This presentation, delivered by InfoCision Chief Marketing Officer Ken Dawson during InfoCision's 2010 Fundraising Summit, explores how demographic and psycho-graphic data and business intelligence helps organizations better understand donors to receive the highest return on investment.
1. DONOR PROFILES & DATA
ANALYTICS 101
The Who, What, When, Where & Why of YOUR Donors
2. 2
Business Intelligence Group
• Dedicated marketing and consulting group.
• Currently purpose driven to deliver the right message at
the right time the right way.
• Using higher levels of personalization in direct marketing.
• Dedicated IT group.
• Creates internal as well as external products and
services.
• Manages and oversees InfoCision Consumer Database.
3. 3
InfoCision Consumer Database
•A nationally compiled list of marketing and
demographic data.
•Data is updated every two months.
•Data Base details:
•205 million records
•110 million households
•80 million phone numbers
•56 million distinct phone numbers
5. InfoCision Consumer Database Fields
Sample
• Age
• Income
• Length of residence
• Level of education
• Family position
• Net worth
• Voter party
• Mail order buyers/responders
• Financial services information
• Internet shoppers
• Religion
• Ethnicity
• Language preference
• Country of origin
6. 6
Business Intelligence Services
• We are able to turn our Business Intelligence into
actionable results by tying the information from our
Consumer Database into:
• Segmentation Strategies
• Data Overlays
• Predictive Models
• Profiles
• Inscription
• P.U.R.L.
• Lifetime Value Studies
• Market Mapping
• Campaign Enhancement
• Marketing Tests
• List Development
• Analytic Studies/Research
• Data Warehousing
• Targeted Messaging
• One-to-One Direct Mail
• Inbound Routing Strategy
7. 7
Profiles and Modeling
Modeling and analytics enhances the
efficiency and effectiveness of
marketing endeavors , particularly:
Current donor Retention
New donor Acquisition
Up-SellUp-Selling and Cross Selling
8. 8
Profiles and Modeling
Define the
current donors
with profiling
Segment the
current donors
Model the
current donors
to target for
cross selling and
acquisition
Step 1 Step 2 Step 3
9. 9
Define the
current donors
with profiling
Profiles and Modeling
• Matching your current donors
against the 165 attributes in the
database allows you to know
who your donors are.
10. 10
Application of Profiles and Models to
Donors
• How do we leverage what we now know about our
donors and customers?
• Predict lapsing donors
• Target donors for sustainer efforts
• Target donors for major gifts/bequests
• Target donors for special event/volunteerism
• Build predictive models for acquisition of new donors
11. 11
Profiles and Modeling
• Using data to segment files to
develop marketing strategies
and variable offers allows for
improved response rates.
Segment the
current donors
12. 12
• Problem
• Donations and response rates were flat.
• Solution
• Profile current donors.
• Use income and home value indicators to drive segmentation for
variable offers.
• Lower gift asks to lower incomes and home values to drive up response
rates.
• Higher gift asks to higher incomes and home values to drive up average
sale.
• Bottom Line
• Call results showed positive response rate growth in the low group
(+18%) and positive average gift (+$12) in the high group.
Profiling - Case Study #1
13. 13
Profiles and Modeling
• Next step is a predictive model
to rank a prospect list on their
propensity to donate.
Model the
current donors
to target for
cross selling and
acquisition
Decile
Total Record
Universe
% of Total
Records
Cumulative %
Records Total Customers
% of Total
Customers
Cumulative %
Customers
Decile Response
Rate Cumulative Rate Marginal Lift Cumulative Lift
Most Likely 1 5,000 10.0% 10.0% 1,512 13.4% 13.4% 30.24% 30.24% 134 134
2 5,000 10.0% 20.0% 1,447 12.9% 26.3% 28.94% 29.59% 129 132
3 5,000 10.0% 30.0% 1,369 12.2% 38.5% 27.38% 28.85% 122 128
4 5,000 10.0% 40.0% 1,304 11.6% 50.1% 26.08% 28.16% 116 125
5 5,000 10.0% 50.0% 1,259 11.2% 61.3% 25.18% 27.56% 112 123
6 5,000 10.0% 60.0% 1,190 10.6% 71.9% 23.80% 26.94% 106 120
7 5,000 10.0% 70.0% 1,049 9.3% 81.2% 20.98% 26.09% 93 116
8 5,000 10.0% 80.0% 877 7.8% 89.0% 17.54% 25.02% 78 111
9 5,000 10.0% 90.0% 681 6.1% 95.1% 13.62% 23.75% 61 106
Least Likely 10 5,000 10.0% 100.0% 555 4.9% 100.0% 11.10% 22.49% 49 100
50,000 100.0% 11,243 100.0% 22.49%
XYZ Corporation
Gains Table Based on Development Sample
Overall Results
14. 14
•Problem
• Client wanted to acquire new donors.
•Solution
• Build a predictive model scoring current donors.
• Based on scoring, rank order prospects on their likelihood
to respond.
