Improving the Research Request Process

In the Division of University advancement

At the University of Miami

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.

What business objectives are supported by the highest priority project?

The Prospect Management Project is strongly related to the Increase the endowment, Improve the
national ranking of the University, Improve student experiences and Improve interdisciplinary
research presidential objectives.
Problem Statement
According to the Research staff, they are overworked and understaffed due to the fact that the
report requestors are asking for types of profiles that are more than the requestor needs and are an
inefficient use of the researchers time and energy. The Development Officers (DOs) feel that their
needs are not being met and that the process is reactive. This means that many of them do their
own research and completely bypass the Research department. The DOs are spending a lot of their
time doing research when they should be out cultivating and soliciting potential prospects. In
addition, the Research staff is spending all of their time doing research reports and does not have
time to identify new prospects for the Development Officers.
Goal Statement
The goal for the Research Request project is to improve the process of efficiently delivering the right
type of research request to the customer for what they need it for. Also the time of the researchers
must not only be spent on doing research reports but more importantly on identifying new donor
prospects.
Project Objective
To transform the current Research Request process into a proactive and efficient process.
What are the process boundaries?
The starting point for the project is when the Research Requestor makes an online Report Request.
The stopping point for the project is when the Research profile is delivered to the Research
Requestor.
What are the benefits of the project?
The soft benefits include eliminating complaints from the Research Department for being
overworked. It will also include making more use of the researchers time so they can spend time in
other areas such as prospect identification. There will also be soft benefits to the requestors of the
research as they will receive better quality reports due to researchers not having to rush to meet
deadlines. Another benefit will DOs spending less time doing research and more time cultivating and
soliciting prospects.

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

Report requestor fills out

online report request
Form

Research Director
receives request

Request is assigned to
researcher

Researcher completes
report

No

Research Director
Reviews report

Does report meet

standards?

Yes

Researcher sends profile

to requestor

End

(3) Analysis Data for Voice of the Customer Data

Raw VoC Data

Affinity Diagram
Theme (Focus Point)

Not an efficient process

Not a proactive process at
all, it is completely
reactive
Not a good system
I am on my own
My assistant and I do all
of our own research
Research cannot manage
the flow, they start doing a
profile for her then get
pulled away to do a profile
for Sergio or President
Shalala
I do my own research

DOs end up doing their

own prospect research
when there is a Research
Department to do it.
(Product related)

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.

(3) Establishing the Baseline Capabilities for CTQ

A dot plot and descriptive statistics were completed to show the cycle times to complete all the requests
from 2004 and 2005.
On average it took 10.271 days to complete a report in 2004 and 2005.
Variable
C2

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

Dotplot of Cycle Time to Complete Report (in days)

25

50

75

100
CTQ

Each symbol represents up to 9 observations.

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.

I-MR for Days to Complete

1

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

Analyze Phase Research Requests

Process map identifying and linking the Xs and Ys:
Start

Report requestor requests

specific report from online
report request Form

X1 = Number of reports requested from each

requestor per year

Research Director
receives request

Request is assigned to
researcher

Researcher completes
report

No

X2 = Type of report ordered (L1, L2, B, T)

X3 = Priority (Due Date - Date Requested)
X4 = Inappropriate profile requested

Research Director
Reviews report

Does report meet

standards?

Yes

Researcher sends profile

to requestor

Y1 = Cycle time to complete research report

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

Operational Definition for X4: Inappropriate report requested

Criteria for priority of the report request: Open the database and pick a report request.
Test: Find the value for the type of report requested and the step in the prospect management process.
Decision: If the report is a Level 1 and the step in the prospect management process is either Cultivate
or Solicit then the request is appropriate, if the report is a Level 1 and the step in the prospect
management process is Qualify then the request is inappropriate. If the report is a Level 2 and the step
in the prospect management process is Solicit then the request is appropriate, if the report is a Level 2
and the step in the prospect management process is Qualify or Cultivate then the request is
inappropriate.
Gage R&R Studies of Each X
Gage R&R studies were not conducted on the Xs because their values come from a database and are
inherently obvious. Therefore, the assumption is that all of the measurements for the Xs are acceptable.

