CHAPTER 1 corruption that often occurs when mainframe data is
converted into a server/PC character set (ASCII).
ABOUT ACL ANALYTICS
ACL Analytics - a tool that allows auditors and
others to perform data analysis for the purpose
of gaining greater insight into the accuracy and
validity of corporate data by ensuring it was
transacted in compliance with internal policies
and procedures and/or external regulations. It
can also be used for other purposes, such as
fraud detection, data scrubbing and data
conversions.
Who uses ACL Analytics?
Existing ACL Analytics users include:
o Auditors - Internal & External
o Fraud Examiners
o Operational Departments - Finance, Accounts
Payable, IT, Production, etc.
o Most Industries - Manufacturing, Airlines, Retail,
Insurance, Banking, Healthcare, Utilities, etc.
o Companies of all Sizes - Includes Fortune 100 & 500,
Global 500, etc.
o Government Agencies - Federal & State
Why use ACL Analytics?
ACL Analytics is dedicated to data analysis and offers an
expansive range of options, allowing users to perform all of
their tests in one convenient place.
Key benefits
o Easy to use
A vast array of data analysis tools assist with the
complexities of data analysis. An intuitive interface
promotes accessibility and comprehension, allowing for
a remarkably small learning curve.
o Read-only application protects data integrity
Because ACL Analytics is a read-only application, it
prevents data from being accidentally or deliberately
altered. This increased protection of data integrity
means that far greater reliance can be placed on the
accuracy of data analysis results.
o Ability to analyze mainframe data
ACL Analytics can read mainframe data in its native
character set (EBCDIC). This eliminates the risk of data
o Vast array of functions to support analysis
ACL Analytics has over 100 functions that can be
incorporated into expressions. These functions play a
vital role in many data analysis objectives including the
creation of mathematical and financial calculations,
data searches, data harmonization, data scrubbing, etc.
o Analysis steps automatically documented
Most analysis steps are recorded by ACL Analytics into
a log file, creating a documented audit trail to support
data analysis results. The log file allows for accurate
repetition of the analysis steps in two ways: it serves as
a manual reference point (for you or a colleague), and it
can be copied into a script for automated repetition.
o Ability to compare data from disparate systems
Tables that originated from different system types can
be combined (joined) in order to make comparisons
between them. Consider two data files that need to be
compared, one from a mainframe and the other from a
server. Provided that common key fields exist in both
tables, ACL Analytics can use them to combine the
tables to identify matched and/or unmatched records
between the two data files.
o Analyze complete data sets
Oftentimes, large data sets must be broken into smaller
subsets due to file size limit restrictions. ACL Analytics
has no file size limits, meaning that data sets can be
analyzed in their entirety. Analyzing data sets in their
entirety improves data analysis in several ways: reduces
complexity, improves speed and efficiency, and is a
necessity when performing pattern analysis for
indicators of fraud.
o Fast processing
ACL Analytics processes large data files rapidly.
o Automate analytical procedures
ACL Analytics Scripts can be programmed to automate
many analysis procedures, saving time and ensuring
consistency.
The impact of ACL Analytics' use
o Validate effectiveness of Internal Controls
o Drive Down Cost of Regulatory Compliance
o Enhance Business Performance & Bottom-Line
Results
The enabling technologies of ACL Analytics
o Enterprise-Wide Data Testing
o Powerful Analytics
o Continuous Monitoring
How data is stored
There's four components of data storage: files, records,
fields, and bytes.
o Files - A named collection of information
stored and processed as an individual
entity, such as an accounts payable
transaction file or a vendor master file.
o Records - The lines (rows) of data stored within the
file.
A vendor master file containing info. for 1,500 vendors
would contain 1,500 records, one for each vendor.
o Fields - Columns within a file where
particular pieces of information are
stored, such as vendor ID, purchase date,
amount etc. Each entry in a field is known
as a value (where a field and record
intersect).
o Bytes - The individual components of a field that
determine its length; these can be printable and non-
printable characters.
The vendor number AH-3456 is seven bytes long.
o Record Bytes - A record layout is required
for each data file imported to ACL
Analytics. By providing information about
fields within the file, record layouts
contextualize data. Record layouts include
information for each field: name,
description, data type, start position,
length, decimals, format (Datetime
fields).
Fields (Data Types)
There are four field types: character, numeric, datetime,
and logical.
Field Types
Each field must have a defined data type that identifies
how each value in that field will be interpreted by ACL.
Some source file types contain information fully describing
the fields within (ex. Excel files have the field name and
data type), whereas others have a combination of letters,
numbers, and symbols that are meaningless until context is
applied.
What might this string of numbers represent?
o 09082016
This eight byte string could be interpreted as:
o Character fields have values with printable
(alphanumeric and symbols) and non-printable
characters (line feeds, tabs etc.).
Unsure if a field is character or numeric?
o Numeric fields contain values that can be used for
mathematical calculations.
Example
o Datetime. In order for ACL Analytics to properly
interpret datetime values, datetime fields must include
the format, such as MM/DD/YYYY. Since ACL uses
this format to interpret the values within the field, it's
important that all the values follow the format.
o Logical fields are filters that evaluate as true or false
(discussed later).
