Professional Documents
Culture Documents
Improving The Research Request Process
Improving The Research Request Process
Rick Melnyck
Introduction
The Division of University Advancement at the University of Miami is charged with the
responsibility and authority to coordinate all fund-raising activities of the university.
President Shalala and other central, medical and school-based development officers are
responsible for qualifying, cultivating, soliciting and stewarding donors for the University of
Miami.
The Advancement Research Department identifies, analyzes, tracks and compiles
biographical and financial information on potential donors for President Shalala and central,
medical and school-based development officers.
Define Phase Research Requests
(1) Business Case
What is the name of the process?
The name of the process is Research Report Requests.
What is the aim of the process?
The aim of the process is to get the Development Officers, VPs, AVPs and President Shalala the right
amount of information on a particular prospect at the right time while making more efficient use of
the Research Departments time.
Why do the project at all?
The Research Requests project should be conducted to satisfy the needs of the Development
Officers, VPs, AVPs and President Shalala in regards to information they need on potential donors.
Why do the highest priority project now?
There seems to be a crisis in the Research office as the Researchers feel that they are overworked
and understaffed. Many of the Development Officers do not use the Research Department and a lot
of their time that should be spent cultivating and soliciting donors is spent doing research. Also, the
Research staff is spending 100% of their time doing research reports and do not have time to
identify new donor prospects for the University.
What are the consequences of not doing the highest priority project?
The consequences of not doing the Research Requests project would be twofold. Firstly, the
Researchers would continue to feel overwhelmed with the amount of work they receive which could
lead to frustration and possibly turnover. The Development Officers will continue to spend part of
their time doing research when their time could be better spent with donors. Finally, if the
researchers do not have time to identify new prospects the donor pool may start to dry up for the
Development Officers.
Hard benefits
Research department
40% misclassified research requests
X 520 requests per year
= 208 misclassified reports/year
X 10 hrs average extra time spent due to misclassification of report
= 2080 extra hours spent on misclassified reports
X $25/hr researcher salary (including benefits)
= $52,000 saved per year
New prospects identified per month
The researchers will have over 2000 more hours to work on prospect identification.
Since the researchers will have more time they will be able to identify prospects for the DOs to
cultivate and solicit, those fundraising dollars would be added to the hard benefits of this project.
Development Officers
The hard benefits for the Development Officers are a little tougher to calculate. However having a
couple more hours per day to meet with prospects could result in many more donations than would
have resulted had they been doing their own research.
(2) SIPOC Analysis
Suppliers
Development
Officers
Vice-President for
University
Advancement
President Shalala
Associate VicePresidents
Other
Development
Staff
Inputs
Name of
Prospect
Name of Report
Requestor
Type of Report
Needed
Date Report
Requested
Date Report
Required
Process
See flowchart
below
Output
Research Profile
Type of Report
Total Number of
Pages
Number of Reports
Requested
Customers
Development
Officers
Vice-President for
University
Advancement
President Shalala
Associate VicePresidents
Other
Development Staff
Start
Research Director
receives request
Request is assigned to
researcher
Researcher completes
report
No
Research Director
Reviews report
Yes
End
Affinity Diagram
Theme (Focus Point)
I identify my own
prospects, I might get one
from research once in a
blue moon
I do research to find my
prospects which takes
away from the time I could
be out there raising
money
I do most of my own
prospect identification and
it is labor intensive
Driving Issue
CTQ
Reactive process
Cycle time to
complete report
(number of days
until report
completed)
Ability to identify
new prospects
# of new prospects
identified by
research per month
Project Objectives: To decrease cycle time to complete reports by 30% and to also increase the
number of new prospects identified by the Research department from the current level of zero, both of
these objectives should be realized by December 31, 2006.
Measure Phase Research Requests
Introduction
The measure phase contains three steps: operationally defining each CTQ, conducting Gage
Reproducibility and Repeatability (R&R) studies on each CTQ, and developing a baseline for each CTQ.
(1) Operational Definitions
From the define phase, it has been determined there is one CTQ, cycle time to complete the research
report (the number of days it takes to take the request and deliver the completed report to the
requestor).
The CTQ will be operationally defined in terms of the following three parameters:
1. Criteria
2. Test
3. Decision
Operational Definition for CTQ: Cycle time to complete report (number of days until report
finished)
Criteria for the cycle time to complete report: Open the database and accumulate all of the different
reports for each requestor for each year.
Test: Find the date the report was completed and the date the report was requested
Decision: Subtract the requested date from the completed date, determine the number of days.
(2) Establishing the Validity of the Measurement System for Each CTQ
CTQ: Cycle time to complete report (number of days until report finished)
There is no measurement error to speak of. For each report requested we simply look at the date
completed and subtract the date requested from it to get the cycle time to complete the report.