Establish Baseline Process Capabilities for Xs

Baseline Data for X1: Number of reports requested from each requestor per year
The control charts for the number of reports ordered per requestor for 2004 and 2005 shows special
causes of variation. After talking to people involved with the process it was determined for both years
that those who were above the upper spec limits were either President Shalala, the Vice-President, AVPs
and a couple people who simply did not understand how the process works and ordered reports for
everything. Those who were under the lower spec limits were people who do their own research
The years 2004 and 2005 are separated is because the turnover for the Development Officers is
ridiculously high! It is more meaningful to separate because in some cases some of the schools and
colleges have had 2 or 3 different Development Officers in that 2 year period. For example, the College
of Arts and Sciences, Paul Mischler had the 2nd most requests in 2004, he left and Holly Davis took over
and she had the 4th most requests in 2005, digging deeper we found they both misclassified A LOT. If I
put the data together for 2004 and 2005 they may not stick out as much and I may have missed them.
After looking at the Pareto charts it was determined that 6 requestors including President Shalala, the
Vice-President, a couple Assistant Vice-Presidents and a couple of users who are uneducated about the
process account for 60% of the reports ordered.
C Chart - Reports by Requestor - 2004

C Chart - Reports by Requestor - 2005

1

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

Pareto Chart - Requests by Requestor - 2004

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

Pareto Chart - Requests by Requestor - 2005

100

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

Count 10477 39 38 30 29 23 20 18 18 15151310 7 7 6 6 6 5 4 3 3 3 3 2 2 2 2 2 2 1 5

Percent
2015 8 7 6 6 4 4 3 3 3 3 3 2 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 1
Cum % 203542505561 65697376 7982848688 8990919393 9495959697 9797989898 9999100

Baseline Data for X2: Type of report ordered (L1, L2, B, T)

The Level 2 report was the most requested report in 2004 and 2005, with the Blurb being the second
most requested and the Level 1 and Trustee report following. This is significant due to the fact that a
Level 2 report takes much longer to complete than any of the other reports which in turn increases the
overall cycle time.

Percent

80

Type of Report Requested - 2005

Type of Report Requested - 2004

Number
107
156
126
23
3
3
24

Pareto Chart - Types of Reports Ordered - 2005

Pareto Chart - Types of Reports Ordered - 2004

500

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

Baseline Data for X3: Priority (Due Date Date Requested)

The average report was needed in 11.228 days in 2004 and 2005.

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

I-MR for Priority

1

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

Baseline Data for X4: Inappropriate report requested

After speaking with some of the Research staff it was realized that some of the reports may have been
misclassified, meaning that the report requested was not the report that was actually needed (see
operational definition). After taking 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.
Misclassification of Report requests - 2004

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

Misclassification of Report requests - 2005

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

Pareto Chart - Requests by Requestor - 2005

500

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

Count 10477 39 38 30 29 23 20 18 18 15151310 7 7 6 6 6 5 4 3 3 3 3 2 2 2 2 2 2 1 5

Percent
2015 8 7 6 6 4 4 3 3 3 3 3 2 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 1
Cum % 203542505561 65697376 7982848688 8990919393 9495959697 9797989898 9999100

Percent

500

X2: Type of report ordered (L1, L2, B, T)

The dot plot for Cycle Time to Turnaround Reports (CTQ) by Types of Report Ordered indicates:
1. The cycle time for Blurb reports has the lowest variation which makes sense as they are the
shortest and quickest report.
2. The cycle time to complete Level 2 reports has the largest variation as they involve much more
work than the others.
This means that more Level 1 and Level 2 reports that are ordered will drastically affect the cycle time to
turnaround reports.
Dotplot for Cycle Time to Turnaround Reports

Dotplot for CTQ by Types of Report Ordered

Level 1

Level 2

Blurb
Trustee

25

50

75

100

125

150

Other

175

25

50

75

CTQ

100

125

150

150

175

CTQ

Each symbol represents up to 9 observations.

Each symbol represents up to 12 observations.

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

Dotplot for Cycle Time to Turnaround Reports

25

50

75

100

125

150

175

25

50

75

Each symbol represents up to 9 observations.

100
CTQ

CTQ
Each symbol represents up to 8 observations.