ACL for Windows
ACL for Windows is a single, downloadable product that
contains ACL Analytics, and links to resources that will be
of use to you.
Key Features
o Profile and Help - View or update your user profile,
sign out of ACL Launchpad, or view online Help.
o Organization selector - Switch between accounts
(organizations) you have access to using this drop-
down list.
o Recent Analytics Files - Open recently
accessed ACL projects or analysis apps.
o Sample Files - Open pre-built ACL projects that
include a variety of sample data.
o Open - Open an existing ACL project, analysis app, or
GRC project.
o Create - Create a new Analytic Project or Workflow in
Results Manager for organizing, tracking, and
remediation of exceptions.
o Community - Ask questions, share product ideas,
contact support and access templates, content, training
and other resources as part of your ACL subscription.
Interface and basic navigation
1. Menu bar - The menu bar provides access to most ACL
Analytics features including menu commands for
working with ACL projects, performing audit analysis,
and configuring options and connection settings.
2. Navigator - The Navigator displays information about
the project using three tabs at the bottom of the pane:
o Overview - displays all the items that belong to the
project.
You can right-click any project item to perform an
action.
To organize items in the Overview, right-click the
project icon and select New > Folder.
You can drag any project item into the folders that you
create in the Overview.
o Log - displays the ACL Analytics command log, where
all actions taken with the project are recorded and
organized chronologically.
Double-click log entries to open them, and right-click
log entries to perform an action.
o Variables - displays the names, values, and data
categories of any variables in the project. The contents
of the tab are dynamically updated as variables are
created, deleted, or changed.
3. Status bar - The status bar displays information such
as the name of the active ACL table, the record count,
the details of any filters currently applied to the table,
and the name of any currently running script.
4. Command line - The command line allows you to
enter ACL Script commands. To display the command
line, if it is not visible, in the menu bar, click
Window > Command Line.
5. Filter text box - The Filter text box and drop-down list
allow you to apply a filter to the data displayed in the
View tab. You can also enter one or more search terms
to perform a quick search of the data.
6. Display area - The display area is your core view
which displays data, results from your analysis and
scripts.
7. View tabs - The tabs available here, allow for multiple
views to exist for the same set of data, allowing you to
switch between tables using the different view tabs that
you can create and customize. You can pin the View
tab if you want to open more than one table.
The ACL Analytics Environment
Project
The project is similar to a file folder within a filing cabinet;
supporting documentation items for each audit are filed in
separate, clearly named folders. For each new analysis or
audit, a new project should be created to contain the
relevant documentation. There are several item types
within an ACL project such as table layouts, Views, the log,
folders, and scripts.
Table layout
When an ACL Analytics project is first created, it's empty
since the source file(s) are stored outside of the project.
Creating a table layout links the source file to the project.
In the ACL Analytics Environment Process Diagram, the
source data file (AP2016.dat) is linked to the project via the
table layout AP2016. Table layouts contain various
information:
o Data path to the location of the source data file
o Field names, start positions, lengths, types, number of
decimal places and datetime formats
o Named expressions (calculated fields and named filters)
o View definitions
View
A View determines how data is displayed on the screen
within ACL Analytics. Whenever a table layout is created,
a View (named Default_View) is automatically generated
that contains all of the fields within the table.
If you only want to work with certain fields, you can create
a new View that has the desired fields. A table layout can
have multiple Views. Changing a View does not affect the
underlying table layout or other Views.
Log file
The log records most analysis steps and results. While you
can access the log from within the ACL project, it's stored
as its own file (eg. ACL Demo.log) outside the project.
In the ACL Analytics Environment process diagram, the
log is named APAudit2016.log.
Example
Folder (not pictured)
You can use folders within the project to organize project
items.
Quick Sort and Quick Filter
Quick Filter and Quick Sort are ACL features that allow
you to quickly perform basic tasks by selecting a field or
value, right-clicking, and selecting the desired option.
o Quick Sort allows you to quickly sort fields in
ascending or descending order
o Quick Filter allows you to build basic filters based off
a selected value
Commands
ACL Analytics commands are pre-defined routines and are
organized into three menu bar choices: Data, Analyze, and
Sampling.
Data Manipulate (combine or reorder) data.
Analyze - Understand (test) data.
Sampling - Create and evaluate sample pools.
Executing basic commands
The basic commands in the Analyze tab of Analytics will
be useful in helping you analyze and get a statistical picture
of the data. Click play for an overview of these.
To follow along, use the project TNE.acl which is located
in the folder 'Demo Training Data'.
What's a command parameter?
Command parameters detail what information or data needs
to be included in order for the command to run. Some
parameters are required, meaning that they need to be
fulfilled in order for the command to run. Other parameters
are optional, meaning that the command will still run even
if they aren't fulfilled. If you attempt to run a command
without fulfilling a required parameter, the command will
fail and you will receive an error message.
One example of a required command parameter is the
"value" parameter which is used to specify the field(s) that
the command will run on. If you do not select a field when
attempting to run a command, you will receive an error
message like the one shown below:
About the log
The log provides a chronological record of work performed
on a project. The log automatically records every step taken
in a project. This provides an audit trail of the analysis, and
allows users to retrace and/or re-create the steps taken to
produce their results, print a report, or create a script to
repeat the analysis process. Each analysis step is written to
the log as an entry, with log entries organized into log
sessions.
