Embed Size (px)
Citation preview
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Investment advisory services are offered through CliftonLarsonAllen Wealth Advisors, LLC, an SEC‐registered investment advisor. | ©2017 CliftonLarsonAllen LLP
Understanding Your Data and Best Practices in Data AnalyticsCCIA Fall Conference 2017October 5, 2017
Matt Anderson
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Learning Objectives
This course will help you understand your data utilizing a proven data analytics methodology, including best practices in performing data analytics using a variety of tools. Specific learning points include:
– Interpreting your data to decrease time and increase effectiveness
– Normalizing your data for efficient analysis– Robust data analytics methodology– Importance of completeness in data analysis– Best practices in data analytics
2
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
CBOK 2015 Practitioner Survey
• The use of monitoring and data analytics increased by 14% from 2006 to 2015.
• Use of continuous/real‐time auditing increased by 7% from 2006 to 2015.
• Only 40 percent of Chief Audit Executives (CAEs) say their use of technology is “appropriate or extensive”.
• 20 percent of CAEs say their departments rely primarily on manual testing.
Source: www.theiia.org/goto/CBOK
3
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
CBOK 2015 Practitioner Survey
4
80 percent of CEOs say data mining and
analysis is strategically important to their organizations
53 percent of internal audit departments use a software tool for data mining or data
analysis
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Quick Poll• How well is your team utilizing Computer Assisted Audit
Techniques (CAAT)?• What tools are being used for data analysis?• What is keeping us from performing better data analysis?
IDEA / ACL
Data
ExcelIt is difficult to explain what data analysis is, but you know it when
you see it.
5
©2017
CliftonLarsonA
llen LLP
CLA’s Robust Data Analysis Methodology
6
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Multi‐faceted Team Combining Different Skills
The CLA Team:
Technical Data Analysis
Deep Industry
Specialization
Fraud and Forensic
Experience
Compliance, Internal Control Process
Improvement
The CLA Process:
1. Plan and Identify Risk
2. Access Data
3. Analyze and Interpret
4. Report and Document
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Robust Data Analysis Methodology
• Technical analysis• Interpret and Corroborate results
• Identify anomalies• Reassess risk
• Achieve desired objective• Document procedures and results
• Identify best practices and other useful information
• Communicate with IT• Understand what data is available
• Data extraction• Completeness testing
• Identify objectives, understand systems, obtain desired skill set
• Discuss risks, understand control environment, get input from interested users, rank risks Plan and
Identify Risk
Access Data
Analyze and
Interpret
Report and
Document
8
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING 9
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Access Data
• Do not settle for “No” – The data exists, we need to find a way to obtain it
• Do not settle with what you get if it is not what you asked for
• Consider IT personnel involvement– The accounting personnel may not be the best resource– Save time by getting data in the correct format
• Perform a cost benefit analysis• Excel is not always the best format
10
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Access Data (continued)
• Common information types in Higher‐Ed include, but are not limited to: – General Ledger Detail– Accounts Receivable– Accounts Payable– Cash Disbursements– Payroll – Credit Cards – Student Loan– Course Registration– Procurement Records– Travel and Entertainment
Expenses
– Along with the above noted data sets, the following information is also needed to adequately interpret the data files:◊ Data Dictionary◊ Chart of Accounts◊ File Source Listing – indicating from which systems and locations the data was extracted
– Vendor and Employee Master Files– Log files– Unstructured data sources
11
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Technical Data Analysis
12
Verify proper import (anyone)
Test data for completeness (anyone)
Normalize data and prepare for analysis (needs an understanding of “data” and types of tests to be performed)
Design analytics to address identified risk and accomplish objectives (team effort)
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Proper Import
• The second most important step in the data analytic process
• Document source of data (screen shots)• Is your source data format ideal?• Utilize field statistics• Verify data field types are correct
13
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Four Main Types of Data
• Character – text data; this should be used for data that is not, numeric, date, or time. Also, if you are unsure whether a data field maintains the same structure throughout the entire database, character should be used as the “default” selection.
• Numeric – numeric data; generally, the rule of thumb is that if you are going to use (or may use) the field as a part of a mathematical calculation or index key, it should be defined as numeric; the field should contain only numbers (Most DA software will recognize other non‐numeric characters, such as commas, parentheses, periods).
• Date – date data (a consistent mask is required, such as “MM/DD/YYYY”).• Time – time data (a consistent mask is required, such as “HH:MM:SS”).