N
962
N*
0
Variable
C2
Maximum
168.000
Mean
10.271
SE Mean
0.386
StDev
11.961
Minimum
0.000000000
Q1
4.000
Median
7.000
Q3
14.000
25
50
75
100
CTQ
125
150
175
An Individuals-Moving Range chart was completed to show a baseline for cycle time to complete the 962
research requests in 2004 and 2005. The y-x axis represents the length of time (in days) it took to
turnaround a research request, each value on the x-y axis represents each of the 962 requests made in
the two year span. The chart shows a wildly out of control process.
Individual Value
160
120
1
1
1
1
1
80
1
111
11
1 1
5
5
40
0
2
2
2
2
2
2
2
2
2
2
6
6
2
6
15
2
6
6
6
6
2
1
52
6
62
2
2
2
2
2
2
2
2 6
26
2
6
2
26
6
6
66
6
6
2
2
2
6
2 6
8
8
6 2
96
192
1
5
5
2
5
6
6
6
1
5
5
22
2
2
2
2
2
2
2
2
288
2
2
2
2
2
6
6
6
2
2
2
384
2
2
2
2
2
2
2
2
2
2
26
2
1
1
6
2 2
6 6
6
2
1
11
51
88
66
6
6
2
22
2
2
2
2
2
2
2
2
480
Observation
68
6
6
6
2
2
2
64
6
6
6
2
6
6
2
4 6
1
8 5
8
2
2 2
2
2
2
2
2
6
6
6
66
6
2
2
6
6
2
2
2
226
2
2
2
2
2
2
8
8
6
6
666
6
6
6
6
6
6
2
2
6
6
2
2
6
2
66
2
6
6
6
6
672
768
86
1
1
160
Moving Range
1
1
6
6
6
2
2
68
576
11
1
1
120
1
1
1
1
80
1
1
1
1
1
1
1
1 11
1 1
40
0
1
1
2
2
2
2
2
22
2
96
1 1
2 2
2
222
2
2
2
2
2
2
2
2
192
1 1
2
2
2 2
2
2
2
2
2
2
2
22
2
2
2
288
2
2
2
2
2
384
2
2
2
2
2
2
2
2
2
22
2
1
1
1
1
11
1 1
1
1
1
2
2
2
2
22
2
2
2
2
2
2
2
480
Observation
2
2
576
1
1
1
1
1 11
1
11 1
1
2
2
2
2 2
2
2
2
2
2
2
2
2
2
2
2
2
2
2 2
2
2
2
2
2
2
672
Process capability cannot be calculated for this process as it is out of control. Further, cycle time and
priority are correlated.
768
2
2
2
2
86
Research Director
receives request
Request is assigned to
researcher
Researcher completes
report
No
Research Director
Reviews report
Yes
End
The sources of the Xs came from process knowledge gathered by interviewing the Research Director, the
Research Analysts as well as the Development Officers (DOs).
Operational definitions of Xs
Operational Definition for X1: Number of reports requested from each requestor per year
Criteria for the number of research requests per year per requestor: Open the database and look at the
table containing research requests for each year.
Test: Check to see if each request has the following criteria:
1. The name of the requestor
2. The area of the requestor
3. The name of the prospect being profiled
4. The type of report requested
5. The data the report was requested
6. The requested completion date
Decision: If all six criteria have been met then a research request has been submitted, if not all six
criteria have been met then a research request has not been submitted. Add up the total number of
research requests per year per requestor.
Operational Definition for X2: Type of report ordered (L1, L2, B, T)
Criteria for type of report requested: Open the database and pick a report request.
Test: Check to see the type of report in each report request.
Decision: The report will be one of the following: Level 1, Level 2, Blurb, Trustee or Special.
Operational Definition for X3: Priority
Criteria for priority of the report request: Open the database and pick a report request.
Test: Find the value for the date the report request was submitted and the date the requestor wants the
report completed by.