Basic ACL Analytics environment
o Source files, such as an Excel file reside outside of the
ACL project.
o Corresponding .FIL files are created for each source
file. .FIL files also reside outside of the ACL project
and are referenced by the table layout. In a way, the
table layout acts as a window to the .FIL file.
o You can view the log in the Navigator of your project.
Sub Files
When a new table is created as the result of running a
command, a .fil sub file is created. In the image above, the
source file AR 2016.dat is 129 KB and contains Accounts
Receivable data for all of 2016. If we use the Extract
command to create a new table, AR_Q1_2016, that
contains only the AR data from the first quarter, a sub file,
AR_Q1_2016.fil is created. Since AR_Q1_2016 is a subset
of the AR_2016 file, it makes sense that it is only 32 KB.
o AR_2016 contains data for the full year, but analysis is
to be performed on Q1 data.
o Use the Extract command to create a sub-file
containing Q1 data.
o Subsequent analysis will be faster on the smaller, sub-
file.
o ACL Analytics commands that create new tables always
create new sub-files with .FIL extensions.
Data Analysis Cycle
Every data analysis project follows a cycle consisting of
five phases:
1 Planning
1 Importing data
1 Preparing data
1 Analyzing data
1 Reporting issues
To ensure accuracy and efficiency, its important to
consider all phases from the beginning. By 'efficiency,' we
mean thoroughly (anomalies otherwise undetected can be
identified in ACL Analytics) and timely (ACL Analytics
supports speedily completion of complex tests). To
remember the five phases of data analysis, use the acronym
 PIPAR, which stands for Planning, Importing, Preparing,
Analyzing, and Reporting.
CHAPTER 2
EXPRESSIONS
Working with expressions
Expressions are statements used primarily to create filters
and computed fields. They perform calculations, specify
logical conditions, or create values that do not exist in the
data file. Expressions can be named and saved as part of a
project or else created just for immediate use.
The output of an expression can be returned in any of the
four data types: logical, character, numeric, and datetime.
Filters (Logical expressions)
Filters are logical expressions, because it produces
outcomes based on if a particular condition is met -
evaluating each record as being true, or false. Because
logical expressions restrict the records that can be viewed
or analyzed to those that meet the expression criteria, they
are known as filters. For example, you can create a filter
that selects only records that fall within a specified range of
dates.
Note
Filters can be named or unnamed. To save a filter, name
it.
Computed Field (Character, Datetime, Numeric)
A computed field is a virtual field, derived from a
calculation that references either physical field(s) and/or
previously created computed field(s). Computed fields can
return character, numeric, date, or logical values, and can
be conditional or unconditional. Once created, they can be
treated as if they were physical fields.
In the table above, you can see that the last three
expressions all create a computed field (convert, calculate,
and add). These result can be seen in the table by adding
the new computed field into the view.
Note
A computed field must be named (i.e. always saved).
Filter or Computed Field?
When trying to discern if an expression you are creating
will be a filter or a computed field, consider whether or not
data is being filtered to see the results (apply a filter) or
created as a physical field (a computed field).
Considerations
ACL Analytics evaluates expressions from left to right,
according to the following rules:
o Operators are evaluated in order of arithmetic
precedence.
o Use parentheses ( ) to modify the order in which the
expression is evaluated.
o Each operator works only if its operands are of an
acceptable type.
Using expressions
ACL expressions are combinations of values and operators
that perform a calculation and return a result.
Expressions are a valuable and flexible tool. You can use
them to:
perform a wide range of calculations
create filters
prepare data for analysis
create computed fields
The content of expressions
Expressions can include data fields, functions, literals,
constants, and variables, which can be combined using
arithmetic or logical operators.
You can enter expressions manually, or you can build them
using the Expression Builder, which provides a standard
utility throughout ACLfor creating expressions.
The complexity of expressions
Expressions can be as simple or as complex as needed.
A simple expression could return the result of a basic
arithmetic operation. For example, Quantity * Cost .
A more complex expression could reference a number of
fields and functions and use operators to combine the parts
of the expression.
For example:
PROPER(first_name) + " " + PROPER(last_name)
 converts all the names in o literal numeric values, without quotation marks limited to
the first_name and last_name fields to proper case (initial digits, a minus sign if needed, and a decimal point if
needed
capital letter followed by lowercase), and joins the first and
last names with a single space between them.
Datetime
Types of expressions
Contains any of the following:
ACL supports four types of expressions, which correspond
o datetime fields
with the four supported data categories, or data types:

o variables that contain datetime data

character
o functions that return datetime values
numeric
o quoted datetime values (datetime literals)
datetime
The Datetime data type encompasses three subtypes: date,
logical datetime, and time.
Quoted datetime values require backquotes for
For example:
example, `20141231` or `20141231.235959` . The
backquote (`) is the lowercase key at the upper left corner of
Amount + 1 is a numeric expression because it the keyboard.
performs an operation on numbers and returns a
numeric result. Logical

Amount > 1 is a logical expression because it Contains any of the following:

makes a comparison and returns a logical result of
True or False.