14
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Completeness
• The most important step in the data analytics process
• Financial data– Reconcile your general ledger detail, by account, to the beginning and ending trial balance
• Operational data– Report totals– System screen shots– Verify and/or observe query scripts used to extract data
15
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Data Normalization
Data Normalization
More Efficient Analysis
More/Less Detailed Analysis
Better Relational Analysis
16
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Common Data Normalization Fields• Absolute Value• Month• Year• Year Month (YYYY_MM)• Account Number Segmented• Fund (parent and child fund)• Fund Type and Fund Subtype• Financial Statement Class • Subsidiary Ledger Identifier• Functional expense allocation
• Workday / Weekend • Day of the week• Days to posted• Balance Sheet/Income Statement• Single Audit/ UGG Compliance
– CFDA / CFSA Number and Description• Shortened version of account number or
vendor field• Statistics by vendor/line item description
17
Data Normalization Now = Significant Time Savings Later
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Technical Data Analysis – Design Analytics
18
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Types of Analytics
• Population analytics — Understand the entire population. – Example test: Stratification by $10,000 increments
• Grouping analytics — Summarize transactions into meaningful groups. – Example tests: Summarization of manual journal entries by accounts, locations, departments
• People analytics —Garner insight into who benefits from and who is responsible for a transaction.– Example test: Count and sum of journal entries by inputting user
19
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Types of Analytics (continued)
• Trending Analytics — Show results through time. – Example tests: Rolling monthly financials, sum cash disbursements by month, faculty course loads and head counts over time
• Transaction Analytics — Isolate transactions exhibiting particular traits. – Example test: Transactions of a certain type and dollar value threshold
20
©2017
CliftonLarsonA
llen LLP
Case Study 1: Student Loan Analysis
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 1: Background• State authority that provides students and families with resources to pursue education beyond high school through grants, scholarships and student loans
• Higher Ed provides student loans to borrowers at fixed rates and has over 250,000 unique loan application numbers totaling in the billions of distributions
22
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 1: Planning
• Our main objective was to understand the makeup of the loan portfolio, get our “arms around” the transactions and run specific queries to address perceived risks
23
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 1: Risk Assessment
• Risk areas we were looking for:– Loan disbursement was less than the outstanding amount– Multiple names with the same social security number or account ID #
– The loan outstanding was greater than $200K– Borrowers who had 10 or more loans totaling over $100K– Loans that had a high balance in FY13 but in FY14, it was either zero or a small balance
– Distributions by fund by month– Borrowers who had 10 or more loans totaling over $100K
24
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Total distributions by month from 2000 through 2014
25
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Distributions by fund by month
26
Funds
Months Trends
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Specific Analysis that would be difficult/impossible without Forensic Data Analysis:
• Total Outstanding Loans by Individual, 2014 totals compared to 2013 where difference is greater than 100k
27
2014 2013
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Specific Analysis that would be difficult/impossible without Forensic Data Analysis:
• 10 or more loans outstanding totaling greater than $100k outstanding
28
©2017
CliftonLarsonA
llen LLP
Case Study 2: Non‐investigative Non‐Instructional Time Analysis
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Background• State college system of over a dozen campuses• System internal audit group identified one campus of concern with applicability to all campuses
• Non‐investigative data analysis• When contracts are issued to faculty or adjuncts, duties may not be adequately explained and documented– Class schedules are clear– Non‐instructional duties are not clear
30
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Planning• What are the known risks and concerns?• What data systems are they using?• What data is available?• Are there common identifiers or links between different data sources?
31
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Risk Assessment
32
• Departments may not be assigning extra duties or paying consistently for those duties– One full time person was spending more than half of time on non‐
instructional duties– Full time are supposed to be teaching at least 12 credit hours per term– Leverage to adjuncts is inconsistent– Course load and students per course variability
• Non‐instructional time– Updating a class or program– Department chair duties– Coaching
– Committee– Lab– Advisor
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Data Sources
33
Employee census/personal
Job assignments (instructional and non‐instructional)
Job authorizationCourse/Section and enrollment
Compensation
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Non‐Inst
34
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Non‐Inst
35
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Non‐Inst
36
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Non‐Inst
37
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: New Risk ID
38
• Data analysis is an iterative process– Step 4 Interpret Results: Continuously reassess risk and improve procedures
• The initial data analysis allowed us to identify a new risk
• The data presented trends of courses with low enrollment
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Low Enrollment
39
New Objectives:• Identify departments with low enrollment by term• Identify EEs with high credit hours per term (> 21)
– Some had credit hours > 100 in a given term– Isolated specific EEs and departments
• Find EEs teaching a high number of credit hours who also had numerous low enrollment course sections
• Enrollment by EE by course/section by term
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Low Enrollment
40
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 2: Zero Enrollment
41
©2017
CliftonLarsonA
llen LLP
Case Study 3: Compensation for Non‐Existing Courses
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 3: Background• Rural college that uses more adjuncts than faculty• Investigative data analysis• Identified risk of improper compensation payments to adjuncts
43
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 3: Data Sources
44
Course Schedule Data
Job Assignments
and Authorizations
Compensation Detail
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Case Study 3: Connecting the Dots
45
Schedule
Job Assignments
Compensation
©2017
CliftonLarsonA
llen LLP
Conclusion
©2017
CliftonLarsonA
llen LLP
WEALTH ADVISORY | OUTSOURCING | AUDIT, TAX, AND CONSULTING
Implementation Considerations
• Build the right team and collaborate…often!• Use a consistent approach.• Perform data analytics early on.• Don’t lose sight of the overall objective .• Are you able to use analytics to replace antiquated
procedures and make them more efficient and effective?• Save import definitions, scripts, and commonly used
formulas.• Share best practices, tips, and tricks.
47
twitter.com/CLAconnectfacebook.com/cliftonlarsonallen
linkedin.com/company/cliftonlarsonallen
©2017
CliftonLarsonA
llen LLP
CLAconnect.com
Questions?Matt Anderson, CPA/CFF, CFE, ASAPrincipal303‐265‐[email protected]
48
