Data analytics

February 2014

The role of data analytics in fraud prevention

Background savvy and our experience indicates that they monitoring of business data for potential fraud
are increasingly exploiting weaknesses in IT indicators. Organisations commonly limit their
Business data is increasingly being managed and related controls to perpetrate fraud. When the fraud data analysis to quantification of the
With increased stored by IT systems. The pressure to improve fraud is hidden in large data volumes, manual financial impact when fraud is detected by some
accessibility of business efficiencies and integrate supply chains has meant checks for fraud are simply insufficient. other means.
that many organisations are now heavily reliant
data from internal on IT systems to support business processes. Such Analytics for fraud monitoring However, data analytics techniques can also
have a significant role to play in the early-
and external sources, systems have also reduced the level of human
While the technology enablement of processes warning, detection and monitoring of fraud.
intervention required, which traditionally acted
organisations now as a fraud control. As a result, organisations are
has meant that the data is now available in a These techniques can allow your organisation
structured form for further analysis, the EY to extract, analyse, interpret and transform
have the opportunity placing more reliance on automated controls to both
fraud risk management survey found that your business data to help detect potential
prevent and detect fraud.
more than ever to many organisations are not capitalising on this instances of fraud and implement effective
Additionally, fraudsters are becoming more IT- development and are not performing proactive fraud monitoring programmes.
benefit from using
analytics in their
fraud prevention
programmes.
Julie Fenton, Partner
 Data analytics process

1. Fraud test definition 3. Data cleansing 5. Reporting and monitoring

Define the fraud indicators you Clean the data and convert to Business-focused reports which are easy to understand,
wish to test for based on experi- a format suitable for analysis. summarise results and provide data insights for process
ence, typical business rule limits Import into analysis software owners. If required, the tests can be re-performed on a
and common fraud schemes. for test execution. periodic basis to facilitate continuous monitoring.

Payroll
Ghost employees
Duplicate bank details
Operational
source systems
Payments analysis
Accounts payable
Adherence to limits
Weekend payments
Benfords Law analysis
Payments to unauthorised vendors
Trend analysis

Financial statement close

Manual journal postings/adjustments
Journals not balancing to 0
Journals posted after hours

2. Data identification and extraction

4. Data analysis
Identify source IT system(s) which store
Translate the fraud tests into suitable technical data
the data required and extract this data in a
tests and perform analysis using data interrogation
controlled environment. The data required
techniques to identify unusual trends, data anomalies
for testing should be formally requested in
and control breakdowns.
advance.

The role of data analytics in fraud prevention 2

 Developing analytics tests

Standard tests Customised solutions Data analytics challenges

Why use data analytics? Based on previous experience of fraud in It is also important to recognise that standard While data analytics offers many benefits as a
Improved efficiency Automated your organisation and your knowledge of fraud tests may not be suitable in certain fraud prevention measure, it is also important
method for detecting and monitoring common fraud indicators, you can translate this circumstances. In these cases, it is necessary to recognise the challenges associating with
potentially fraudulent behaviour. knowledge and experience into analytics tests to customise analytics tests specifically for the performing these techniques:
to help detect potential instances of fraud. This needs of your organisation, based on the known
Repeatable tests Repeatable fraud Data quality The results from analytics
set of standard fraud analytics tests can then issues, key risk areas and data available. This
tests that can be run on your data at tests are only as good as the input data.
be run against your organisations business data can include consolidating and analysing data
any time. Before performing tests, it is important
on a periodic basis as part of a fraud monitoring from multiple sources in order to provide the
to assess the quality of data and perform
Wider coverage Full coverage of programme. This can facilitate the automated data insights required. Additionally, custom
validation/cleansing if required.
testing population rather than spot analysis on data from a variety of business techniques are also necessary when you are
checks on transactions better chance accounting cycles such as Accounts Payable, required to drill-down into the source data and Data volumes There may be significant
of finding exceptional items. Accounts Receivable, Procurement Cards, perform additional testing to investigate specific data volumes supporting certain business
Payroll, Expenses, General Ledger, and more, all issues. processes. Your data analytics testing
Early warning system Analytics
designed to look for the red flags of fraud. infrastructure should be capable of
solutions can help you to quickly
handling
identify potentially fraudulent
these volumes.
behaviour before the fraud becomes
material. Data security It is essential that
appropriate security protocols are
considered throughout the extraction and
analysis to protect the confidentiality and
integrity of
source data.

Skillsets Data analytics requires a

combination of
business and technical skills to define the
tests, perform the analysis and interpret
the results in order to provide meaningful
insights.

The role of data analytics in fraud prevention 3

 EY | Assurance | Tax |
Transactions | Advisory

About EY

About us EY is a global leader in assurance, tax, transaction and

advisory services. The insights and quality services we
deliver help build trust and confidence in the capital
markets and in economies the world over. We develop
outstanding leaders who team to deliver on our
promises to all of our stakeholders. In so doing, we play
a critical role in building a better working world for our
people, for our clients and for our communities.
EY refers to the global organisation and may refer to
one or more of the member firms of Ernst & Young
Global Limited, each of which is a separate legal entity.
EY fraud analytics services Our team Learn more Ernst & Young Global Limited, a UK company limited
by guarantee, does not provide services to clients. For
EYs fraud investigation and technology Our dedicated global team of professionals To learn more about how EY can assist you with more information about our organisation, please visit
ey.com.
specialists have developed a set of standard specialise in delivering fraud data analytics your fraud data analytics solution needs, please
fraud analytics tests that can be tailored to solutions for a variety of clients across industry contact: 2014 Ernst & Young. Published in Ireland. All Rights
meet your specific business requirements. sectors. Reserved.
Julie Fenton
These tests can be periodically run against your
This combination of business fraud investigation E-mail: [email protected] 4085.indd 02/14. Artwork by the BSC (Ireland)
organisations business data in order to detect
and technology skills provides an integrated Tel: +353 1 221 2321
or indicate the presence of fraud. These tests ED None.
approach to fraud investigation.
are based on our industry knowledge and fraud Eoin OReilly
Our specialists, who are part of a global The Irish firm Ernst & Young is a member practice of
investigation experience. E-mail: [email protected] Ernst & Young Global Limited. It is authorised by the
analytics network, have significant experience
Tel: +353 1 221 2698 Institute of Chartered Accountants in Ireland to carry on
However, at EY we also recognise that our in data extraction, data cleansing and investment business in the Republic of Ireland.
standard fraud tests may not be suitable in sophisticated data analysis techniques.
Ernst & Young, Harcourt Centre, Harcourt Street,
certain circumstances.
Dublin 2, Ireland.
In these cases, we create custom analytics tests
specifically for your requirements. Information in this publication is intended to provide
only a general outline of the subjects covered. It should
neither be regarded as comprehensive nor sufficient
for making decisions, nor should it be used in place
of professional advice. Ernst & Young accepts no
responsibility for any loss arising from any action taken
or not taken by anyone using this material.

ey.com

The role of data analytics in fraud prevention 4