The content of any expression you create must correspond
with the expression type:
o an operation that produces a logical result of True or False
(T or F)
o functions that return logical values

Character If T or F are part of the expression, they must be entered

without quotation marks.
Contains any of the following: Note
A logical expression can reference fields, variables, or
o character fields literals of any data type.

o variables that contain character data How ACL evaluates expressions

ACL evaluates expressions according to the following rules:

o functions that return character values

Operator precedence - Arithmetic and logical precedence

o quoted character strings (character literals)
dictates the order in which operators are evaluated.
See Operators in ACL expressions.
Numeric
Use parentheses ( ) to modify the order in which the
Contains any of the following: operators are evaluated.

o numeric fields

o variables that contain numeric data
o functions that return numeric values
Operand data type - Each operator works only if
its operands are of a compatible data type.
Function parentheses - All ACL functions require
parentheses. Everything inside a function's
parentheses is evaluated first, before any other
 parts of an expression outside the function's
parentheses.
Comparing character strings - By default, when character
strings of different lengths are compared, the shorter of the
two lengths is used.
If the Exact Character Comparisons option is selected in
the Tables tab in the Options dialog box, the longer of the
two lengths is used. For more information, see Table tab
(Options dialog box).
Decimal precision - If numbers with different precision are
mixed in numerical operations, the result retains the decimal
places of the operand with the largest number of decimal
places in the expression.
For example:
o 4 + 5.0 = 9.0
o 1.1 * 1.1 = 1.2
o 6 * 2.000000 = 12.000000
Note
You can use the SET MATH command to change the number
of decimal places that result from a mathematical operation.
Operators in ACL expressions
Listed below are the some of the basic operators in ACL.
The operators are listed in decreasing order of precedence.
When operators of equal precedence appear in an
expression, they are evaluated from left to right unless
you use parentheses to specify an operator precedence.
( ) Parentheses modify operator precedence,
or enclose function parameters
NOT - Logical NOT
- Unary minus minus sign, indicates a
negative number
^ Exponentiation raises a number to a power
* Multiply
/ Divide
+ Add
- Subtract
+ Concatenate character strings
Greater than
< Less than
= Equal to
>= Greater than or equal to\
<= Less than or equal to
< >Not equal to
AND (or &) Logical AND
OR (or |) Logical OR
About Filtering
A filter sifts records, only returning those that meet
specified conditions; filters evaluate as true or false and
return records that meet the specified conditions. The
values in the filter syntax govern which records will be
returned after applying the filter.
Filters are constructed using a combination of field names,
operators, and strings, dates, or values.
Considerations
o The benefit of filters is that they allow filter results to
be visually inspected without having to extract the
filtered records into a new file.
o When a command is run with an active filter, it only
produces results on the filtered records, but the entire
file still has to be processed. This can be inefficient
when a small number of records are filtered on a large
file.
o Only one filter can be in effect at a time.
o When a filter is active the filter condition is visible in
the filter window (at the top of the work area) and the
status bar (at the bottom of the work area). On the
status bar, the record count is preceded with a question
mark. To generate a record count of filtered items, run a
command like Count or Total. These can be found
under Analyze on the menu bar.
Creating filters using the Expression Builder
The Expression Builder can be used to create both
filters and computed fields.
Considerations
o When filters arent saved, they are not added to the
table layout. Prior to AN10, this meant that they could
 not be reused. Since AN10, the last ten unsaved filters
are remembered by the table and can be re-selected by
clicking on the dropdown arrow to the right of the
Filters bar.
o In the table layout, saved filters appear with an L
(logical) in the category column.
Troubleshooting Filters
Common error message
If you click the Edit View Filter button to access the
Expression Builder, the expression you create in the
Expression Builder must be a logical expression (filter). If
you create an expression that does not evaluate as logical
(true/false), you'll receive an "Expression type invalid"
error message like the one shown below.
In this example, the expression entered was InvValue *
UnitCost, which produces a numeric output instead of
logical.
Searching
Search for a Starting String
Filters can be used to find specific words.
By default, the search behaves like a wildcard:
o looks for every occurrence where the value begins with
the filter string
o it doesnt matter what is after the string
o Only character fields can be used for word searches.
o The search can be performed on any characters
within the field (A-Z, 0-9, or symbols).
o The string being searched for will only be picked up
if it's at the beginning of the string within the field
(ie. nested words are not returned)
o Alpha characters (A-Z) are case sensitive
Did you know?
The wildcard behavior is controlled by Exact Character
Comparisons (ECC). If ECC is on, the remainder of the
field value after the search for string must be blank in order
for it to be picked up. Consider the filter LastName =
"SMITH". The table below shows how values will evaluate
if ECC is turned on or off.
ECC can be found under Tools > Options >Table.
About Quick Searching
Quick searching allows you to find whole or partial words,
exact phrases, numbers, or datetime values. All the source
data in the table is searched, not just the data displayed in
the current view. Entire records are searched, rather than
specific fields. You can subsequently modify the search if
you need to search a specific field (character data searches
only).
Quick Searching a table
1 In the Filter text box:
o Type one or more search terms separated by a space
and press Enter.
