Download - Make data work harder
MAKE DATA WORK HARDER SUCCESSFULLY EMBED PREDICTIVE ANALYSIS IN YOUR FUNDRAISING STRATEGY
Attitude
• Data analysis does not replace fundraising skill, it compliments it.
• Analysts must work in partnership with fundraisers to accomplish common goals.
Appetite
• Find your champion
• Demonstrate worth on small low risk project
2010 ROI =
= 138%
Introduction of predictive model
2011 ROI =
= 294%
2010 CRP =
= £0.72
Introduction of predictive model
2011 CRP =
= £0.34
Communicate
• Understand your audience
• Practical analytics not data science
• Easy to go too far
What is a predictive model?
Find those that look like your donors and
you will have a better chance of producing
more donors!
• Gather data about your constituents
• Find data with predictive power
• Combine data to produce a model
What gives data predictive power?
What does the average donor look like?
• Predictive models use distinguishing characteristics not
common characteristics
• Do not look only for similarities between your donors
• Look for distinguishing qualities between your donors
and the rest of your constituents
What does a donor look like?
The questions
Is there any point looking at legacy pledges
to find new donors?
Do these results give email address more
predictive power?
The answers…
It is impossible to tell.
Why?
We have ignored our non donors.
The complete picture…
The answers…
Email address = COMMON characteristic
Legacy pledge = DISTINGUISHING characteristic
MAJORITY of donors have email yet MINORITY of
those with email are donors.
MINORITY of donors have pledged legacy yet
MAJORITY of legacy pledgers are donors.
The question is NOT “Why do people give?”.
xkcd.com
Selecting Variables
Giving history Age
Wealth indicators Questionnaire/Survey responder
Interests Email clicks
Affiliations Twitter/facebook
Gender Events attended
Sign up/subscriptions Family relationships
Employment/positions Address
Marital status Email
Degree Phone
Mailing preference (opt outs) First gift amount
Volunteers Proximity
Prepare your data file
• Excel v SPSS
Constituent ID
Is a donor? Attended Event?
Has email? Over 40?
A 1 1 1 1
B 1 0 1 1
C 0 1 1 0
D 1 1 0 1
E 0 0 1 1
Evaluate
Score Decile
Non donors
Donors Total Donor Ratio
1 3611 59 3670 1.61%
2 4672 54 4726 1.14%
3 3351 145 3496 4.15%
4 4906 172 5078 3.39%
5 3698 275 3973 6.92%
6 3813 351 4164 8.43%
7 3511 489 4000 12.23%
8 3575 593 4168 14.23%
9 3593 802 4395 18.25%
10 3190 1010 4200 24.05%
Baseline 37920 3950 41870 9.43%
Evaluate
0%
5%
10%
15%
20%
25%
30%
1 2 3 4 5 6 7 8 9 10
Do
no
r R
atio
Constituent Decile
Conclusions….
• The average donor and the average non-donor
may look the same.
• Look for distinguishing characteristics not
common ones.
• Don’t look at donors in isolation. Compare data
for donors with data for everyone.
Conclusions….
• Data modelling can help you focus your resources on the best prospects.
• Demonstrate worth on low risk segments.
• Consider your audience. Communicate results so that everyone can understand.
Paul Weighand
Insight Manager
University of Edinburgh
@paulweighand