Decision: Subtract the date the report request was submitted from the date the requestor wants the
report completed by to get the priority for the report request in days, for example, a report is submitted
on the 14th and is needed on the 17th, so, the priority is 3 days
100
70
60
50
1
1
30
+3SL=23.29
+2SL=19.74
_
+1SL=16.18
C=12.63
-1SL=9.07
-2SL=5.52
-3SL=1.97
20
10
2 2 2 1 1
12
15
1 1
18
21
Sample
24
27
30
Sample Count
Sample Count
40
80
60
a r o s z f a n k s s z a s a n a s s o r n x r e n s r e o n r
al le r i e l e of r o ni o e e g r r e in r e n e a o le id a e e s tr la e
al c h s a Jon z a ian sha Nix lot od rvi c om r ie oge rr e oh ba we on S a alk em C shc r b am ar nllingHou as oo O th
i A Br B e
C D
Sh Mis Ro Gon T al Be D. P r a S S e G No R He is C Ro Po A ngles W stl
M
e
y
r
an D
is i
ar
Ca
an
C a tee
m
Ch
Am Ch
ex om
Di
le
us
Al Na
o
r
C
T
Count 75 52 48 36 32 29 19 19 15 11 1110 10 10 9 7 6 5 5 4 4 3 3 3 2 2 2 2 2 1 1 4
Percent 17 12 11 8 7 7 4 4 3 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 1
Cum % 17 29 40 4855 62 66 70 74 76 7981 83 85 87 89 90 9193 93 94 95 96 96 9797 98 98 99 99 99100
40
Count
60
Percent
Count
80
1 1
12
16
20
Sample
24
28
32
36
500
400
C1
1 1 1
100
+3SL=25.3
+2SL=21.6
_
+1SL=17.8
C=14.1
-1SL=10.3
-2SL=6.6
-3SL=2.8
20
500
200
33
300
40
20
0
C1
100
400
80
300
60
200
40
100
20
ez la io is a es ez ff e rs n es ik es n er os ro er a ra ox nc er tz ck ke n a se ns ra er
al ala ar av ieg on m no bi d ge ixo ic otn on uso ttn od st ng ain r e C la de ar Be Bla ool adz in ou r so ha th
nz h os DNor J Go alia A r Ro . N erv Pl ng r g Ma S C a elli ob Her LeB Ra chw
D r H Pa Bes O
T
e Ga
A Fe a r a
D R
Go S R
i De S
S
an
r m x is
C
oms te
Di
a
No Ale
u
N Tr
Percent
80
60
200
40
100
Le
2=
l
ve
Percent
Number Ordered
80
300
Type of Report
Number Ordered
100
400
20
In
2/
th
ep
-d
b
ic
te
i le
ial
as
lur
of
ec
ida
/B
Pr
=B
Sp
nd
e
l1
a
=
BL
v
e
S
v
ti
eC
Le
cu
te
xe
1=
us
Tr
=E
E
=
C
156
35.3
35.3
Count
Percent
Cum %
126
28.5
63.8
107
24.2
88.0
24
5.4
93.4
23
5.2
98.6
3
0.7
99.3
Number
78
217
3
182
25
1
2
12
Type
1=Level 1/Basic
2=Level 2/In-depth
B=Bio Summary
BL=Blurb
C=Trustee Candidate
E=Executive Profile
M=Memo
S=Special
r
he
Ot
Type of Report
500
100
400
80
300
60
200
40
100
20
l
y
e
th
rb
ia
si c
ar
at
lu
ep
ec
Ba
id
m
-d
=B
Sp
1/
m
nd
BL
/In
el
Su
S=
Ca
2
v
o
e
l
Le
Bi
te
ve
1=
us
Le
B=
Tr
2=
C=
Count
Percent
Cum %
3
0.7
100.0
217
41.7
41.7
182
35.0
76.7
78
15.0
91.7
25
4.8
96.5
Variable
C2
N
962
N*
0
Variable
C2
Maximum
169.000
Mean
11.228
SE Mean
0.397
StDev
12.308
Minimum
0.000000000
Q1
4.000
Median
8.000
Q3
15.000
Individual Value
160
120
11
1
1
1
80
1
1
111
111 1
6
40
0
1
1
5 15
5
2
22
8
2
2
2
2
2
2
6
2
2
2
2
2
2
2
2
22
2
22
2
2
2 6
62
6 8
1
5
222 2
2
2
2
2
2
2
2
2
2 6
2 2
2 62
2
2 6
6
62 6 2 6
96
192
1
5
5
288
384
2
2
2
2
2
2
2
26
88
66
6
66
11 1
1
1
51
5
5
5
2
2 2 22
2 22
2
2
2
6
6
6
68 8
6
6
2
2
2
2
2
22 6
6
6
6
2
2
6
2
24 6
6
480
Observation
576
1
1
2
6
6
2 6
1
81
5
8
1
1
1
1
11
11
1
51
2
6
6
2
2
2
2
2 22
2 2 66
26
2
2
2
2
2
2
2
2
6
6
6
66
6
6
6
2
2
2
28
8
6
6
66
6
6
2
2
2
2
2
2
2
2
2
6
66
6
6
6
2
2 6
6
6
6
2
22
26
6
6
672
768
864
+3SL=30.5
_
+2SL=24.1
+1SL=17.6
X=11.2
-1SL=4.8
-2SL=-1.6
-3SL=-8.0
960
1
1
160
Moving Range
11
120
1
1
1
1
80
1
1
40
11
1
1 11 1
11
1
1
1
2
2
22
2
2
2
96
22
2
2
2
2
2
1
2
2
2
22
192
1
2
2
2
288
1
1
1 11
1
11
11
1
1
2
2 2
22
2
22 2
2
2
2
2
2
384
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
22
2 2
2
2
2
2
2
2
2
2
480
Observation
576
1
11
1 1
1 1 1
672
11
1
1
1
1
1 1
2
22 2
2
2
2
2
2
2
2
2 22
2
2
2
2
2
2
768
2
2
2
864
1
1
1
11 1
11
2
2 2
2
2
2
2
2
2
2 2
2
2
960
+3SL=23.6
__
+2SL=18.2
+1SL=12.7
MR=7.2
-1SL=1.8
-3SL=0
-2SL=0
12
2.3
98.8
3
0.6
99.4
er
th
3
0.6
100.0
Percent
Type
1=Level 1/Basic
2=Level 2/In-depth
BL=Blurb
C=Trustee Candidate
E=Executive Profile
M=Memo
S=Special
Name
Donna Shalala
Mischler
Rosario
Jones
Gonzalez
Talianoff
Total
#
75
52
48
36
32
29
272
%
27
20
17
13
12
11
100%
N=100
N1=27
N2=20
N3=17
N4=13
N5=12
N6=11
Number
Misclassified
6
13
5
4
3
9
40
Number ok
21
7
12
9
9
2
60
Number
Misclassified
4
5
11
12
3
5
40
Number ok
29
19
1
0
7
4
60
Name
Gonzalez
Shalala
Rosario
Davis
Noriega
Jones
Total
#
104
77
39
38
30
29
317
%
33
24
12
12
10
9
100%
N=100
N1=33
N2=24
N3=12
N4=12
N5=10
N6=9
Develop Hypotheses about the Relationships Between the High Risk Xs and the CTQs
X1: Number of reports requested from each requestor per year
The number of reports requested from each requestor per year seems to be a large contributing factor to
cycle time to turnaround reports. Upon further investigation it seems that only 1 researcher is assigned
to President Shalala, the VP and AVPs even though they account for 60% of the research requests.