With multiple search terms,
the quick search performs a logical OR operation
and finds records that contain at least one of
the search terms.
o To search for an exact phrase, enclose the entire
phrase in double quotation marks and press Enter.
The quick search finds only those records that
contain the exact phrase.
o When searching for numbers or datetime values,
you need to match the source data formatting
rather than the formatting in the view. For more
information, see About quick searching.
1 To search a specific field:
o Modify the automatically generated filter in the
Filter text box by typing a comma after
the search term, and then add the name of the field.
For example, modify FIND("casino")
to FIND("casino", Merchant).
You must use the physical name of the field,
which may not be the same as the display name of
the field in the view. To check the physical name,
right-click the appropriate column header and
select Properties. If necessary, copy the physical
name from the Column Contents text box at the
top of the Modify Column dialog box. Do not use
the Alternate Column Title.
To search in a related field you must specify the
fully qualified name of the field (that
is, table.field name). For
example: FIND("casino", Vendor.Vendor_Name)
o Press Enter.
Unconditional Computed Fields
An unconditional computed field applies the same
expression to every record in order to produce a new field.
 Note: A computed field must be named and always saved.
What's a computed field?
We've already looked at the four data types (numeric,
character, logical, and datetime). These are fields that
physically exist in the data, but what if you need additional
fields? Such as a field that combines the first and last name
fields to produce a full name? You can create
a computed field that calculates values based on existing
fields (eg. First_name + Last_name).
Considerations (all computed fields)
o Only exist in the table and are not a part of the data
source.
o Must be named.
o Can return a character, numeric or date output.
o The rules for using the Expression Builder to create a
computed field are similar to those for creating a filter
as detailed in the Expression Builder page.
Conditional Computed Fields
When different calculations need to be applied based on
different conditions within the file, one expression, known
as a conditional computed field, can be created. In a
conditional computed field, values are calculated
conditionally.
Note
A computed field must be named and always saved.
About conditions and values
In a conditional computed field, different values are
assigned when a specified condition is met. Every
conditional computed field has a default value which is the
value that is assigned if none of the conditions are met.
Order in which conditions are evaluated
The conditions are evaluated in the order in which they are
listed. Once a record has evaluated as true, that record
cannot be re-selected by any subsequent condition. In the
examples below, the order that the conditions are listed
does not matter since none of the records will meet more
than one condition, but in some cases they will.
Considerations
o The default value is the value to apply to any records
that dont meet any of the conditions. The default value
can be character, numeric,
logical, or datetime. Consider the calculation for a
discount. If transactions are less than $1,000, no
discount applies. So the default value here is 0.
o Conditions must be logical (such as TotalAmt >=
1000).
o Values must be the same data type as the default value.
If the default value is 0.0000 then a valid value would
be TotalAmt * .0625.
o If the expression is numeric, the number of decimal
places is controlled by the number of decimals in the
default value.
o You can enter as many condition and value pairs as
needed.
Functions
In an expression, functions are used to apply calculations
for a wide variety of purposes.
Commonly used functions
PROPER( ) - Changes strings to proper
(sentence) case.
UPPER( ) - Changes strings to upper case.
LOWER( ) - Changes strings to lower case.
DEC( ) - Determines the number of decimal
places that values are rounded to.
The BETWEEN( ) function produces a logical output,
meaning that it returns all records that answer true to the
parameters. BETWEEN( ) works with datetime, character,
or numeric fields, filtering all the records between and
including the minimum and maximum values. Examples of
how BETWEEN( ) works with different field types are
shown below.
Syntax
The syntax for the BETWEEN( ) function is:
BETWEEN(value, min , max)
where:
PARAMETERS
Value - The field, expression, or literal value to
test.
Min - The minimum value of the range.
Max - The maximum value of the range.
o Each parameter is separated by a comma (a space can
be used instead of the comma).
 o Brackets denote optional parameters.
o No space between the function name and the opening
parenthesis.
The ABS( ) function produces a numeric output
and converts all negative values to positive. It
can also be used to create a logical output
(filter)
Output
o Character
o Numeric
o Datetime
o Logical
Considerations
o Learn more about a specific function by selecting
Help > Contents and entering the function name.
o Access functions from the function list (right-hand
side of the expression builder).
o By default, the Paste Parameters check-box is
checked, meaning that when a function is selected
from the list, the parameter prompts are pasted into
the expression.
Fixed-Point Arithmetic
When multiplying or dividing, ACL Analytics uses fixed-
point arithmetic. This means that the result is rounded to
the largest number of decimal places used in the
expression.
Note
If the result contains more decimal places than the
number with the largest in the expression, it will be
rounded up or down in order to have the correct number
of decimal places.
Expressions with multiple calculations
Fixed-point arithmetic applies for each calculation in an
expression. For expressions with multiple calculations, the
largest number of decimal places is taken from each
calculation.
CHAPTER 3
IMPORTING DATA\
Identifying data sources
There are several common resources and techniques that
help ensure that the correct data is identified: IT
partnership, pre-audit meeting, data dictionaries, sample
data, system reports, data entry, ODBC, and Direct Link.
