2. Ludo Longin
Administrateur Délégué Direct Social Communications
My favourite song: Queen
Geert Verstraeten
Partner at Python Predictions
My favourite song: Absynthe Minded
3. Analytics in
Fundraising
How knowing donors
helps in growing donors
Ludo Longin Geert Verstraeten
Direct Social Communications Python Predictions
4. Direct Social Communications nv
= communication agency
°1985
1 activity:
fundraising for humanitarian organisations
Team:
12 enthousiastic people
5. Health Food
Belgium Belgian Assocation for Burn Injuries Food Banks
Belgian Cystic Fibrosis Assocation Restaurants of the Heart
Kom op Tegen Kanker (Cancer Assoc.)
World Mercy Ships
Chain of Hope
Damien Foundation
Medics Without Vacation
6. Children Animals
Belgium Collective Research and Expression
Youth Village
Pelicano Foundation
World If Child Help Veterinarians Without Borders
7. Handicap People
Belgium Blind Care Light & Love Flemish Autism Association
World Handicap International Pilots Without Borders
Sensorial Handicap Cooperation Mamas for Africa
The Voice of the Lebanese Women
Friends of Sister Emmanuelle
8. ± 400 fundraising campaigns using dm
> 8,000,000 letters to private individuals
> 1,500,000 inserts in newspapers
> 600,000 donations per year
> 24,000,000 euros
9. Typical dm-donor in Belgium
Female
60+ years old
religious
Fully owned appartement/house
Adult children have left the house
10. Normal results of dm-campaigns
Response rate in housemailing:
◦ 8%
◦ 10 %
◦ 12 %
◦ 14 %
◦ More than 15 %
11. Normal results of dm-campaigns
Response rate in Acquisition campaigns:
◦ 2%
◦ 2,5 %
◦ 3%
◦ 4 or More %
12. Recruitment campaigns
Where do we get our donors from?
CONSUDATA
D.S.C.
+ criteria
Other
± 220,000 databases
6 million
addresses
addresses
14. Python Predictions
Core business: Predictive Analytics
Since 2006
Based in Brussels
References:
15. Cloning in a Nutshell
STEP 1 Select the best donors:
16. Cloning in a Nutshell
STEP 2 Calculate area conversion rate:
17. Cloning in a Nutshell
STEP 3 Calculate area profile
Residents
Housing
Neighbourhood
18. Cloning in a Nutshell
STEP 4 Build Predictive Model
Residents
Housing
Neighbourhood
19. Cloning in a Nutshell
STEP 5 Validate Predictions
4,0%
Cloning
Response
3,0% Geo Segmentation
Age
2,0%
1,0%
0,0%
0 50000 100000 150000 200000 250000 300000
Number of Households Targeted
20. Cloning in a Nutshell
STEP 5 Validate Predictions
Households we
would target
Age category
Households we
9%
8% would not target
7%
6%
5%
4%
3%
2%
1%
0%
0‐4 5‐9 10‐14 15‐19 20‐24 25‐29 30‐34 35‐39 40‐44 45‐49 50‐54 55‐59 60‐64 65‐69 70‐74 75‐79 80‐84 86‐89 90‐94 >95
21. Cloning in a Nutshell
Households we
STEP Validate Predictions
would target
5 Households we
would not target
House surface Number of bathrooms
21%
9%
26%
8%
N/A < 35 m2 35 till 55 till 85 till 105 till > 125
54 m2 84 m2 104 m2 124 m2 m2 0 1 2 or more
22. Campaign Results (First Test)
Response Percentage Average Donation Amount
2,71%
2,29% 28,6 €
1,98% 25,5 €
21,6 €
Age > 55y Cloning Cloning Age > 55y Cloning Cloning
Top 100.000 Top 10.000 Top 100.000 Top 10.000
improvement of 37% in improvement of 32% in
response rate donation amount
23. Campaign Results (First Test)
Break
(revenue per letter sent)
0,78 €
0,43 €
0,58 €
Current and
Future Usage
Age > 55y Cloning Cloning
Top 100.000 Top 10.000
improvement of 82% in
revenue per letter sent
24. QUESTIONS?
Or later?
Ludo Longin Geert Verstraeten
ludo.longin@dsc.be geert.verstraeten@pythonpredictions.com
Tel +32 2 280 00 74 Tel +32 2 762 69 00
Direct Social Communications Python Predictions
www.dsc.be www.pythonpredictions.com