When a large amount of requests from them comes in all of the other researchers put their own projects
on hold to work on these projects.
Pareto Chart - Requests by Requestor - 2004
100
400
200
100
0
a r o s z f a n k s s z a s a n a s s o r n x r e n s r e o n r
al le r i e l e of r o ni o e e g r r e in r e n e a o le id a e e s tr la e
al c h s a Jon z a ian sha Nix lot od rvi c om r ie oge rr e oh ba we on S a alk em C shc r b am ar nllingHou as oo O th
i A Br B e
C D
Sh Mis Ro Gon T al Be D. P r a S S e G No R He is C Ro Po A ngles W stl
M
e
y
r
an D
is i
ar
Ca
an
C a tee
m
Ch
Am Ch
ex o m
Di
le
us
Al Na
o
r
C
T
Count 75 52 48 36 32 29 19 19 15 11 1110 10 10 9 7 6 5 5 4 4 3 3 3 2 2 2 2 2 1 1 4
Percent 17 12 11 8 7 7 4 4 3 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 1
Cum % 17 29 40 4855 62 66 70 74 76 7981 83 85 87 89 90 9193 93 94 95 96 96 9797 98 98 99 99 99100
C1
60
40
Count
300
Percent
Count
80
20
0
C1
100
400
80
300
60
200
40
100
20
ez la io is a es ez ff e rs n es ik es n er os ro er a ra ox nc er tz ck ke n a se ns ra er
al ala ar av ieg on m no bi d ge ixo ic otn on uso ttn od st ng ain r e C la de ar Be Bla ool adz in ou r so ha th
nz h os DNor J Go alia A r Ro . N erv Pl ng r g Ma S C a elli ob Her LeB Ra chw
D r H Pa Bes O
T
e Ga
A Fe a r a
D R
Go S R
i De S
S
an
r m x is
C
oms te
Di
a
No Ale
u
N Tr
Percent
500
Level 1
Level 2
Blurb
Trustee
25
50
75
100
125
150
Other
175
25
50
75
CTQ
100
125
150
150
175
CTQ
The reason for the outliers in the dotplots above is due to the reports requested not being needed right
at the moment. The reports were requested well in advance of them being needed meaning the
researcher worked on other reports that were a higher priority (needed sooner) and worked on the
outlier reports right before their due date.
X3: Priority (Due Date Date Requested)
The dot plots for Cycle Time to Turnaround Reports (CTQ) and Priority to Receive Reports indicate:
1. The values of Cycle Time to Turnaround Reports seem to be close to those of Priority to Receive
Reports, meaning that the priority to receive the reports is a contributing factor to cycle time to
turnaround reports (CTQ)
Dotplot for Priority to Receive Reports
25
50
75
100
125
150
175
25
50
75
100
CTQ
CTQ
Each symbol represents up to 8 observations.
125
N
40
N*
0
Mean
11.48
SE Mean
2.62
StDev
16.55
Minimum
0.000000000
Q1
5.00
Median
7.00
Q3
13.75
14
28
42
56
CTQ
70
84
98
2005
Variable
C1
N
40
N*
0
Mean
12.80
SE Mean
1.63
StDev
10.32
Minimum
1.00
Q1
5.00
Median
10.00
Q3
15.75
Maximum
44.00
12
18
24
CTQ
30
36
42
Misclassification of Reports
After speaking with some of the Research staff as well as some of the Directors of Development it was
indicated that some of the reports may have been misclassified, meaning that the report requested was
not the report that was actually needed. In the Analyze phase we took a stratified random sample of the
6 people who made the most requests (in both 2004 and 2005) it was found that there was a
misclassification rate of 40% in each year.
After some investigation it is clear that this problem is caused by the requestor filling out the online
report request form above and asking for a report that they do not need. The requestor ends up getting
more information than they will use and the research department spends valuable time doing research
for nothing as well as many hours of overtime.