IT Partnership
IT departments frequently need to be involved in the data
access process. Their knowledge of corporate databases
and the technologies involved is often essential when it
comes to locating data, but auditors frequently encounter
obstacles when it comes to speaking with the right IT
personnel and/or obtaining the correct information from
them. There are two major factors which influence this:
o Lack of process
IT departments usually operate a ticket process for
requests. This process is often generic and doesnt lend
itself to the detail necessary when it comes to locating
the correct data. It is extremely difficult to complete a
ticket if you dont really know what it is you are asking
for.
o Disconnect
There is often a disconnect between you and IT
whereby you understand the business processes and the
reasons why the data is required for the various tests
that need to be performed whereas the IT department
understands the technologies involved but often has
limited appreciation as to why the data is being
requested. The net result is that you will very likely
receive incomplete data.
Training Audit staff in ACL Analytics usage is only one
of the steps involved in setting them up for efficient and
successful optimization of ACL Analytics. It is essential
that there is correct infrastructure that allows the
Auditors to communicate effectively with IT.
Oftentimes, it is necessary for senior personnel within
the IT and Audit departments to meet in order to ensure
there is a workable process for Auditors to request data.
This process should include opportunities for Auditors
to meet with IT personnel to discuss the detail of their
requests; a revision of the ticket system may be
necessary in order to allow for comprehensive
information to be included within the request ticket and
the solution ticket.
Pre-Audit Meeting
During the pre-audit meeting, try to obtain as much
information as possible from the auditee concerning the
available data. Depending on the scope and complexity of
the audit, it may be beneficial to have an IT team member
 present so that they can better understand the data
requirements and offer insight into the process.
Data Dictionaries
Each corporate database should have a data dictionary
describing its tables and fields. The various data
dictionaries can provide a useful reference for identifying
the available files and fields for an audit. IT knowledge
will still be needed to help you understand the data
dictionaries, but the more that they are used, the more self-
sufficient you will become.
Sample Data
Databases often contain obscure field names, which makes
it difficult to know whether the data contained within them
is necessary for the audit / analysis in question.
Consideration should be given to requesting a sample file
containing a small number of records and with every field.
This sample file can be reviewed in ACL Analytics in order
to determine which fields are required for the audit. Over
time, a central library can be built to contain sample files
that can be reviewed for content when needed.
System Reports
Obtain reports from the business area under audit/ analysis.
When meeting with the IT Department, it can be useful to
show reports containing the fields required for the
audit/analysis in order to help them identify which tables
and fields need to be accessed.
Data Entry
Meeting with data entry personnel can provide context and
enhance understanding of data; there are several benefits
that can contribute to this:
o Screen shots taken of the data entry screens can often
help IT ascertain in which table(s) the required data is
located.
o Some applications can identify field names by right-
clicking on the various elements of the data entry
screen(s).
o Data entry personnel can highlight workarounds that
were implemented when the application being used
doesnt fully meet the requirements of the business i.e.
social security numbers being entered into a memo
field.
o Change logs can be reviewed to ascertain if it is
possible to make non-compliant entries. This will help
identify where potential data quality issues may exist.
o The data entry process can be compared to the
appropriate policies and procedures to ensure the
process is in compliance.
ODBC
Open Database Connectivity will be discussed in detail in
this chapter. It allows ACL Analytics to connect directly to
relational databases, allowing you to create your own
downloads. This has the benefit of allowing you to peruse
the database and download data without reliance on the IT
Department.
Direct Link
Direct Link is an ACL product that allows ACL Analytics
to connect directly to SAP databases. With this tool, it is
possible to peruse the database and search for specific
keywords within file and/or field names.
Each organization has its own challenges when it comes to
the identification of data. There is no magic "cure," but
embracing some of the techniques suggested here will yield
huge benefits. The truth is in the transactions so it is
incumbent on the Audit team to do what they can to gain
more knowledge of their corporate database technologies in
order to understand where the data resides. With this
knowledge, the Audit Department will increase its
effectiveness and greater reliance can then be placed on the
completeness, quality, and accuracy of audit issues.
Acquisition Methods
There are three ways to acquire source data:
o Source Data Access
o Database/Warehouse Access
o Receive a Copy of the Data
Source Data Access
You are provided with a user name, password, and
instructions to access the host system. You must extract the
data yourself. Many systems have an extraction tool that
you can use to create a file that ACL Analytics can read.
These tools often allow you to choose the download file
type, so consider which file types contain the highest
quality of data. For example, given a choice between Excel
or delimited, delimited is the preferred choice.
Database / Warehouse Access
From ACL Analytics, you are granted access to either a
production database or a data warehouse. ACL Analytics
reads the data directly from the source using data access
technologies such as ODBC (Open Database Connectivity).
The benefit of this approach is that it allows you to
download the required data directly into ACL Analytics
without having to make data requests from IT or other data
providers.
 For SAP users, ACL Analytics offers a product called
Direct Link, which allows direct access to SAP databases.
It provides search tools specific to the SAP environment
that are not available when using ODBC.
Receive a copy of the data
The data is requested, often from the IT Department, and
provided in a workable file format. Ask which file formats
are available and select the one that is most likely to
contain the highest quality of data.
To ensure the data provider fully understands your
requirements, consider the following when making data
requests:
Location - The name of the system where the
data resides.
