project Customer
Customer Network
Network Finance
Finance
Sponsor
Sponsor Services
Services Operations
Operations

Working Group

Project
Project Customer
Customer Network
Network Finance
Finance
Director
Director Services
Services Operations
Operations

Service
ServiceRecords
Records Charges
Charges Disputes
Disputes
IT
ITTeam
Team Team
Team Team
Team Team
Team

Steering Group Chaired by the project sponsor, the steering group meets monthly to
review overall project status, to set priorities and to resolve policy
issues. It includes representatives from the major affected
departments.

Page 17 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Working Group Chaired by the project director, the working group meets weekly to
review detailed project status and to resolve issues escalated by the
implementation teams. It includes representatives from the major
affected departments.

Implementation Each implementation team is responsible for a specific aspect of the

Teams solution, led by a project manager who reports to the project director.
Team composition will include people permanently assigned to the
revenue assurance group, expert consultants and people on loan from
affected departments.

The IT team handles all computer-related aspects including hardware,

operational systems changes, infrastructure and reporting and
analysis tools.

The other three teams are organized by major process flow, rather
than by department. This is to ensure that full attention is paid to
interdepartmental boundary issues, often the most common source of
leakage.

These teams are responsible for mapping the process flows and
identifying database audits, mirrored calculations, statistics, error
analysis and spot-check requirements and for developing ongoing
revenue assurance procedures.

• The Service Records team is responsible for all processes that

maintain or change information about customers and their
services, including sales, ordering, provisioning, network and
billing system records.

• The Charges team is responsible for all processes that generate

and / or calculate charges, deliver invoices and handle payments,
including both retail billing and interconnect settlement from
switch to payment.

• The Disputes team is responsible for the complaints, adjustments

and collections, and for identifying discrepancies, potential
synergies and ongoing monitoring processes in these areas.

If an external consultant is involved, their expert staff would normally be placed in

“shadowing” roles in this organization structure, working in close partnership with the
corresponding project director and project managers. There is considerable value in
seeding the project organization with experienced subject experts, but AYM Gael would
strongly recommend that the bulk of the team members come from the carrier’s internal
organization.

Page 18 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Typical Timeline
Implementing a revenue assurance operation is an ongoing process. The first objective is
to establish basic processes and supporting software to monitor the most vulnerable
areas. Following that, an ongoing series of projects will extend coverage and adapt to
changes in the business.

The timeline for the initial implementation will depend on availability of project
resources and competing initiatives may affect it. The sample timeline for the initial
launch given below is therefore for illustration purposes only. The situation analysis
phase will develop a more realistic timeline.

Task Name Month 1 Month 2 Month 3 Month 4 Month 5 Month 6

Project Initiation
Situation Analysis
Procedure Design
Institutionalization
Framework Implementation
Audit Tools
Statistics Collection
Analysis Tools
Mirrored Calculations

In this illustrative timeline initial operations start about five months after project start.
The critical path is defined by the time taken to ramp up the project team, conduct the
initial situation analysis focusing on quick hit opportunities, design the initial set of
procedures and commence operations.

Project Initiation
The project initiation phase establishes the scope, framework and preliminary schedule
and resource plan for the project, brings the key players on board, prepares a preliminary
business case for approval for work to start and prepares a change management plan.

At the end of this phase the carrier will have staffed up for the situation analysis and
framework implementation phases, with a budget and detailed project plan for these
phases. If the carrier has approved a mirrored calculation solution, the project team for
implementing this solution and initial schedule will also be in place.

A preliminary schedule and budgetary estimates for implementing the primary and
secondary waves of the revenue assurance solution will also be provided at this stage.

It is critical to get key influencers in the major affected departments on board during this
phase. There may be emotional resistance to implementing a set of processes for
monitoring people’s work and reporting problems. As stated earlier, it is important to
emphasize that the purpose is not to audit individuals but to look for ways to improve the
process. However, some individuals are bound to feel threatened. The change
management plan should attempt to address these natural concerns.

Page 19 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Each steering committee meeting and each working committee meeting should include a
generous time allocation for discussing change management. With some carriers,
contract and labor union issues have become a serious impediment to plugging revenue
leaks. If the project is presented positively to the affected departments and employees
and their concerns are addressed the return will be far higher than if the project is treated
as a purely technical implementation.