The prospect management cycle consists of the following stages and the type of report needed at each
stage in the process is indicated in parenthesis:
1. Identify the prospect no report is needed
2. Qualify the prospect blurb report is needed (can often be provided by DODs assistant)
3. Cultivate the prospect Level 1 report is needed
4. Solicit the prospect Level 2 report is needed
5. Steward the prospect no report is needed
In order to solve this problem the online report request form above will be modified so that the report
requestor does not ask for a specific report but instead states the stage of the prospect management
cycle that they are in. The Director of Research will then determine upon receiving the request which
report will be completed depending on the stage in the prospect management cycle the prospect is in.
Reports do not meet standards
Another problem with the process is the fact that the Research Director must send many reports back to
the researchers before sending them to the requestors because the report does not meet quality
standards.
Many of the same mistakes are made over and over again and waste not only the Research Directors
time but also the individual researchers time as well.
To solve this problem we will create a quality checklist for the researchers to go over before they submit
the report to the Research Director for review. In addition, they will submit their report to their peers for
a peer review before sending it to the Research Director.
PM Status
Qualify
Components of Research
Job Title
Brief Business Description (Include spousal occupation only if
relevant. i.e. Education Prospects)
Home Market Value
Estimated Gift Capacity
UM Affiliations
Current Contact Information
Note: A Prospecting Blurb should not exceed 3-4 sentences. Customize, within reason, for your particular DOD.
While some prefer simple one liners, stating only Estimated Gift Capacity and Affiliation, others prefer all info.
available for such a request.
Miscellaneous
Event Blurb
* Usually reserved for Presidential
and Provost related
events.
Job Title
Brief Business Description
Significant Directorships
Significant Charitable/Civic Affiliations
Spousal Information, if Attending
UM Affiliations
Summarized UM Giving
Basic - Level 1
Cultivate
In-depth - Level 2
Ready to Solicit
NOTE: By this point, the DOD should be very familiar with prospect. A level 1 profile is already on file and does
not need to be repeated.
If on a rare occasion they have no prior research, you would have to inlcude Level 1 information as well.
Depending on the DOD and their preferences,
you may also combine/update your Level 1 with the Level 2 for a final product. Some may prefer simply the
taregted info.
Memo Report
Miscellaneous
cultivation meeting
(i.e. Recent sale of a company, pending lawsuit, or simply a
current address and phone number)
Also remember to check the following:
Spelling
Grammar
Style (AP)
Peer Review
NOTE: During a Peer Review, the reviewer should always write their name at the top of the draft reviewed. Apart
from the elements listed above,
remember to look out for information that does not add up, or seems off the mark...
This document is for internal use within the Research Department only.
Exponential
-0.5 + EXPO(1.23)
0.021524
The 4 different types of reports each have their own processing times.
processing times and they were as follows (in hours):
Blurb Report
POIS(4.81)
Level 1 Report TRIA(16,30,42)
Level 2 Report TRIA(36,48,60)
Trustee Candidate Report TRIA(4,5.95,8.95)
A simulation model and experiment was created using the ARENA simulation software package without
making any changes to the system (see below). The results were compared to the actual system
performance to verify your model.
The results obtained in Arena were very close to the actual system.
Actual
System
962
10.27
# of reports in 2 years
Cycle time (in days)
95% CI
898
9.25*
(866, 931)
(8.52, 9.98)
* Difference may be related to the fact that in the actual system there is a step where the reports are checked
by the Research Director and sent back for revision if not up to specifications. This was omitted from the
model due to the fact that there is currently no Research Director and a good estimate of the time and
frequency of that part of the process could not be made.