125

X4: Inappropriate report requested

The dot plot and descriptive statistics for Cycle Time to Turnaround Reports (CTQ) by Misclassified
Reports indicates:
1. The dotplot shows that misclassification of reports increases the time it takes to complete
reports. The average of 11.48 days for misclassified reports in 2004 and 12.8 days in 2005 are
both higher than for all of the requests (10.22)
2004
Variable
C1

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

Dotplot for Misclassifed Reports 2004 vs CTQ

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

Dotplot for Misclassifed Reports 2005 vs CTQ

12

18

24
CTQ

30

36

42

Hypotheses Regarding Relationships Among the Xs and CTQ

The analyze phase resulted in the following hypothesis:
Cycle time to turnaround report = f (X1: Number of reports ordered per requestor/year, X2: Type of
report ordered, X3: Priority, X4: Inappropriate report requested)
The reason for the extreme points in the CTQs and their relationships to the extreme points in the Xs is
due to the correlation between cycle time and how quickly the report is needed (priority). If a report is
not needed for 90 days, the researcher wont work on it until close to the date it is needed instead
focusing on reports that are needed sooner, hence, the outliers in the CTQs.

X1: Number of reports requested from each requestor per year

There is no optimum number of reports requested per year. It all depends on the number of prospects
they are going to qualify, cultivate or solicit, which depends on a bunch of factors that are not
measurable.
X2: Type of report ordered (L1, L2, B, T)
There is no optimum type of report ordered. Blurbs are the easiest and least time consuming then come
Level 1 reports and finally Level 2s are the most time consuming. As you will see in my improve phase
it all depends where the prospect is in the cultivation cycle as to the type of report ordered.
X3: Priority (Due Date Date Requested)
There is no optimum level for priority. If the DO needs the report in 2 weeks because they are meeting
with a prospect then the priority is 14 days.
X4: Inappropriate report requested
The optimum level for inappropriate report requested is zero. The more misclassified reports you have
the more time your researchers are wasting doing research that most likely will not be used. This is a
pretty easy fix in the improve phase.
A Design of Experiments is not needed in the Analyze phase due to the fact that we know the optimum
level of factor levels due to process knowledge (see above). The problems with the Xs can be solved by
modifying the procedures in the Advancement Research Office which will be detailed in the Improve
phase.

Improve Phase Research Requests

The Research Report Request process begins with the report requestor going online and filling in the
form below.

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.

Research Report Quality Checklist

Prior to beginning any research report, remember to always do the following:
Check Assignment of Prospect
Check Prospect Management Status/Cultivations
Contact Requesting Officer to Discuss Specific Needs
Type of Report
Prospecting Blurb

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

Confirmation of Home Address/Contact

Personal Information (Spouse, children, birthdate)
Confirmation of Business Address/Contact
Current Business Dealings/News
Significant Directorships
Significant Charitable/Civic Affiliations
Basic On-Line News Search (up to 3 years)
Basic Wealth Indicators (i.e. Home Value, Stockholdings, Salary,
Business Revenue - if owner of company)
UM Affiliations (if not previously delivered)
UM Giving History (if any)

In-depth - Level 2

Ready to Solicit

Indepth Online News Search

Extensive Wealth Profile (including family relations if applicable)
Indepth Business Information/Career History
Indepth Analysis of Past Charitable Giving
Philanthropic Affiliations
Updated UM Giving History

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

Targeted to a specific question that a DOD may need for a

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.

Reports requested by requestor

The final problem with the Research Report Request process has to do with the fact that only 1 out of 6
researchers is assigned to the President's Office, Vice-President's Office and Trustee Services even
though over 60% of the research report requests come from those 3 areas.
In order to solve this problem a computer based simulation of the process was created to optimize the
workload of the Research Analysts.
The ARENA Input Analyzer was used and it was determined that an Exponential distribution fit the interarrival time data the best. See below.

Inter-arrival Rate of Report Requests Distribution Summary

Distribution:
Expression:
Square Error:

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)

I used Input Analyzer to fit the

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)

Simulation Model Average

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.

[Simulation of current process]

A s s g
i n
Co m

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

Cor por at e and Foundat ion

Relat ions

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

Send t o Cor r ect

Resear cher

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.

Average time in queue for Shalala/VP reports can be summarized as follows:

nd
(note: all of the average times in queue can be found in the 2 table above)
Current System
(55.28, 75.58) 95% C.I.
Alternative #1
(6.14,7.8) 95% C.I.
Alternative #2
(6.14,7.8) 95% C.I.
Alternative #3
All in 1 queue (.22,.42) 95% C.I.
rd
Clearly the 3 Alternative appears to be the best as the 95% C.I. for its mean is the lowest and does
not overlap any of the others. However this alternative would require all Research Analysts to work
on all of the reports and use 1 queue. This would not be feasible due to the fact that Development
Officers are used to working with their own Research Analysts and have developed strong working
relationships with them.