Situation Analysis
In the situation analysis phase, typically lasting about six weeks, the objectives are to

• Develop an overall map of the business processes and related computer applications,
and the information flows. Annotate the map with risk areas where there is revenue
leakage potential (see section 0)

• Derive initial metrics with which to quantify the carrier’s operations and compare
them to industry norms. The team may need one-off software to derive metrics, and
may use sampling techniques to measure the quality of critical paths in the
information flow. There is no accurate information on industry norms, but a qualified
consultant will be able to provide representative figures.

• Based on these metrics and industry experience, identify the primary areas where the
highest return on investment is likely, and the secondary and tertiary areas. The
higher the revenue volumes and the higher the risk, the higher the score.

Categorization may be difficult. The most practical way to distinguish between

primary, secondary and tertiary priorities may be to rank the problem areas, roughly
estimate the effort needed to address each of the first twenty or so problems, then
arbitrarily assign them to primary, secondary or tertiary based on the available budget
for the first two implementation phases. Perhaps the top five will get into the primary
class, the next ten will get into the secondary class and the remainder will fall into the
tertiary class for later consideration.

• Develop a schedule and resource plan for procedure design, audit tools and statistics
collection and analysis tools to address the primary risk areas.

Procedure Design
In this phase, the teams define and design the revenue assurance procedures. Operational
procedures are required for database audits, collection of statistics and mirrored
calculations. Manual procedures must be defined for:

• Statistical analysis and reporting

• Spot checks: different procedures for different types of spot check

Page 20 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

• Error analysis

• Dispute monitoring

• Follow-up initiation and tracking

Representatives of the affected line departments must be involved in procedure

definition, must approve the procedures and must be committed to supporting them.

Framework Implementation
This phase, starting as soon as the executives have funded the project, establishes the
technical framework for the revenue assurance project. It includes staffing, hardware and
system software acquisition, database configuration, tools selection and installation,
technical standards definition and development of model subroutines and programs for
statistical extracts, load and reporting.

Statistics Collection
The team will initiate a series of projects to collect statistics to support revenue assurance
in the primary risk areas. Although the revenue assurance team will fund and coordinate
these projects, they will normally farm out the work to the IT groups responsible for each
of the source systems.

As stated earlier, the emphasis should be on gathering the low-hanging fruit. All
information is useful, but any information is better than no information. If it is possible
to acquire some statistics at relatively low cost and others will take longer, the team
should split up the projects so they can achieve at least some of the deliverables as early
as possible.

Analysis Tools
To be meaningful, statistics must be filtered, correlated and analyzed to detect trends and
anomalies.

Filtering ensures that when a task is re-run, statistics from the re-run replace the statistics
from the earlier run.

Correlation includes matching statistics from different processes that should correspond
exactly. For example, a record count out of a mediation run should exactly match the
record count from the billing run that processed the mediation output file. Correlation
also includes deriving ratios between different statistics, where anomalies may indicate
revenue leakages. The correlation processes may raise alarms.

Page 21 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Analysis typically looks at ongoing trends, looking for sudden shifts in daily statistics.
Anomalies like this should also raise alarms.

The statistics will depend on the carrier’s local procedures, and will only be meaningful
in terms of the carrier’s business model. Filtering, correlation and analysis will often be
best done be specialized software tailored to the local environment.

However, it is worth investigating mechanized tools in this area. Artificial Intelligence is

a term that has fallen out of favor, but some of the AI concepts remain valid. There are
tools that perform large numbers of random correlations, measure the stability of the
correlations and report discrepancies. Through a blind number-crunching approach they
may find hidden but highly suggestive relationships between apparently unrelated bits of
information. Perhaps a particular class of customer has exceptionally high long-distance
calling volumes and also has exceptionally high dispute volumes. Root cause analysis
may uncover new ways to save revenue.

Audit Tools
Database audit tools, comparing information from different sources, will require custom
development on a case-by-case basis. As with collection of statistics, it will often be best
to break down database audit projects into sub-projects. Crude comparisons will provide
useful information, and can be refined later. For example, it may be relatively easy to
match single-line accounts, but much harder to match multi-line accounts. If this is the
case, the first version of the audit program should focus on reporting discrepancies in
single line accounts, and a later refinement can handle the multi-line problem.

As with statistics, any information is better than no information.

Institutionalization
With procedures designed and the initial mechanized tools available, the permanent
revenue assurance employees are trained and operations commence. Different specialists
will handle different areas, such as usage error monitoring and follow-up, dispute /
collections monitoring and statistics analysis.