Re q u e s t o r
m
Com m unic at o
i n Type of
Repor t
u n c
i a t o
i n
1 0 0
E s
l e
Ds
i p o s e
A s s g
i n
0T
Re q u e s t o r
Engineer ing
E n g n
i e e r n
i g
r u e
F a s
l e
Ds
i p o s e
0
A s s g
i n
M
Re q u e s t o r
u s c
i
M usic
A s s g
i n
Re q u e s t o r
S t u d e n t
A f f a ri s
St udent
A s s g
i n
Re q u e s t o r
B o a r d
o f
T r u s t e e s
0T
r u e
0T
r u e
F a s
l e
Af f air s
F a s
l e
0
Tr ust eesT
Boar d of
r u e
F a s
l e
Ds
i p o s e
0
A s s g
i n
Judaic St udies
Re q u e s t o r
J u d a c
i
6 7
E s
l e
A s s g
i n
Re q u e s t o r
Nu r s n
i g
Nur sing
7
1 1
E s
l e
Re s e a r c h e r
A s s g
i n
Re q u e s t o r
RS M
# 1
Re c o r d
Re s e a r c h e r
1
Re c o r d
Re s e a r c h e r
RSM AS
A S
1 0
5 0
E s
l e
A s s g
i n
Re q u e s t o r
Co r p o r a t e
Re s e a r c h e r
# 2
2
A s s g
i n
L e v e l
Re p o r t
T m
i
3 8
E s
l e
A s s g
i n
Re q u e s t o r
Re c o r d
Ar chit ect ur e
Re s e a r c h e r
3
1 2
5 0
A s s g
i n
L e v e l
Re p o r t
T m
i
E s
l e
A r r iv e s
A s s ig n
T im
E n t e r n
i g
Ar ea Request
Fr om
Co m
Re c o r d
Com ing
p e
l t e d
Re p o r t
A l
S e n t
t o
Re q u e s t o r
0
1 . 3 6 1 3
. 3 1 4 1
. 4 1 8 8
A s s g
i n
Re q u e s t o r
E s
l e
. 3 1 4 1
3 . 7 6 9 6
. 3 1 4 1
2 . 9 3 1 9
2 . 0 9 4 2
5 . 0 2 6 2
3 . 3 5 0 8
Ds
i p o s e
Re s e a r c h e r
# 3
A r c h ti e c t u r e
Re p o r t
5 . 1 3 0 9
1 . 6 7 5 4
3 . 2 4 6 1
. 8 3 7 7
3 . 1 4 1 4
3 . 8 7 4 3
Educat o
i n
A s s ig n
B lu r b
Re c o r d
Re s e a r c h e r
Re p o r t
T m
i
Re q u e s t o r = = 1 9
1 0
2 9
Re q u e s t o r = = 1 8
E s
l e
0T
Re q u e s t o r
Re q u e s t o r = = 1 5
Re q u e s t o r = = 1 4
r
u e
Law
A s s g
i n
T r u s t e e s
Re p o r t
T m
i
Re q u e s t o r = = 1 3
Re q u e s t o r = = 1 2
L a w
Re q u e s t o r = = 1 1
Re q u e s t o r = = 1 0
Ds
i p o s e
A s s g
i n
Re q u e s t o r
F a s
l e
At hlet c
i s
0T
0T
u e
Lowe M useum
L o we
Re q u e s t o r
A u
l m
F a s
l e
Alum ni Relat o
i ns
n i
7 0
7
E s
l e
A s s g
i n
Re q u e s t o r
Libr ar ies
u e
L b
i r a r e
i s
F a s
l e
Ar t s and Sciences
A s s g
i n
Re q u e s t o r
A r t s
a n d
S c e
i n c e s
3 5
4 5
E s
l e
A s s ig n
Re q u e s t o r
P r e s d
i e n t
S h a a
l a
l
Pr esident
Shalala
4 9
1 3
E s
l e
A s s ig n
Re q u e s t o r
V P
Vic e Pr esident s O f f ic e
4 2
1 2
4 2
E s
l e
Re c o r d
Re s e a r c h e r
Re s e a r c h e r
# 5
5
Re q u e s t o r = = 7
Re q u e s t o r = = 6
Re c o r d
Re s e a r c h e r
Re q u e s t o r = = 1
Re s e a r c h e r
# 6
6
Re q u e s t o r = = 3
Re q u e s t o r = = 2
2 3
3 2
A s s g
i n
Re q u e s t o r = = 9
Re q u e s t o r = = 8
Re q u e s t o r = = 5
Re q u e s t o r = = 4
E s
l e
Re q u e s t o r
A t h le t ic s
A s s g
i n
Re q u e s t o r = = 1 7
Re q u e s t o r = = 1 6
E s
l e
A s s g
i n
9 . 7 3 8 2
1 5 . 9 1 6 2
3 6 . 5 4 4 5
Re s e a r c h e r
# 4
4
E d u c a t io n
The different scenarios and modifications of the system that were experimented with were the following:
1. Alternative #1 One additional Research Analyst was hired @ $40,000 per year and assigned to
help Research Analyst #1 with the reports for President Shalala, VPs and the Board of Trustees.
This was due to the fact that the queue and cycle times were the highest for that area.
2. Alternative #2 In this scenario Research Analyst #5 (the least busy) was re-assigned to help
Research Analyst #1 with the reports for President Shalala, VPs and the Board of Trustees while his
reports were re-assigned to Research Analyst #2 (he had the 2nd lowest utilization)
3. Alternative #3 All Research Analysts work from a single queue and the first available Research
Analyst works on the next report in the queue.
See the charts below for the results of the 95% confidence intervals of the following three
performance measures after 15 replications of each alternative were run for 24 hours a day for 730
days (2 years):
