Expected Net Revenue - Amazon Web Services

Download Report

Transcript Expected Net Revenue - Amazon Web Services

Shaping Your Incoming Class and
Maximizing New Revenue
OKLAHOMA CITY UNIVERSITY
Kevin Windholz, Vice President of Enrollment Management
Jacob Dearmon, Associate Professor of Economics
About Us
• Kevin Windholz
– Vice President of Enrollment Management,
Oklahoma City University (2012-present)
– Associate Director of Admission Operations,
Saint Louis University (2006-2012)
– Assistant Director of Admissions,
Washburn University (2003-2006)
– Admissions Counselor,
Washburn University (2001-2003)
About Us
• Dr. Jacob Dearmon
– Education
• Chemical Engineering, B.S.
• Economics, Ph.D.
– Work
• Quorum Business Solutions:
– Software Development
• OCU:
– Associate Professor of Economics, 2012 Faculty Fellow
• Devon Energy:
– Spatial Modeling and Machine Learning
• Economic Research and Policy Institute:
– Sales Tax Forecasting for the City of OKC
– Energy Policy Analysis
Challenges
•
•
•
•
Less overall ability to pay
Increase in unpaid balances
Fear of a tuition hike
Significant increase in discount rates to
preserve headcount
• Less university revenue
• Harder to prove our value
Parent Plus Loans at OCU
total amounts borrowed
7,000,000
6,000,000
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
0
03
04
05
06
07
08
09
10
11
12
Financial Awareness Campaign
• Done extensively with fall 2013 incoming
students, but started small with fall 2012
incoming
• Combination of items new at Oklahoma City
University and those brought from Saint Louis
University
• Involves communication plan in various forms,
and constant review of different types of
financial data
Financial Aid Communication Plan
• Senior Year Plan
– Always ‘proudly’ display cost, with detailed financial
aid and scholarship info close by
– October:
• ‘Financing Your Education’ mailer
– Personal accounts of financial aid
• Financial Aid webinar
• Parent newsletter covering full process
• Scholarship notices begin
– November
• FAFSA 4Caster information
Financial Aid Communication Plan
• December
– File FAFSA Card
• January/February
– Financial Aid (FAFSA Process) Chats
• March
– Awards Packets Begin
• April
– Financial Aid Interviews (with admissions)
Financial Aid Communication Plan
• May/June
– Explanation of charges at STARS 101
• July
– Student Accounts webinar
– Statements are mailed
Financial Aid Interviews
• Admissions counselors are trained as financial
aid counselors
• Give every admit the opportunity to have a
one on one interview about their financial aid
package and potential charges
• Benefits to doing this pre-STARS 101
*credit to Mark Steinlage, Associate Director of Freshman Recruitment,
Saint Louis University
Award Packets
•
•
•
•
•
•
Converted to paper for all new students
Paper letter and cost calculation sheet
Booklet with full explanation of loans
Glossary of financial aid terms
Instructions on how to accept online
Introduction of OCU email
*credit to Cari Wickliffe, Assistant Vice President of Enrollment and Retention
Management and Director of Student Financial Services, Saint Louis University
Paper Statements
• Once a new student has enrolled, they receive
a letter in the email explaining the timeline of
when charges will be billed
• Statement comes in the form of paper (new
students only), with an attachment of how to
understand terms and further explanation of
payment plans
• Instructions on how to pay online are
enclosed
Results: Summer Melt
Freshman First Week Melt
1
11
35
Total Freshman Melt
Total Transfer Melt
2013
77
28
61
2012
Results: Incoming Statistics
• Grew headcount by 7%
• Raised annual incoming freshman net revenue
almost $800,000
• Lowered discount rate by a percentage point
• Freshman-Sophomore Retention Rate – 82%
– up from 80% year prior
Discount Drop
• Modeled the class in house the summer prior,
then set firm budgets with discount giving
areas
• More ‘financial education’ caused less
requests for need based aid
DATA, DATA, DATA
• Looked at Zip Code Household Buying Power
(HBP) for applicants
• FAFSA Data for admits
– Look at EFC levels
– FAFSA positions
FAFSA Positions
2013 OCU Freshman
1st (57%)
2nd (11%)
3rd (9%)
4th (3%)
5th (1%)
Other (19%)
Negotiated Amounts
Jacob Dearmon
Introduction
• Scenario:
– An admitted Business student with a 27 ACT who