Modeling - Case Study #2
Decile Records Customers RR%
Average
Score
BP
Variance
1 5,000 1,512 30.24% 25.19 775.40
2 5,000 1,447 28.94% 24.56 645.40
3 5,000 1,369 27.38% 24.22 489.40
4 5,000 1,304 26.08% 24.00 359.40
5 5,000 1,259 25.18% 23.75 269.40
6 5,000 1,190 23.80% 23.54 131.40
7 5,000 1,049 20.98% 23.31 -150.60
8 5,000 877 17.54% 22.92 -494.60
9 5,000 681 13.62% 22.39 -886.60
10 5,000 555 11.10% 21.67 -1,138.60
Total 50,000 11,243 22.49% 188.44 0.00
Top Test Cume 15,000 4,328 28.85% 24.66 636.73
Bottom Test Cume 15,000 2,113 14.09% 22.33 -839.93
XYZ Corporation Model Scoring Performance
15. 15
•Bottom Line
• Call results showed the model worked well as a predictor
of donating behavior with four deciles performing above
average.
Modeling – Case Study #2
MODEL DECILES
Total
Universe
Completed
calls Sales
1 56,624 5,630 1,707
2 56,623 4,897 1,290
3 56,623 4,408 1,092
4 56,622 4,442 1,070
5 56,622 4,332 990
6 56,622 4,254 959
7 56,622 3,769 842
8 56,622 3,460 705
9 56,622 1,732 303
10 56,622 1,891 268
Grand Total 566,224 38,815 9,226 23.77%
InfoCision Management Corporation
Start Date: 4/1/09
End Date: 4/30/09
14.17%
17.49%
20.38%
22.34%
22.54%
22.85%
24.09%
24.77%
26.34%
30.32%
RR%
Campaigns: XYZ0001
16. 16
What tools do we have to unlock the value?
• Inscription (Script on Screen)
• Customizable screen prompts
• Variable and branch scripting
• Script changes can be made in real-time
• Variable objection handling
• Quantitative results tracking beyond final
disposition
• Dynamic gift asks
• Custom built CRM solutions
• Variable print campaigns.
• Intelligent call routing on inbound.
17. •Now that the audience is scored and segmented
• How do we now impact the offer?
•Analyze various affluence indicators and their
relationship to gift amounts
•Apply this information to develop a dynamic gift
ask utilizing variable scripting technology
Modeling – Case Study #3
18. •Findings:
• Household income displayed the highest
correlation to gift amounts
• Household incomes were then broken into five
income bands ranging from low to high
• Each income band was given a specific gift ask
• The key metrics we were looking to influence
were:
• Response rate
• Average gift
• Dollars per call
• Efficiency
Modeling – Case Study #3
19. • The second step was to index these incomes by the Cost
of Living Index to normalize data
Estimated Household
Income
$1 - $39,999
$40,000 - $74,999
$75,000 - $124,999
$125,000 - $249,999
$250,000 +
Modeling – Case Study #3
20. Control GRC
$/CC CC RR% Avg. Gift
$3.08 3,073 8.14% $37.47
Dynamic GRC
$/CC CC RR% Avg. Gift
$3.92 2,549 9.42% $41.66
Modeling – Case Study #3
21. • Dynamic GRC results against control:
• Revenue per call increased by 27%
• Response rate increased by 16%
• Average gift increased by 11%
• Also showing an increase were credit card rates at
12%
• Not only were gross conversions impacted but fulfillment
rate and ROI dramatically improved
Modeling – Case Study #3
22. 22
Market Channel Coordination
•Understanding the audience allows for true
integrated marketing approaches
• TM, DM, Interactive, Social, etc.
• Timing, message, offers can all be coordinated to truly
“reach” donors in a way that they choose to respond.
•Many industry studies have shown that, if
coordinated appropriately, these channels do not
“cannibalize” each other.
23. 23
De-Bunking the Myth!
•Myth
• Telemarketing has a negative impact on Direct Mail.
•Approach
• Study Direct Mail results as they relate to TM utilizing large
nonprofit partner and 3rd
party analysis
•Reality
• Multiple studies across various client data have shown a
positive impact on mail results with coordinated TM.
28. 28
Market Mapping
• CBSA and DMA data is available to do regional
analysis and targeting.
COLUMBUS
29. DONOR PROFILES & DATA
ANALYTICS 101
The Who, What, When, Where & Why of YOUR Donors
Notas del editor
As a second step, we studied various economic indicators to determine if they had an impact on giving.
We found that Household Income showed the strongest correlation to giving.
We used this information to derive variable gift asks.
Other indicators we looked at were net worth, home values, education, and zip level income percent.
It is my belief that income is the best indicator due to the nature of our business. We are asking for a monetary response in a matter of 5 seconds. People will quickly think about how much money they have readily available to give. It is also my belief that net worth is a better indicator for direct mail as people do not have to make a split second decision and as such think about what they can afford in a broader scope.
In analyzing results, we found these 5 buckets worked well for our variable gift asks.