How to avoid some of the pitfalls when deploying legacy targeting models david dipple - adroit data and insight
- 1. David Dipple
- 3. Fellow of Royal Statistical Society Worked with Not For Profit and Charity Clients for over 25 years Recognised as an expert data modeller Trained numerous analysts and fundraisers in the use of analysis in fundraising Worked with charities in UK and mainland Europe
- 4. An approximate answer to the right question is worth a great deal more than the precise answer to the wrong question. -The first golden rule to applied mathematics The formulation of a problem is often more essential than its solution which may be merely a matter of mathematical or mental skill. •A. Einstein
- 6. Question Gubbins Answer
- 7. The only point where there is interaction is at the start – no time is allocated for re-visiting the question
- 8. Needs, wants and requirements Question Initial Brief Marketing Analysis Initial Brief Brief Analysis Analysis reqs Initial results Final analysis Marketing Results Answer Analysis workshop
- 9. Forget complex relationships – simplicity is your friend Analysis follows the 80/20 rule ◦ 80% of the analysis can be done in 20% of the time. ◦ The last 20% takes 80% of the time
- 11. Binary Clustering: Charity Sector Humanity 3rd Word & Overseas Environment Disability Cancer & Nature Health Medical Research Wildlife Animal Welfare
- 12. Our Target? Or
- 13. Traditionally many legacy campaign have been designed and devised around a message they are not shaped around supporters needs and requirements To fully tap the legacy potential of the base a more supporter lead strategy would match supporter interests and propensity to legacy message
- 14. Method ◦ Mail ◦ Phone ◦ Event ◦ Online The halo effect
- 16. Behavioural Recency, Frequency, Value, Forms of help. Segmentation Demographic Attitudinal Lifestage, Age, Gender Questionnaires, Geodems Interests & Beliefs
- 17. Payment Type Interests Amount Lifestyle Date Cause Name Address Gender LTVs Donor & Age RFVs Demographic Income Scores Details Media codes Responses Method
- 18. Donor Information Communications Attitudinal Donor & Demographic Details Database Donations Derived
- 19. Geo-Dems are great for cold and certain aspects of warm targeting For small population analysis they tend to be less useful ◦ For one model that I created by using a geo-dem it added 0.5% to the power of the model Take care with including or excluding people based on their geo-dem coding
- 20. Academic Centres, Students and Young Acorn Description Professionals Personicx Retired - Low income - Aged in the City Description Suburbs
- 21. People tend to be interested in people ◦ But why are they interested? ◦ What aspects of your cause excites them? ◦ What motivates them to give you money?
- 22. What data do we currently have? ◦ What is its quality What data would we like to have? ◦ What barriers are there to getting it?
- 24. But what type of model? ◦ Legacy ◦ Pledger ◦ Legacy & Pledger ◦ Residuary/Pecuniary The past determines the future ◦ Lifetime Model ◦ Time Limited Model ◦ Something Else
- 25. SPSS Excel FastStats MapInfo & MapPoint My own software
- 26. Modelling techniques ◦ Binary Logistic ◦ Discriminant ◦ Multinomial Logistic ◦ CHAID ◦ Proxy
- 27. Type of Data ◦ Number of Relationships ◦ Supporter Lifetime ◦ Number of Gifts ◦ Age of Supporter ◦ Gift Aider Time is not our friend!
- 28. Beware of False Relationships Gender Response Age Response Male 8% Young 12% Female 10% Old 12% Population Response: 10% Gender: Male Gender: Female Response: 8% Response: 12% Age: Young Age: Old Age: Young Age: Old Response: 15% Response: 5% Response: 10% Response: 16%
- 29. c Clas sification Table Predicted a b Selected Cas es Unselected Cas es Legator Percentage Legator Percentage Obs erved 0 1 Correc t 0 1 Correc t Step 1 Legator 0 776 134 85.3 908940 153597 85.5 1 173 725 80.7 83 272 76.6 Overall Perc entage 83.0 85.5 a. Selected c as es sel_var EQ 1 b. Unselected c ases sel_v ar NE 1 c. The cut value is .500 Multiple ways of understanding if a model has worked. Most of the output can be ignored by non statisticians and the key – The key is finding what needs to be communicated to marketers and in what form. used to determine power.
- 31. Selected High Score Supporters Even with a small population outcome models – test down the model to reduce the Tom Smith effect. Model Score
- 32. Building legacy models has so far been carried out by building statistical propensity models. These need previous results to determine what will happen. But if there are no previous results you can’t build a model or can you?
- 33. The factors that increase propensity to make a pledge or leave a legacy are fairly well know – as we saw earlier Create binary flags for each of the data items given earlier and then add them all up. The higher the result, the more likely to make a pledge (and it works).
- 34. Analysis of a legacy campaign tends to be point based, That is how many responded to being contacted To truly understand the effect of legacy campaigning the relationship over time needs to be examined, including the effect on non legacy messages – that is the full supporter journey
- 36. Message 1 Message 2 Message 3 Message 4 Single model that determines both who should be No Contact Model contacted and with (at this point…) what message. Warehouse
- 37. The biggest barrier to producing efficient models is lack of data – especially demographic and attitudinal data Understand what the data is saying and then use an appropriate model - There is no one perfect solution There is no certainty in modelling – models are built from past behaviour and if you change what you are doing it can take a while for the data to catch up Examine the whole supporter journey to understand the full relationship Define the question and the answer will be much easier – remember a model is not a panacea