lives 10 miles away has come into your office
asking for more aid.
• Questions:
• Will you give him additional aid?
• If so, what amount will you give?
• What factors play into your decision?
Introduction
• The scholarship negotiation affects two key
items:
– The probability that an admitted student will
enroll.
– The net revenue that an admitted student will
generate should he/she enroll.
Introduction
• As the student’s aid increases, so does the
probability of enrollment.
Probability
Add’l Sch. Amt.
Introduction
• Trade-Off:
– Give the admitted student too little scholarship,
he/she won’t come.
– Give the admitted student too much, he/she
won’t generate any revenue.
• Key Question:
– Where is the sweet spot???
Negotiated Amounts
• We use data-mining and optimization processes
to identify the sweet spot and help inform
scholarship allocation decisions.
• This process
–
–
–
–
Is Automated
Uses Historical Data
Is Able to Identify the Student’s Sweet Spot
Is Able to Recommend Scholarship Allocation Across a
Pool of Admitted Students Based on the Greatest
Return
Negotiated Amounts:
Student Level
• This process depends on Expected Net
Revenue which consists of two parts:
– Probability of enrollment
– Net revenue should the admitted student enroll
• The next slide shows the relationship between
scholarship, enrollment probability and expected
net revenue for an individual student.
Negotiated Amounts:
Student Level Graphs
Prob. Of Enrollment Increasing
Exp. Net Rev. Increasing
Negotiated Amounts:
Student Level Graphs
Prob. Of Enrollment Increasing
Exp. Net Rev. Increasing
Negotiated Amounts:
Student Level Graphs
Prob. Of Enrollment Increasing
Exp. Net Rev. Increasing
Negotiated Amounts:
Student Level Graphs
Prob. Of Enrollment Increasing
Exp. Net Rev. Increasing
Negotiated Amounts:
Student Level Graphs
Prob. Of Enrollment Increasing
Exp. Net Rev. Increasing
Negotiated Amounts:
Student Level Graphs
Prob. Of Enrollment Increasing
Exp. Net Rev. Decreasing
Negotiated Amounts:
Student Level Graphs
Prob. Of Enrollment Increasing
Exp. Net Rev. Decreasing
Negotiated Amounts:
Student Level Graphs
“Sweet Spot”
Student Level Report:
Recommendation and Background
Student Level Report:
Add’l Scholarship Data
Current
Rec.
Student Level Report:
Graphs
Negotiated Amounts:
Managerial Concerns
• The availability of discounts might be limited.
• Given this constraint, how should discounts be
allocated across a pool of admitted students?
• How does one control for input quality?
Global Scholarship Allocation
• Allocate dollars based on some user defined
increment up to some maximum amount on a
per student basis.
• Allocation proceeds based on some userdefined combination of gains in Expected ACT
and Expected Net Revenue.
• In other words, dollars are allocated based on
the “biggest bang for the buck”.
Aggregate Level Report:
Current
Aggregate Level Report:
Configuration Settings
Aggregate Level Report:
Optimized Values
Aggregate Level Report:
Scholarship Allocation Rec.
Aggregate Level Report:
Forecast Maps
Aggregate Level Report:
Census Maps
Details: A Closer Look
• Expected Net Revenue depends on the
probability of enrollment.
• How is probability of enrollment determined?
Probability Estimate:
Possible Factors
• Academic Input Quality- HS GPA, ACT
• Financial Aid- Tuition, Fees, Scholarships
(Discount and Endowed), Grants, Loans,
Remission
• Financial Need- EFC
• Demographic Characteristics
• Location- Distance, Census Data
• Preference for OCU
• Economic Data
• School or College
Probability Estimate:
Objective and Considerations
• Objective: 1) To determine a robust estimate of the student’s
enrollment probability accounting for the student’s characteristics
and 2) to capture its sensitivity to changes in the aid amount.
• Considerations:
– How are the appropriate set of factors determined?
• How does one account for the uncertainty regarding the true set of factors?
– How is the appropriate relationship between these variables and
enrollment determined?
• How many methods should be used?
• If more than one, how should they be combined?
– How should results be assessed?
Conclusion
• We have extended, or are in the process of
extending, this program to other areas as well
including:
– Setting the Tuition Amount
– Setting the Scholarship Matrix Amounts
– Assessing Retention Outcomes