1. Cycle time (in hours)
2. Average time report is in queue (in hours)
3. Utilization of Research Analysts.
Cycle Time (in hours) Comparison Between Current System and 3 Alternatives
Current System
Alternative #1
Alternative #2
Alternative #3
Avg
Avg of all
Researcher 1
Researcher 2
Researcher 3
Researcher 4
Researcher 5
Researcher 6
Researcher 1&7
Only 1 queue
Researcher 1&5
65.99
89.44
24.32
42.15
27.87
41.57
37.56
-
Half
width
5.81
10.04
2.37
1.24
1.01
1.73
2.06
-
95% CI
Avg
(60.18,71.8)
(79.4,99.48)
(21.95,26.69)
(40.91,43.39)
(26.86,28.88)
(39.84,43.3)
(35.5,39.62)
-
32.96
24.32
42.15
27.87
41.57
37.56
30.98
-
Half
width
.49
2.37
1.24
1.01
1.73
2.06
.86
-
95% CI
Avg
(32.47,33.45)
33.31
36.26
42.15
27.87
37.57
30.98
(21.95,26.69)
(40.91,43.39)
(26.86,28.88)
(39.84,43.3)
(35.5,39.62)
(30.12,31.85)
-
Half
width
.53
1.89
1.24
1.01
2.06
.86
95% CI
Avg
(32.78,33.84)
(34.37,38.15)
(40.91,43.39)
(26.86,28.88)
(35.5,39.62)
(30.12,31.85)
27.12
27.12
-
Half
width
.30
.30
-
95% CI
(26.82,27.42)
(26.82,27.42)
-
Average Time in Queue/Report (in Hours) Comparison Between Current System and 3 Alternatives
Current System
Alternative #1
Alternative #2
Alternative #3
Avg
Researcher 1
Researcher 2
Researcher 3
Researcher 4
Researcher 5
Researcher 6
Researcher 1&7
Only 1 queue
Researcher 1&5
65.43
1.80
6.62
3.99
3.42
6.49
-
Half
width
10.15
1.05
.65
.75
1.29
1.62
-
95% CI
Avg
(55.28, 75.58)
(.75, 2.85)
(5.97, 7.27)
(3.24, 4.74)
(2.13, 4.71)
(4.87, 8.11)
-
1.80
6.62
3.99
3.42
6.49
6.97
-
Half
width
1.05
.65
.75
1.29
1.62
.83
-
95% CI
Avg
(.75, 2.85)
(5.97, 7.27)
(3.24, 4.74)
(2.13, 4.71)
(4.87, 8.11)
(6.14,7.8)
-
5.96
6.62
3.99
6.49
6.97
Half
width
1.62
.65
.75
1.62
.83
95% CI
Avg
(4.34,7.58)
(5.97, 7.27)
(3.24, 4.74)
(4.87, 8.11)
(6.14,7.8)
.32
-
Half
width
.1
-
95% CI
(.22,.42)
-
Utilization of Research Analysts Time Comparison Between Current System and 3 Alternatives
Current System
Alternative #1
Alternative #2
Alternative #3
Avg
Researcher 1
Researcher 2
Researcher 3
Researcher 4
Researcher 5
Researcher 6
Researcher 1&7
Only 1 queue
Researcher 1&5
.6938
.06
.21
.12
.10
.19
-
Half
width
.02
.01
.02
.01
.01
.01
-
95% CI
Avg
(.6738, .7138)
(.05, .07)
(.19, .23)
(.11, .13)
(.09, .11)
(.18, .20)
-
.06
.21
.12
.10
.19
.35
-
Half
width
.01
.02
.01
.01
.01
.01
-
95% CI
Avg
(.05, .07)
(.19, .23)
(.11, .13)
(.09, .11)
(.18, .20)
(.34,.36)
-
.1631
.21
.12
.19
.35
Half
width
.01
.02
.01
.01
.01
95% CI
Avg
(.1531,.1731)
(.19, .23)
(.11, .13)
(.18, .20)
(.34,.36)
.23
-
Half
width
.01
-
The different performance measures for all three alternatives in relation to the current system came
out statistically significant, as none of the 95% confidence intervals of their means overlapped with
the 95% confidence interval of the mean of the current system.
95% CI
(.22,.24)
-
In regards to cycle time (in hours), the most important metric the results can be summarized as
follows:
Current System
(68.16, 79.84) 95% C.I.
Alternative #1
(32.47,33.45) 95% C.I.
Alternative #2
(32.78,33.84) 95% C.I.
Alternative #3
(26.82,27.42) 95% C.I.
The results from Alternatives 1 and 2 appear to be fairly close and there is no statistical significance
between them as the 95% C.I.s of their means overlap. Alternative #1 would require the Division of
University Advancement to hire an additional Research Analyst at an annual salary of $40,000. The
increase in performance does not appear to be enough to justify this expenditure.
Therefore amongst Alternatives explored in this study Alternative #2 is the preferred solution.