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.

Revised Flowchart of the Research Report Request Process

Start

Report requestor requests

report online stating stage
of Prospect Management
cycle prospect is in

X1 = Number of reports requested from each

requestor per year

Research Director
receives request

Request is assigned to
researcher depending
area requestor is from

Researcher completes
report

X2 = Type of report ordered (L1, L2, B, T)

X3 = Priority (Due Date - Date Requested)
X4 = Inappropriate profile requested

Researcher completes
quality checklist and peer
review and submits report
to Research Director for
review/approval
No

Research Director
Reviews report

Does report meet

standards?

Yes

Researcher sends profile

to requestor

End

Y1 = Cycle time to complete research report

Pilot Test Results

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%

Dotplot of Cycle Time to Complete Reports (in days)

Dotplot of Cycle Time to Complete Report (in days) - Pilot Test

9
12
timetocomplete

15

18

21

Descriptive Statistics: timetocomplete

Variable
timetocomplete

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

Prospects Identified by Research Department

With the re-organization of researchers and the reduction in the misclassified reports the researchers
have been able to use their time to identify new donor prospects for the university
Since August 2006 over 400 new good prospects were identified, prior to this DMAIC project the
researchers did not identify any new prospects due to the fact they were spending all of their time doing
non-value added research reports.
In addition, that number may grow much higher as there are a total of 84 prospects who have been
identified but the researcher is waiting to hear back from the Development Officer as to whether they are
good for that School/College or not.

Before/After I-MR Chart of Cycle Time (CTQ #1)

I-MR Chart of Cycle Time by Before/After

1

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

I-MR Chart of Cycle Time (after)

1 1 1 11

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

The Control phase is comprised of 4 steps:

1. Standardization
2. Mistake-proofing
3. Documentation
4. Giving control of the process back to the process owner

(1) Standardization

Who is involved at the revised steps of the process?

The Director of the Research department as well as all of the Research Analysts are involved in the
revised steps of the process.

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.

Why should they follow the revised standard operating procedures?

There are a few reasons they should follow the revised standard operating procedures:
1. To prevent misclassifications of report requests.
2. To ensure reports are sent out with the highest quality.
3. To allow proper tracking of newly identified prospects.
4. To ensure Development Officers are entering prospect management information into the system.
5. To ensure the Report Requestor gets exactly what they need.
6. To be able to ensure the Research Department gets the credit they deserve for helping bring in
donations.

Where/when should they be doing the revised process procedures?

The revised process procedures should be done throughout the Research Report Request Process.
Documentation will be given to the Research staff to tell them exactly where/when they should be doing
the revised process procedures.

How much will it cost to do the revised process procedures?

There will be no cost to do the revised process, everything is being done in-house.

Is additional training needed to perform the revised process procedures?

Yes, additional training will be conducted for the Research staff as well as those who submit Research
Report Requests.

How often should the revised process procedures be monitored?

The revised process procedures will be monitored monthly.

Who will monitor the revised process procedures?

The Research Director as well all Research Analysts will be trained on monitoring the revised process
procedures. The Research Analysts will be trained in order to help foster intrinsic motivation on their part
so that they feel like they are part of the improvements that are taking place.

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 Effects Analysis (FMEA)

Risk
Elements

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

Requestor enters wrong Prospect Management status on Report Request form.

Upon receiving online report requests the Research Analyst will look up the prospect in the database to
ensure that the correct Prospect Management status was entered. If they do not match the Research
Analyst will call the Research Requestor for clarification.

Research Analyst neglects to conduct peer review/use quality checklist.

Weekly exceptions reports will be created for the Director to monitor peer reviews and use of the quality
checklist for reports rendered that week.
Researcher neglects to change Research Status for newly identified prospects.
Weekly exception reports will be created for newly identified prospects whose status is R (being
reviewed) to remind them to call their Development Officers to see if the prospects they have identified
are good or not.

(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

Type of Report Given

Blurb
Level 1
Level 2

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.

Research Requests Procedures

Upon receiving the Research Request the Research Analyst should do the following:
7. 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.
8. 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.
9. If there is anything on the request that is unclear the research analyst should call the requestor
for clarification.
10. 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)
11. 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.
12. 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).
Prospect Identification Procedures
When working on prospect identification Research Analysts should 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.

(4) Giving control of the process back to the process owner

This concludes the project. This process will now be turned over to the process owner with
control plans for future PDSA cycles.