The carrier should implement the new procedures very carefully. Change management
will have been addressed throughout the project, but the strongest reaction will come
from the affected line department workers at the point when they start to see their results
are being checked and their mistakes analyzed, and to realize that some of the
transactions they handle may be bogus, test transactions designed to measure the process.

Again, it is essential to emphasize that the goal is to look for process improvements to
improve the financial health and competitive position of their company. The goal is not
to look for non-performing individuals or to eliminate job functions.

Page 22 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Conclusions
Introduction of a comprehensive revenue assurance solution is a large undertaking,
touching many aspects of the carrier’s operations. It is a large job, but may be phased in
to capture the main benefits as soon as possible. And the benefits may be very large
indeed, ranging from 5% to 15% of total revenue.

Each project is unique. There is no generic, one size fits all solution. The revenue
assurance solution for a carrier must be carefully tailored to their unique operational
configuration. External consultants can help by bringing experience from other projects
and providing objective opinions in controversial areas. But internal employees must do
the bulk of the work, and the bulk of the software cost will be for changes to existing
internal systems.

One of the most important factors in ensuring that the project is successful is early and
ongoing involvement from all affected departments, and careful change management. If
managed carefully, a revenue assurance project will not just increase revenue. It will
promote a broader and deeper awareness of overall business processes, and will provide
invaluable input to business planning.

Page 23 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

Appendix – Commonly Asked Questions

Q. If a one-shot audit gains many of the benefits, why go to the cost of building a
permanent solution?

A. Several reasons:

1. A lot of the work in building a permanent solution is the same as for a one-shot
audit. Once the analysis has been done, the database audit programs written and
the statistics collection software has been built, there is relatively low
incremental cost to re-running the software and processes again and again.

2. An audit between the switch and the billing system will detect unbilled services
and features, which can be corrected. But until the underlying problem has been
fixed, fresh audits will continue to find additional unbilled services and features.
It takes time to fix the underlying problems, and in the meantime systematic
audits are needed to detect and recover revenue leaks.

3. The world does not stay still. New products, new network technology and new
business approaches all introduce new potential revenue leaks. A permanent
solution will detect and stop these leaks before they cause serious loss.

Q. Why not focus on one area, such as billing, and then look at other areas later?

A. Revenue leakage often occurs at process boundaries. A holistic approach is the only
way to detect these leakages. Ordering and Provisioning may be working well,
Repair is working well. Billing is working well. Treatment is working well. But
millions of dollars are being lost in the cracks between these processes.

Q. What are the top areas where revenue is normally lost? Why not focus on them?

A. It is difficult to get hard numbers. Carriers are understandably reluctant to publish

quantitative information on revenue loss. A number of surveys have been conducted
by consulting firms, with results published on the internet, but results are distorted
by the way the questions were asked.

For example, a survey asking how many correspondents feel they are experiencing
revenue loss in each of eight areas will give an indication of where losses are most
commonly known to occur, but will not show how large the losses are, and will not
give any information on areas nine and ten.

The pie charts shown below are typical of survey results. At most, they indicate that
revenue losses appear to be spread fairly evenly over different areas of the business
and different categories of loss.

Page 24 of 25
 Telecommunications Revenue Assurance
Implementing a Solution

The numbers for a given carrier are unlikely to conform to the average. Each carrier
has unique issues. The most that can be confidently asserted is that there will be
large areas of revenue loss, and they are likely to be found anywhere.

Fraud
Unbilled features Bad Debt
Stale charges (> 3 months) Error File
Unmatched Toll Customer Records
Unbilled service fees Lost records
Adjustments
Incorrect Rates
PIC/Care
Fail to create records
Lost calls
Corrupted calls
Records too late

Customer acquisition

Ordering

Provisioning

Activation

Usage Collection

Rating
CDR Recycling

Billing

Interconnect Settlements

Collections

Adjustments

Q. If all the detailed work is done by in-house employees, why hire a consultant?

A. A consultant will be able to supply people who have worked in this area for other
carriers. They will be able to jump-start the process, giving informed and neutral
advice on areas to be investigated, the approach and the techniques: what works and
what does not work. Their main contribution will be at the front end of the project.
Once the framework has been defined and the project has gained momentum, the
carrier’s internal employees should be able to carry it through to completion.

Page 25 of 25