Research Director
receives request
Request is assigned to
researcher depending
area requestor is from
Researcher completes
report
Researcher completes
quality checklist and peer
review and submits report
to Research Director for
review/approval
No
Research Director
Reviews report
Yes
End
Misclassified Reports
August 1- October 3, 2006
Total number of reports requested
Total number of reports misclasified
% misclassified
100
14
14%
The changes which were implemented reduced the misclassification reports from 40% to 14%
9
12
timetocomplete
15
18
21
N
100
N*
0
Variable
timetocomplete
Q3
13.000
Mean
7.340
Maximum
23.000
SE Mean
0.717
StDev
7.169
Minimum
0.000000000
Q1
1.000
Median
6.000
Individual Value
160
120
11
11
1
80
1111111 1
40
1 1
1
1
11
11 1
1
11
1
11
11
1
1
1
1
1
11
111
11
1
111 11111
1 11
1
_
UCL=16.4
X=7.3
LCL=-1.8
0
1
106
212
318
424
530
Observation
636
742
Moving Range
111
1
80
1060
11
1
1
11
1111 1
40
954
11
160
120
848
1
1
11 11 1
1
1
11
11
1
1
1
11
11
1 111
1 1
1
11 1
1 11
1
1
1
1
1
11
1111
1
11 11
11
__
UCL=11.2
MR=3.4
LCL=0
0
1
106
212
318
424
530
Observation
636
742
848
954
1060
Individual Value
1
111
20
1
1 1 11
UCL=16.45
15
10
_
X=7.34
5
0
LCL=-1.77
1
Moving Range
16
10
20
30
40
50
Observation
60
70
80
90
100
1
1
1
1
12
UCL=11.19
8
4
__
MR=3.42
LCL=0
1
10
20
30
40
50
Observation
60
70
80
90
100
Upon improving the process the mean cycle time went from 10.217 days with a standard deviation of
11.961 days before to a mean cycle time of 7.53 days and a standard deviation of 7.227 days after.
By improving the process with the DMAIC project we were able to decrease mean cycle time substantially
as well as reduce variation.
Control Phase
(1) Standardization
What should they be doing after standardization of the revised standard operating procedures?
Upon receiving a research request they should do the following:
1. Look up the prospect in the system to make sure the report requestor specified the correct
prospect management status on the online report request form.
2. If the prospect management status on the request form does NOT match the prospect
management status in the system the research analyst should call the requestor and get them to
update it in the system.
3. If there is anything on the request that is unclear the research analyst should call the requestor
for clarification.
4. The Research Analyst should log the report request in the Project Log database and make sure
to enter all of the normal information as well as the new fields which were added (prospect
management cycle stage)
5. After completing the report the researcher should complete the Research Report Quality
Checklist and have another Research Analyst perform a peer review to ensure quality.
6. After sending the report to the requestor the Research Analyst should close out the project in the
Project Log database and make sure to enter in the completed date as well as the new fields
which were added in this project (peer review completed (Y/N), checklist completed (Y/N)
With the time saved from this project Research Analysts will have more time to identify new prospects for
the Development Officers. It is important that these new prospects are tracked in order to:
(a) give the Research Analysts the credit they deserve in regards to helping bring in new
donations
(b) help keep track of the hard benefits of this DMAIC project
In order to be able to track these new prospects, Research Analysts will be asked to do the following:
1. Go into the new Prospect Identification database and set the Research Status on prospects they
are researching as potential donors to the letter R, also enter the date, the research analysts name
and the area the prospect is being identified for (School/College).
2. After identifying that prospect as a possible good donor their name is sent to their respective
Development Officer. If the Development Officer confirms the prospect is a good one change the
Research Status to I, if the prospect is not good for that School/College set the Research Status
to NR, if the prospect is not a good prospect for UM at all set the Research Status to N.
Who will make decisions on the future outputs of the revised process procedures?
The Research Director will make decisions on the future outputs of the revised process procedures.
(2) Mistake-proofing
Potential risk elements were identified that could be carried through the process by using failure mode
effects analysis (FMEA). These risk elements are:
1. Requestor enters wrong Prospect Management status on Report Request form.
2. Research Analyst neglects to conduct peer review/use quality checklist.
3. Researcher neglects to change Research Status for newly identified prospects.
Failure
Mode
Requestor
enters wrong
PM status on
request
Possible
misclass. of
report
Researcher
neglects to
conduct peer
review/use
checklist
Lower
quality
reports.
More time
by Director
proofing
reports
Research
does not
get the
credit it
deserves
for Prospect
Ids
Researcher
neglects to
change
Research
Status for
newly
identified
prospects
Likelihood
of
Occurrence
9
Severity
Likelihood
of Detection
RPN
Action
441
252
126
Look up
prospect in
system.
Call
requestor
for
clarification
Create
weekly
exceptions
reports for
Director for
monitoring
purposes
Create
weekly
exception
reports for
Director for
monitoring
purposes
Likelihood
of
Occurrence
7
Severity
Likelihood
of detection
RPN
196
72
70
(3) Documentation
Training Documentation for Researchers and Requestors of Research
The revised Research Request process is as follows:
The report requestor will specify the current prospect management status (amongst other information) on
the online report request in order to receive the correct report.
Prospect Management Status
Qualify
Cultivate
Solicit
It is very important that the Requestors of Research realize the importance of specifying the correct
prospect management status on the online report request form in order that the Research Analyst
performs the correct research.
The Research Director or Senior Research Analyst will assign the incoming research requests to
Research Analysts depending on who the requestor is.
With the revised process there are currently 2 Research Analysts assigned to President Shalala and the
Office of the Vice-President for University Development and 4 Research Analysts distributed evenly
among the Schools and Colleges.