Name - The file name, table name, or report
name.
Date bounds - Be sure to specify whether the
request is inclusive or exclusive of the start and
end dates.
Filter criteria - Caution! Minimize filter criteria
in order to reduce the risk of receiving incorrect
data. It is better to create the filter yourself after
the data has been imported to ACL Analytics so
that you are in control of the filter criteria.
File type - The preferred file type(s) that you
would like to receive the data in.
Medium - The medium on which you would like
the data transferred i.e. flash drive or place in a
specified location on the network.
Deadline - The date by which you require the
data.
Controls - Control totals and record counts.
Field names - When requesting a delimited file,
be sure to request field names. If field names
arent available, request a record layout.
Record layout - When requesting a fixed-width
file, be sure to request an accompanying record
layout.
Conversion - If data is provided from a
mainframe, don't allow the data to be converted
from EBCDIC to ASCII. Conversion can cause
corruption of some field types, especially packed
fields. ACL Analytics can read mainframe data in
its native form so conversion isn't necessary.
Source file formats
Source file format refers to the format the data file is in
prior to importing into ACL Analytics. While there are
several file formats that can be imported, some are more
reliable than others; it is important to select the most
reliable file format available. In order of reliability, the
available file formats are: 1) fixed-width files, 2) delimited
files, 3) Excel/Access files, and 4) report (print image)
files. ACL Analytics can also import other file types.
Fixed-width files
A fixed-width file contains data without field definitions. It
can be created from servers or mainframes; each field has a
fixed-width. To ensure field width consistency, values are
padded with blanks. Consider a field with the value "soda"
(four bytes long) and another value "aspirin" (seven bytes
longs). In order to maintain field width consistency, "soda"
would be padded with three blanks, making it seven bytes
long. Oftentimes, data from older mainframe systems is in
the data encoding EBCDIC. Because many applications
cannot read EBCDIC, data providers often convert
mainframe EBCDIC data into ASCII. This can create errors
in some data types, especially packed fields. A packed field
is a mainframe numeric field that has been compressed to
save disc space. Such fields can be correctly decompressed
in table layouts with the Packed data type. ACL Analytics
reads and processes EBCDIC without converting it into
ASCII, preserving data quality and integrity.
Delimited files
Unlike a fixed-width file, a delimited file typically has a
variable record length because variable length fields are not
padded with blanks. Instead, a delimited file contains
delimiters to separate fields. A delimiter, such as a comma,
specifies the boundary between fields.
What's a text delimiter?
The most common type of delimited file is a CSV (Comma
Separated Value). Delimited files can be in either data
encoding: ASCII (server) or EBCDIC (mainframe). They
can also contain field names, so be sure to ask for them in
your request. If field names cannot be included, you will
need to request an accompanying record layout.
Best practice
Request a delimiter such as a tab or a pipe ( | ) to reduce
the risk of non-delimiters being recognized as delimiters
Microsoft Excel/Access files
Although Excel and Access files can be imported to ACL
Analytics, they are not the preferred formats because the
original data must be converted from server/mainframe to
Excel. Often, during this conversion, data is corruption
(such as dropping leading blanks or zeros) occurs. Due to
Excel file size limits, multiple Excel files may be produced
for one data set. Nevertheless, there will be situations when
Excel/Access files need to be imported to ACL Analytics.
Best practice
Discuss other file options, such as fixed-width files (no
conversion) or delimited files (limited conversion), with
your data provider as these file types have no size limits
Report (print image)
A report file is the soft copy of a printed report. It contains
page breaks with header information repeating on every
page. Only use report files when other file types are not
available or when the report itself needs to be audited /
analyzed for accuracy. Report file structures can vary
significantly and ACL Analytics contains many features to
handle these complexities.
PDF (report)
Although PDF files are increasingly popular, they are not
the preferred file format for accessing data and should only
be used when alternative files, such as fixed-width or
delimited, are not available. Only the text elements of a
PDF file can be imported to ACL Analytics; scanned
images cannot be interpreted. A PDF file that does not
contain any text will be rejected. In ACL Analytics, a PDF
file is treated as a report file because it typically contains
data that is formatted like a printed report.
Other files
ACL Analytics can also import other file types, including
XML, AS400, Cobol Copybook and SAP (Private File
Format).
Importing Fixed-Width Files
Considerations
o Select Platform For Data Source window only opens if
ACL Analytics server integration is enabled under
Tools > Options.
o Ensure EBCDIC data is not converted to ASCII.
o If EBCDIC data contains packed fields, use the
"packed" data type.
o Fields defined as numeric in the Edit Field Properties
screen will be converted to Print, which is a numeric
data type that ignores commas and correctly interprets
parentheses and "CR" as negative. Signs can be leading
or trailing, fixed or floating and any other punctuation
(such as $) is ignored.
Direct Access Method
When an ACL Analytics table layout is created for a fixed-
width file, it will be linked directly to the fixed-width file
without making a copy.
Defining overlapping fields
Importing Delimited Files
Import and Copy Method
When an ACL Analytics table layout is created for a
delimited source file, it cannot be linked directly to the
delimited file, so during the import process a copy will be
written to a fixed-width file with a .FIL extension. The
table layout will be linked to the copy and not to the
original delimited file.
Importing Report (Print Image) Files
Import and copy method
When an ACL Analytics table layout is created for a source
report file, it cannot be linked directly to the report file;
instead, during the import process, a copy is written to a
fixed-width file with a .FIL extension. The ACL table
layout is linked to the .FIL and not to the original report
file.
Reviewing hard copy of report
Prior to beginning the import, review a hard copy of the
report (or the soft copy in Notepad or WordPad):
1 Identify the fields that need to be imported into ACL
Analytics.
1 Identify structural elements of the report.
1 Identify which structural elements contain
o Header Records (optional)
Elements above the detail records that they belong
to.
o Footer Records (optional)
Elements below the detail records that they belong
to.
o Detail Records
These are considered the "anchor" in the report.
Importing PDF Files
Import and copy method
When importing PDF files, the Data Definition Wizard
(DDW) copies the PDF data to a text file. This copied text
file is then imported the same way as a report file (identify
detail, header and footer records and create identification
filters), with an additional DDW step that allows you to
select the number of pages to import. Oftentimes, this
conversion to a text file corrupts field alignment, making it
extremely difficult to properly import the data.
To reduce the risk of field alignment corruption, it is
recommended to open the PDF file in Adobe Acrobat and
save it as a text file by selecting File > Save As > Text and
then use the Data Definition Wizard to import the new text
file. However, if you convert the PDF to a text file in this
way, you will no longer be able to select specific pages to
import.
For directions on how to import PDF files, refer to the
report file import steps.
Importing from an ODBC compliant data source
About ODBC
Short for Open DataBase Connectivity, ODBC is a standard
database access method developed by the SQL Access
group. ODBC was developed to allow access to any data
from any application, regardless of which database
management system (DBMS) is handling the data. ODBC
achieves this by inserting a middle layer, called a database
driver, between an application and the DBMS. This middle
layer translates the applications data queries into
commands that the DBMS understands.
For this to work, both the application and the DBMS must
be ODBC-compliant - that is, the application must be able
to issue ODBC commands and the DBMS must be able to
respond to them. ACL Analytics is ODBC-compliant. To
use it on a DBMS that is also ODBC-compliant, an ODBC
driver is required.
Setting up an ODBC Driver
An ODBC Driver needs to be setup on the same computer
that is running ACL Analytics. This is a one-time setup,
typically done by IT personnel. Installed ODBC drivers can
be found in the Windows Control Panel under
Administrative Tools > Data Sources (ODBC).
Note
Some databases will have restricted access, in which case
youll need a user name and password in order to allow
read-only access.
The Data Access window
The Data Access window is a component of ACL Analytics
that contains data connectors which you can use to import
data to ACL from a wide range of sources. For a complete
list, see Data sources you can access with ACL .
The Data Access window also contains features for filtering
and joining tables prior to importing.
Advanced ODBC imports
You may have noticed a few things in the Data Access
window while you were navigating the activity:
o The toggle. This switches to a
text box where you can freely edit or insert SQL syntax
to extract specific data.
o More than one table can be placed into the staging area,
which allows them to be joined.
These are handy features, however are considered
advanced topics - so we've put together introductory
material of Advanced Importing Concepts , to get you
started in the appendix.
Caution
SQL joins are different from ACL joins. So unless you're
an SQL pro, maybe leave this until the end of the course.
Reusing table layouts
By eliminating field definition and other table layout
definition steps, reusing table layouts can increase
efficiency. We'll look at three methods for reusing table
layouts:
o reusing table layouts of fixed-width files
o refreshing from source
o reusing the IMPORT entry from the log
There is one method for reusing table layouts created from
Fixed-Width source files. There are two methods for
reusing table layouts created from all other source file
types that used either the Import and Copy Method
(Delimited, Excel, Report, and PDF) or Direct Access
Method (ODBC).
Reusing table layouts of fixed-width files
When you have two or more fixed-width files that share the
same record structure, it is only necessary to use the Data
Definition Wizard to define the first one. This table layout
can then be copied and pasted as many times as necessary.
The copies can be renamed and linked to the other fixed-
width files. Compared to the alternative of creating the
same field definitions multiple times, this process saves
considerable time.
Refreshing from source
Whenever the Data Definition Wizard creates a table layout
using the Import & Copy method, table layouts can be re-
used using Refresh from Source. Since the created table is
linked to the .FIL copy of the source file, selecting Refresh
from Source will overwrite the .FIL file to reflect any
changes that occurred in the source file. In ACL Analytics,
the View will reflect any changes made to the .FIL file.
Source file types that can be refreshed from source:
o Delimited file
o Excel file
o Access file
o Report (Print Image)
o PDF
o ODBC
Reusing IMPORT from the log
Whenever the DDW creates a table layout using the Import
& Copy method, table layouts can be re-used by adjusting
and reusing the import statement that was written to the
log. The benefit of this is that the file names within the
import statement can be changed to allow a different source
file to be imported and to prevent overwriting the first .FIL
file and associated table layout.
Reusing IMPORT entry from the log
If data paths are eliminated from the Command
Line syntax, ACL Analytics will automatically use the
default Windows working folder (the Windows folder
where the ACL Analytics project is located).